Prism Hospital Software Inc. v. Hospital Medical Records Institute

Date of Release: August 25, 1994

No. C872267

Vancouver Registry

IN THE SUPREME COURT OF BRITISH COLUMBIA

BETWEEN:

)

PRISM HOSPITAL SOFTWARE INC.,

REASONS FOR JUDGMENT

EDGAR TRUEIT, PENROSE HOLDINGS LTD.)

JOHN JEFFERY and MAINLAND

BUSINESS SYSTEMS INC.

)

OF THE HONOURABLE

PLAINTIFFS

AND:

THE HOSPITAL MEDICAL RECORDS

INSTITUTE, JOHN STENABAUGH,

LYNNE ABBOTT and STEVE IMBERT

MR. JUSTICE PARRETT

DEFENDANTS

Counsel for the plaintiff:

Counsel for the defendant:

Place and Date of Hearing:

Adrian Chaster, Esq.

P. Hamilton, Esq.

G. E. Fisk, Esq.

N. L. Fishman, Esq.

Vancouver, B.C.

Nov. 4, 5, 6, 7, 8,14,

15, 18, 19, 20, 21, 22

25, 26, 27, Dec. 2, 3,

4, 6, 9, 10, 11, 12,

13, 1991; Jan. 6, 7,

9, 10, 13, 14, 15, 16

17, Feb. 3, 4, 5, 6,

7, 10, 11, 12, 13, 14,

17, 18 and July 9, 1992

- 2 -

I.

INTRODUCTION

This action raises factual and legal issues which are

centered on the relationship between an author of a computer

program (or group of programs) and the person for whom the program

was written.

Once the many factual issues are resolved the essential

question to be determined is the scope of protection available to

the author of computer software in contract and under the

provisions of the Copyright Act, R.S.C., c. C-30.

A host of subsidiary issues arise from and around these

general issues. These will be referred to in turn within the

context in which they arise.

It is important to a proper consideration of the issues in

this case to place the relationship of the present parties in the

historical context in which they arose. In this sense I refer to

the development of microcomputer technology itself, which burst on

the scene in the early 1970's and began to revolutionize the data

processing industry and then set in motion a wave of change that

has reached into virtually every aspect of our present lives. The

extent of that development and the present level of our

familiarity and use of the technology serves to perhaps obscure

- 3 -

the rate at which these changes occurred. A proper consideration

of the factual disputes in the present case requires a realization

of the fact that the relationship of the parties began during the

infancy of this technology, at a time of quite extraordinary

change and development.

The first Apple I computer was demonstrated in July 1976.

This was followed by the introduction of the Apple II in 1977. On

August 12, 1981 IBM announced the introduction of what it

characterized as the IBM Personal Computer a product which soon

became known as the "PC". When the parties began their discussions

in August of 1985, the discussions were aimed at developing

software to be utilized on a computer product that from it's

inception had existed for some four years.

II. BACKGROUND

A.

THE PARTIES

1.

The defendant Hospital Medical Records Institute ("HMRI") is

a federally chartered non-profit company. It is governed by a

HMRI

board of directors elected by its members, who are drawn from

hospitals, provincial hospital associations, and provincial

government health ministries.

It funds itself in large part by

- 4 -

charging fees to the hospitals which make use of its services. It

occasionally receives government and other funding for designated

projects, but its normal activities are not government funded.

HMRI maintains and operates a health care information

database on a mainframe computer.

information in charts which contain details of the particular

patient's visits, diagnosis, treatment and procedures. The

Hospitals maintain patient

Ministries of Health in some provinces require hospitals to

collect and report specific data concerning patient visits and

treatment after a patient is discharged. This selection of data

is called an "abstract" and is taken from the medical records

maintained with respect to each patient. The information is coded

based in part on a system developed by the World Health

Organization called the International Classification of Diseases,

Version 9 ("ICD-9").

medical records abstracts for patients following either the death

or discharge of the patient from hospital. HMRI accepts these

abstracts in either "paper" or computerized form. The abstracts

Participating hospitals send to HMRI their

are then added to the HMRI mainframe database. Information from

the HMRI database is used for studying resource utilization,

management reviews, strategic planning, personnel planning,

funding analyses, and medical research.

Collected data also

provides a basis for comparing a particular hospital's data with

national figures or with peer group hospitals.

- 5 -

In addition to charging a per abstract collection fee, HMRI

also charges for certain standard reports to all subscribing

hospitals on their own data on monthly, quarterly and annual

basis. Hospitals may order additional reports on their own data

for management or accreditation purposes.

Various other reports

are issued regularly on a standing order basis. HMRI reports or

raw data are also used by some provinces for funding purposes. By

1985, all Ontario hospitals were required to submit abstracts for

"inpatient separations" (i.e. discharges or deaths) but the

submission of abstracts for patients who were not admitted to

hospitals (i.e. same day surgery and outpatients) was voluntary.

In provinces other than Ontario in 1985, participation in HMRI was

voluntary.

2.

PRISM

The plaintiff Prism had its origin in 1981 when Dr. Eoin Finn

and John Jeffery began to create and market a computerized

abstract collection and reporting system.

In 1981 Finn and

Jeffery met Deanna Hickey, the British Columbia representative of

HMRI, who introduced them to medical record abstracting. Finn and

Jeffery developed an abstracting system for use on Ohio Scientific

(OSI) computers. Ms. Hickey viewed prototypes of the system and

provided advice.

Prism called the OSI software MACS (Medical

Abstract Collection System).

- 6 -

On August 31, 1982, Prism Hospital Software Inc.("Prism") was

incorporated as a British Columbia company and the rights to MACS

were transferred by Finn and Jeffery to the corporation. Between

August and October 1982, MACS was installed at two hospitals in

Vancouver, Grace and Shaugnessy.

commenced work on an enhancement of MACS for installation at

Vancouver General Hospital ("VGH"). VGH wished to collect

At about the same time, Prism

information for its own use in excess of the HMRI requirements and

it also wanted multi-user capability (the ability to have more

than one person working on the system at the same time).

The

project involved ongoing discussions between Prism and VGH staff

members, and the resultant enhanced version of MACS was installed

in VGH in 1983 and called SuperMACS.

In 1982, Edgar Trueit joined Prism, and in October 1983 Mr.

Jeffery and Mr. Trueit bought out Dr. Finn's interest in the

company.

hospitals in the Vancouver area in a multi-user format, so that

both hospitals could share data on the same system. In 1985,

In 1983, Prism installed OSI SuperMACs at two more

Prism secured a contract to instal SuperMACS at St. Paul's

Hospital in Vancouver. SuperMACS was first written in a computer

language known as OSI BASIC and operated on OSI computers, the St.

Paul's Hospital version had to be converted to run on that

hospital's VAX computer in the VAX BASIC language.

Work on the

- 7 -

conversion commenced in June 1985 and was completed in October of

that year.

VGH replaced its OSI computer with a VAX computer in 1986 and

Prism installed VAX SuperMACS and, in February 1987, signed a

contract with VGH granting to the hospital a perpetual licence to

use VAX SuperMACS for the sum of one dollar.

In the licence

agreement, VGH acknowledged Prism's ownership in the VAX based

SuperMACS system comprising the licenced programs as described in

a schedule.

B.

FORMATION OF THE CONTRACT

From about 1981 onward, HMRI encouraged member hospitals to

submit medical record abstracts to it in electronic (computerized)

rather than paper form. HMRI recognized that electronic

submission had advantages over paper submission: data is easier to

enter into the mainframe computer; computer inputting systems

could have built in "edits" to validate the data being entered so

as to catch errors at source; and hospitals with computerized

abstracting systems could create their own databases, giving them

the ability to manipulate and analyze data in-house.

Prior to 1985, most hospitals still submitted abstracts on

paper. Personnel known as "coders" would complete paper abstract

- 8 -

forms ("Form 780") and submit them in monthly batches.

Some

hospitals had developed in-house computer software to collect and

submit abstracts, and some others had licenced software from

software vendors.

Some vendors licenced computer systems which

were "stand alone" (such as Prism), while others provided systems

which were part of larger hospital-wide information systems (HIS).

Paper abstracts were optically scanned by IBM under contract

with HMRI, converting the information to electronic form for entry

into HMRI's database.

In early 1985, IBM gave notice that it

After the

intended to withdraw the optical scanning service.

withdrawal of the optical scanning service, paper abstracts had to

be "keypunched" manually at HMRI to convert the data into

electronic form. The announcement by IBM of the discontinuance of

its optical scanning service spurred HMRI to increase its efforts

to convince member hospitals to submit abstracts electronically.

By 1985, DOS-based microcomputers such as the IBM PC-XT were

becoming a de facto industry standard.

They offered increased

computing power at a lower cost than was previously possible and

so provided a viable alternative to larger computers which for

many hospitals had been prohibitively expensive. With the advent

of affordable, small in-house computers, HMRI decided to contract

for

abstracting system which it would distribute.

envisaged distributing such a medical records abstracting system

the

preparation

of

a

computer-based

medical

records

It initially

- 9 -

to hospitals which submitted more than 6,000 abstracts per year.

The data to be supplied to HMRI from the computer system, as

well as allowable values for each data field, were already defined

in detail in HMRI's abstract specifications. HMRI sought a system

which would collect the data, validate most of it against

allowable values through the appropriate edits, and submit it for

entry into the mainframe computer. HMRI looked for a vendor which

was already in the medical abstracting marketplace as such a

vendor would be familiar with at least some of HMRI's

requirements.

HMRI

discussions resulted in approaches being made to two vendors of

software, one of which was Prism. Discussions with the other

vendor were not fruitful, and in August 1985 HMRI entered into

serious negotiations with Prism. Prism offered to provide a

did

not

formally

screen

vendors,

but

informal

version of SuperMACS to operate on PC (DOS) microcomputers. The

intent was that Prism would licence the software which HMRI would

then sublicense to hospitals.

The system, which was to be

marketed under the name Savoir-Faire, was intended to be usable by

first-time computer users with no special training.

The discussions between Prism and HMRI culminated in a

contract reflected in Prism's proposal of September 20, 1985 ("the

- 10 -

Proposal") and a licence agreement ("the Licence Agreement"). The

Proposal and Licencing Agreement together are termed "the Contract

Documents". Although the Licence Agreement was not executed until

April 1986, in the fall and winter of 1985 HMRI funded Prism's

acquisition of microcomputer equipment and Prism began the

conversion of SuperMACS to a PC version.

Because HMRI intended to market the software in six other

provinces as well as Ontario, it had to incorporate different

requirements for each of seven provinces with two versions for

Quebec (French and English).

own particular reporting requirements, which resulted in specific

additional data being collected by HMRI for that province. As

Each provincial government had its

both the Prism system and the system which HMRI later wrote in the

Q-Pro-4 language were marketed under the name "Savoir-Faire", the

Prism system will be referred to as "Prism Savoir-Faire" and the

HMRI system will be referred to as "Q-Pro-4 Savoir-Faire".

C.

DEVELOPMENT OF PRISM SAVOIR-FAIRE

Under the contract, Prism was to make its best efforts to

provide a demonstration version of Savoir-Faire in November 1985

for HMRI to show in early December, and working versions

incorporating

edit

requirements

of

different

provinces

in

accordance with a schedule set out on page three of the Proposal.

- 11 -

In fact, Prism provided a demonstration version of the software

in early December 1985, and progressive working versions of the

software from February 1986 until January 1987. The first working

version was provided for Ontario, and each succeeding version

either added a province or corrected problems found in the

preceding version. Thus, each succeeding version incorporated all

provinces and functionality of the preceding versions and was

designed to replace all preceding versions in all provinces so

that only one version would be current.

In all, six versions of the software were created,

incorporating the following provinces in the specified order:

first Ontario; second Quebec-English; third Quebec-French; fourth

British Columbia; fifth Saskatchewan and Newfoundland; and sixth

Prince Edward Island and Nova Scotia.

The sixth version had in

turn six "releases" which added functionality or corrected errors.

These six releases of Version 6 were issued between August 1986

and January 1987.

Prism delivered the final version of Prism

Savoir-Faire to HMRI on January 22, 1987. In accordance with the

contract, it was paid $10,000 for each new version of the program.

The development of a customized software program normally

requires that various releases of the software be produced and

tested before a satisfactory version is achieved.

The only

reference in the Contract Documents to testing is found in the

- 12 -

Proposal at page three (entitled "Delivery Schedule"), which sets

out a schedule of deliveries by Prism to HMRI for "Beta testing".

HMRI relies on the industry-standard definition of "Beta testing"

to argue that Prism was to provide a "turn-key" product i.e. a

product that required only a one-time test to see how a user

reacted to the new product.

Prism contends that because of the

tight schedule for delivery of the provincial versions and because

of HMRI's intimate knowledge of the different provincial reporting

requirements, HMRI agreed that Prism would concentrate on

producing the software and would conduct only cursory tests,

whereas HMRI would test each version extensively so as to satisfy

itself that the software was suitable for release.

The versions provided by Prism to HMRI were tested by a

member of the HMRI staff, Wendy King.

She tested the different

versions as she received them and reported any problems to Prism.

Prism would then correct the problems in its next version. Prism

maintains that such problems as were reported to it were to be

expected in a system of the size and complexity of Savoir-Faire

and that it acted expeditiously to resolve those problems of which

it became aware.

HMRI says that it expected to receive a

conversion of the SuperMACS program, that SuperMACS had been

characterized by Prism as a tested and operative system, and that

the number and nature of the problems were totally unreasonable.

HMRI says that certain of the problems were eventually solved by

- 13 -

Prism while others were not.

The Contract Documents did not explicitly discuss whether the

software system was to have multi-user capability. However, the

parties agreed that multi-user capability be provided in an early

version of Savoir-Faire. A total of six multi-user Prism Savoir-

Faire systems were installed in hospitals in Ontario and

Saskatchewan. HMRI alleges that the multi-user versions of Prism

Savoir-Faire were particularly prone to problems. Prism maintains

that the multi-user function worked correctly as delivered to HMRI

but that, as a result of inappropriate changes to the software by

HMRI, it functioned poorly at some user sites and, in some

instances, it did not function at all.

Prism says that the

single-user capability of Savoir-Faire functioned adequately by

the last release of the last version to HMRI.

HMRI says that in converting or "porting" SuperMACS from OSI

BASIC language to VAX BASIC language for use on a VAX computer,

and then to a BASIC language used by PC microcomputers (in this

case, QUICK BASIC), the program was not competently designed by

Prism for the DOS-based microcomputer environment. Prism contends

that its program did make appropriate use of the capabilities of

DOS-based microcomputers as they existed at the time (1985 and

1986). HMRI contends that none of the Prism programming team was

experienced in microcomputer DOS; Prism says its programmers had

- 14 -

sufficient experience with DOS and with computer programming

generally to write programs for DOS-based microcomputers.

HMRI criticizes various decisions made by the Prism

programming team during the course of developing Prism Savoir-

Faire, such as the choice of file manager product, computer

programming language, multi-user hardware, and other components of

the system. HMRI says these choices resulted in various problems

and limitations with the program and that they were the result of

the inexperience and incompetence of the programming team. In all

cases, Prism denies any alleged incompetence or inexperience of

the programming team and defends their choices as appropriate and

resulting in no significant problems or limitations.

Prism contends that the HMRI staff assigned to test and

install Savoir-Faire and deal with users after installation,

principally Wendy King, were inadequately trained in PC computer

use generally and had no experience with programming, as a result

they did not and could not accurately describe to Prism the nature

of the problems being encountered. HMRI contends that its staff

were appropriately trained and that Prism expressed satisfaction

with the work of Wendy King in particular. Prism contends that on

many occasions it attempted unsuccessfully to establish with HMRI

the protocol necessary for testing, maintenance and ongoing

maintenance. HMRI denies this allegation.

- 15 -

HMRI says that by June 1986, the volume of problems reported

by users was such that HMRI delayed installations for about six

weeks so that existing hospital sites could receive upgraded

software. It stopped marketing Savoir-Faire during this period as

well. Prism says that it was given no notice of an extraordinary

volume of problems or of any decision by HMRI to stop installing

or marketing Savoir-Faire. HMRI says that Prism was given notice

and acquiesced.

Prism was made aware of problems with certain Savoir-Faire

multi-user sites although it says it was not informed of all of

the problems later alleged, that it did not receive all relevant

information, and that the information it did receive did not come

in a timely fashion.

It says that while it could resolve some

Prism

problems, others occurred without apparent explanation.

alleges that during preparation for trial it discovered that the

explanation for the most serious problem was that HMRI had made

inappropriate changes to Savoir-Faire for use in the problematic

sites. HMRI denies making changes to Prism Savoir-Faire.

Prism contends that it offered to maintain the problematic

multi-user sites, but that HMRI rejected the offer.

Prism says that HMRI denied Prism direct access to the sites,

making it impossible to identify and solve the problems. HMRI

Further,

- 16 -

says that it gave Prism extensive access to the site having most

problems, Scarborough General Hospital, including paying for Mr.

Jeffery's travel and expenses on several occasions, but that this

access was discontinued at the request of the site concerned,

which complained about Jeffery's work habits and interference with

their staff.

HMRI's request in June 1986 and denies that it was given extensive

access to the site otherwise. Prism further denies any valid

Prism says it provided support at Scarborough at

concerns with Jeffery's work habits and disputes that he

interfered with Scarborough staff.

Under the contract, Prism was to provide software maintenance

until October 1, 1986. Thereafter, Prism was prepared to provide

software maintenance by way of a yearly contract, but HMRI had the

option of taking over maintenance itself. In September 1986, HMRI

hired Steve Imbert as a senior programmer/analyst.

Part of his

duties were expected to be the maintenance of Prism Savoir-Faire

once it became a stable product. HMRI maintains that although it

wanted to take over maintenance of the Prism software on October

1, 1986, it could not do so since a final release of the sixth and

last version had not yet been delivered by Prism. HMRI says that

the parties never contemplated that maintenance would start before

the final version of the software was delivered, and points out

that the Proposal states that the final delivery deadline was

August 1986.

- 17 -

Prism maintains that HMRI had the ability to assume

maintenance of the software as delivered by October 1986, and

could reasonably have maintained the releases of Version 6 which

were delivered thereafter. It further says that the October date

for maintenance was a fixed date and that it had delivered

versions for all provinces as called for in the contract by August

1986. It alleges that delays thereafter arose from HMRI's delays

in testing. The first release of Version 6 was delivered August

20, 1986. Release 2 was sent to HMRI on September 28, 1986 along

with some of the "source code" of the program in anticipation of

HMRI taking over software maintenance.

HMRI says that the second release of the final version of

Prism Savoir-Faire (i.e. Version 6 - Release 2) had a much higher

level of problems than any previous release or version delivered

up to that point.

Programmer Steve Imbert, who had joined HMRI

about two weeks before the September 28 delivery of Version 6 -

Release 2, concluded that Prism had lost control of the program

and therefore recommended that HMRI authorize him to write a

prototype replacement for Prism Savoir-Faire in Q-Pro-4, which he

said was a more modern computer language that would not cause the

problems exhibited by Prism Savoir-Faire.

HMRI acceded to that

recommendation. Prism says that the nature and frequency of the

problems with Prism Savoir-Faire was not such as to justify the

- 18 -

decision to write Q-Pro-4 Savoir-Faire.

D.

EVENTS SURROUNDING AND FOLLOWING THE REWRITE

HMRI says that Imbert was initially authorized to write

Savoir-Faire in Q-Pro-4 only as a contingency in case Prism was

not able to solve problems with Prism Savoir-Faire. Prism

maintains that from the moment HMRI authorized Imbert to rewrite

Savoir-Faire, it intended to replace the Prism product at all

current and future user sites. Prism alleges that HMRI breached

the Licence Agreement by authorizing Imbert to write Q-Pro-4

Savoir-Faire.

On October 9, 1986 Prism delivered version 86.6 release 3 to

HMRI.

for the multi-user portion of Savoir-Faire had become corrupt in

Prism's computer. In mid-October 1986 John Jeffery attended at

This release could not be installed because a subroutine

HMRI's offices to give a training session. It was intended that

Prism give HMRI all of the source code for Prism Savoir-Faire.

The purpose of the training session and provision of the source

code

was

to

facilitate

HMRI's

assumption

of

maintenance

responsibilities for Savoir-Faire.

HMRI contends that at this training session Jeffery disclosed

that not all of the programs were written in BASIC, but that some

- 19 -

were written in ASSEMBLER. HMRI says that Jeffery stated at the

training session that the programs written in ASSEMBLER were

"Prism proprietary" and that he would not provide copies of the

source code to HMRI.

Prism contends that it advised HMRI well

before the delivery of version 86.1 that some routines were

written in ASSEMBLER, so that HMRI knew this fact long before the

training session in October 1986. Prism says that Jeffery brought

with him to the training session all of the source code for the

Savoir-Faire system.

During the training session, Jeffery did not deliver all of

the source code to HMRI, but withheld source code for some

subroutines.

Prism says he did so because specific detailed

questioning relating to certain parts of the Prism software and

the refusal of HMRI staff to hear a complete overview of the

system caused him concern that HMRI might not intend to deal

fairly with Prism. Prism says Jeffery believes he sent the source

code for those subroutines a short time later.

HMRI denies

receiving this source code. HMRI says it was impossible to

maintain the software without the complete source code.

Throughout 1986, HMRI had discussions with the Commission of

Professional Hospital Activities (CPHA), the American medical

records abstract collecting system, with a view to HMRI taking

over CPHA's Canadian hospital members.

These discussions

- 20 -

culminated in an agreement reached in September 1986 whereby CPHA

would stage a phased withdrawal from Canadian provinces starting

with Alberta on April 1, 1987 to be followed a year later by New

Brunswick.

The addition of the former CPHA members to the HMRI

database would require HMRI to become capable of receiving

abstracts under a new database classification system, ICD-9 CM, in

addition to its existing ICD-9 database. This new system was an

American Modification of the ICD-9 codes with "CM" standing for

Clinical Modification.

On November 5, 1986, HMRI advised Prism that it intended to

maintain Prism Savoir-Faire and itself develop Alberta and New

Brunswick versions. On January 27, 1987 HMRI informed Prism that

it intended to replace the existing Prism programs completely with

a new set of programs. In February 1987, Prism learned that these

new programs would be written in a different computer language, Q-

Pro-4.

The Licence Agreement provided, inter alia, that "HMRI may

modify any Licensed Programs and/or merge such Licensed Programs

into any other program material to form an updated work for the

purposes as detailed in this agreement..." (Article 8).

HMRI

contends that Q-Pro-4 Savoir-Faire for Alberta and, later, for all

other Savoir-Faire sites is a modification as discussed in Article

8 of the Licence Agreement and is therefore authorized by the

- 21 -

licence.

Prism contends that this interpretation of the Licence

Agreement is incorrect and is neither fair nor consistent with the

general intentions of the parties as disclosed in the contract

documents.

The Proposal (page 6) was said to be based on the

understanding that HMRI would "at all times apply their best

efforts toward the marketing and distribution of this software."

The Licence Agreement (Article 14) called for HMRI to make every

"reasonable effort to competitively market" Prism Savoir-Faire.

Prism contends that HMRI breached these conditions by marketing Q-

Pro-4 Savoir-Faire instead of the Prism product and by replacing

Prism Savoir-Faire with the Q-Pro-4 product at sites already

installed.

HMRI says that it was entitled to replace the Prism

product and write its own software for other regions because Prism

Savoir-Faire had so many flaws that it could not be marketed

competitively. In the alternative, HMRI says that Q-Pro-4 Savoir-

Faire is a modification pursuant to the Licence Agreement and that

in making its best efforts to market Q-Pro-4 Savoir-Faire, it did

not breach the contract.

At the time this action commenced, in May 1987, HMRI says

that it had not written Q-Pro-4 Savoir-Faire for any province but

Alberta. It says that installation in Alberta had just commenced

and that no abstracts had been transmitted from Alberta sites, so

- 22 -

no per abstract fees were payable, assuming Alberta installations

come within the Licence Agreement. HMRI says further that it had

paid all per abstract fees which were due to Prism to date on non-

Alberta installations.

Prism disputes that Q-Pro-4 Savoir-Faire

was written first or only for Alberta, and it says that further

amounts for per abstract fee payments are due to it.

HMRI began installing Q-Pro-4 Savoir-Faire in Alberta in

April 1987. It subsequently replaced all Prism Savoir-Faire sites

in other provinces with Q-Pro-4 systems and continues to install

Q-Pro-4 in user sites on an ongoing basis.

Prism has developed

and marketed its own medical record abstracting system called PC

SuperMACS, which it says is its own incarnation of Prism Savoir-

Faire.

The two products, Q-Pro-4 Savoir-Faire and PC SuperMACS,

are currently rated number one and number two respectively among

Canadian "stand alone" medical records abstracting systems in

terms of market share.

The summary of positions taken by each side in this

litigation is simply that, a summary, it is not possible nor

desirable to quote extensively from the 940 pages of written

argument provided by counsel. Instead summaries of the positions

will be provided as necessary within these reasons.

- 23 -

III. THE CONTRACT

A. THE CONSTRUCTION OF CONTRACTS

The defendants in this action have submitted that the

precontract discussions between the parties are of little

significance, if not irrelevant to the determination of the issues

because of the existence of the written contract. The defendants

go on in their submissions to suggest that the written contract ".

. . should be conclusively considered to be the whole agreement

between the parties. . ., and that any ambiguity should be

interpreted against Prism."

With the greatest of respect to the argument presented, it is

not reflective of either the circumstances of this case or the

law.

which vary the terms of the written contract documents or

ambiguous terms within them. Extensive evidence has been

presented on the discussions leading to the contract itself and

the information communicated during those discussions. It is of

The plaintiff's do not allege oral collateral agreements

some significance that the licence agreement itself was not

executed until April 5, 1986 although both parties had begun to

act unequivocally to perform the terms of the contract months

before.

The proper approach in the present case is that set out by

- 24 -

Lord Wilberforce in Prenn v. Simmonds [1971] 3 All E.R. 237 at p.

239:

The time has long

passed when agreements, even those under seal,

were isolated from the matrix of facts in

which they were set and interpreted purely on

internal linguistic considerations. There is

no need to appeal here to any modern, anti-

literal, tendencies, for Lord Blackburn's

well-known judgment in River Wear Comrs v.

Adamson provides ample warrant for a liberal

approach. We must, as he said, enquire beyond

the language and see what the circumstances

were with reference to which the words were

used, and the object, appearing from those

circumstances, which the person using them had

in view.

Moreover, at any rate since 1859

(Macdonald v. Longbottom) it has been clear

enough that evidence of mutually known facts

may be admitted to identify the meaning of a

descriptive term.

The surrounding circumstances and discussions, in this case,

are the backdrop to the agreements reached and formalized in the

contractual documents.

The court is entitled to, indeed,

obligated to look to those circumstances in interpreting the terms

of the agreements reached, and placing them in the factual context

in which they belong.

The general principles to be applied in the interpretation

and construction of a written contract are well known, they

include the following principles:

(a)Words of an agreement are to be

interpreted

in

their

plain,

ordinary and popular sense.

- 25 -

(b)The provisions of the agreement should

be

construed

as

part

of

the

consideration of the agreement as a

whole.

(c)Clauses in contracts are not to be

interpreted in isolation, although

they may be the specific focus of

concern.

(d)An inquiry should be made to ascertain

the surrounding circumstances or

the "factual matrix" in which the

agreement

was

made

because

an

agreement should be construed with

reference to its objects.

Five years after the decision in Prenn v. Simmonds (supra)

Lord Wilberforce had occasion to again address the issue in

Reardon Smith Line Ltd. v. Hansen-Tangen [1976] 3 All E.R. 570 at

p. 574:

No

contracts

are

made in a vacuum: there is always a setting

in which they have to be placed. The nature

of what is legitimate to have regard to is

usually

described

as

"the

surrounding

circumstances" but this phrase is imprecise:

it can be illustrated but hardly defined. In

a commercial contract it is certainly right

that the court should know the commercial

purpose of the contract and this in turn

presupposes knowledge of the genesis of the

transaction, the background, the context, the

market in which the parties are operating.

. . .

It is often said

that, in order to be admissible in aid of

construction, these extrinsic facts must be

within the knowledge of both parties to the

contract, but this requirement should not be

stated in too narrow a sense. When one speaks

of the intention of the parties to the

contract, one is speaking objectively - the

- 26 -

parties cannot themselves give direct evidence

of what their intention was - and what must be

ascertained is what is to be taken as the

intention which reasonable people would have

had if placed in the situation of the parties.

Similarly, when one is speaking of aim, or

object, or commercial purpose, one is speaking

objectively of what reasonable persons would

have in mind in the situation of the parties.

It is in this sense and not in the sense of

constructive notice or of estopping fact that

judges are found using words like "knew or

must be taken to have known".

The consideration of evidence of the setting or the factual

matrix in which a contract was arrived at allows commercial

contracts to be viewed and understood in their business setting.

This process is used to understand the genesis, aim and object of

the transaction which, in turn, represents the context in which

the written agreement should be construed.

In approaching this evidence care must be taken to remember

that the test to be applied is an objective test and not a

subjective determination of the parties intentions.

In other

words, the question to be addressed is what would reasonable

persons have in mind in the situation in which the parties found

themselves and the object of the courts intervention is to find

the intention of the parties as recorded in the document.

The difficulties which arise in this process are again best

expressed in the words of Lord Wilberforce from Prenn v. Simmonds

- 27 -

at p. 241 and flow from the fact that:

. . . The words used may, and often do,

represent a formula which means different

things to each side, yet may be accepted

because

that

is

the

only

way

to

get

"agreement" and in the hope that disputes will

not arise. The only course then can be to try

to ascertain the "natural" meaning. Far more,

and indeed totally, dangerous is it to admit

evidence of one party's objective - even if

this is known to the other party.

However

strongly pursued this may be, the other party

may only be willing to give it partial

recognition, and in a world of give and take,

men often have to be satisfied with less than

they want. So, again, it would be a matter of

speculation how far the common intention was

that the particular objective should be

realized.

B.

THE LICENCE AGREEMENT

The contract documents in this case consist of two

documents; a written proposal dated September 20, 1985 and

executed on behalf of the defendant HMRI on January 29, 1986 and

the licence agreement executed on April 5, 1986.

The licence agreement provides:

LICENCE AGREEMENT

THIS

day of

LICENSE

, 1985

AGREEMENT made as of the

BETWEEN:

PRISM HOSPITAL SOFTWARE INC., a company

duly incorporated pursuant to the

laws of the Province of British

Columbia, and having its registered

office at Suite 950, 999 West

Broadway, in the City of Vancouver,

in

the

Province

of

British

Columbia,

- 28 -

(hereinafter

OF THE FIRST PART

referred to as "PRISM")

AND:

HOSPITAL MEDICAL RECORDS INSTITUTE OF

CANADA,

(hereinafter

referred to as "HMRI")

OF THE SECOND PART

WHEREAS PRISM is the

owner of certain licensed programs, all as

more particularly described in Schedule "B"

attached hereto and forming part of this

Agreement and which are hereinafter referred

to as the "Licensed Programs" which it is

prepared to license to HMRI, all on the terms

and conditions hereinafter set forth;

NOW, THEREFORE, THIS

AGREEMENT WITNESSETH that in consideration of

the mutual covenants and promises hereinafter

contained to be performed by the respective

parties hereto, the parties do covenant and

agree with each other as follows:

1.

PRISM hereby grants

a non-transferable, non-exclusive license to

HMRI to use the Licensed Programs in its

business operations in Canada and to sub-

license the Licensed Programs to HMRI's

Canadian

conditions of this Agreement.

clients

under

the

terms

and

2.

It is understood and

agreed that this Agreement shall remain

effective from the date on which it is

executed

by

the

parties

hereto

unless

terminated by PRISM for non-compliance by HMRI

of the terms and conditions hereof.

3.

It is understood and

agreed that upon execution of this Agreement,

that HMRI shall pay to PRISM certain fees, all

of which are more particularly described in

Schedule "A" attached hereto and forming part

- 29 -

of this Agreement.

4.

In addition to the

fees due hereunder, HMRI agrees to pay amounts

equal to any taxes resulting from this

Agreement

or

the

activities

hereunder

exclusive of any tax based on the net income

of PRISM.

5.

HMRI

shall

be

entitled to use the Licensed Programs granted

hereunder in connection with its business

operations in Canada but it is expressly

understood and agreed that HMRI shall not

sell, lease, sub-lease, market or otherwise

make available in any manner the source code

for such Licensed Programs to any person or

entity without the prior written consent of

PRISM and in the event HMRI does so sell,

lease, sub-lease or otherwise make available

in any manner the source code for such

Licensed Programs, PRISM may terminate this

Agreement forthwith.

6.

HMRI

is

hereby

authorized to use the Licensed Programs in

compiled, object code form on its data

processing equipment in the course of their

business operations in Canada. HMRI is hereby

also authorized to sub-license the Licensed

Programs in compiled, object code from to

their Canadian clients for the client's use on

their data processing equipment in the course

of their business operation in Canada.

Save

and except for these purposes, HMRI shall not

have the right to use, print, copy or

otherwise make available in any manner the

compiled

object

code

for

such

Licensed

Programs or other such material as are

supplied hereunder.

7.

The

shipment

or

delivery date for the Licensed Programs will

be specified by PRISM and PRISM's best efforts

will be made to meet those delivery dates.

However, PRISM does not represent or warrant

that the shipment dates will be met.

shipments will be made on standard PC readable

diskettes unless otherwise specified.

All

8.

HMRI may modify any

- 30 -

Licensed Programs and/or merge such Licensed

Programs into any other program material to

form an updated work for the purpose as

detailed in this Agreement, provided that in

the

event

the

Licensed

Programs

are

discontinued as a result of default by HMRI

hereunder,

associated

the

materials

Licensed

will

Programs

be completely

and

removed from the updated work and dealt with

under this Agreement as if permission to

modify and/or merge had never been granted.

9.

Any portion of the

Licensed Programs and/or associated materials

included in such updated work will continue to

be subject to the terms and conditions of this

Agreement.

10.

HMRI will take the

appropriate action by instruction, agreement

or otherwise with any persons permitted access

to the Licensed Programs and/or associated

material so as to enable HMRI to satisfy

HMRI's obligations under this Agreement.

11.

All copies of the

Licensed Programs made by HMRI including

translations, compilations and partial copies

with modified or updated work are the property

of PRISM, HMRI will reproduce and include the

copyright notice on any such copies in

accordance with such copyright instructions as

may be provided to HMRI by PRISM.

12.

The

Licensed

Programs and related material as supplied to

HMRI are warrantied to meet their respective

official published HMRI specifications and

software maintenance will be provided until

the 1st day of October 1986.

13.

In the event of any

breach or default on the part of PRISM in

connection with this Agreement or the subject

matter thereof in such circumstances as would

otherwise entitle HMRI in law to damages

and/or to be discharged or released herefrom

and whether or not any such breach or default

on the part of PRISM is in the nature of a

breach of express condition or breach of

fundamental

term(s)

or

a

fundamental

- 31 -

breach(es) or otherwise whatsoever in law, the

parties hereto having specifically considered

the matter, expressly agree that PRISM shall

not be liable for damages of any nature

whatsoever whether direct, special, indirect,

consequential or otherwise even if PRISM has

been advised of the possibility thereof

(including but not limited to increased

expenses, failure to realize expected savings,

loss of goodwill, loss of profits or any other

economic loss of any kind).

14.

HMRI

hereby

expressly agrees to make every reasonable

effort to competitively market the Licensed

Programs and/or related material in every

Province and Territory in Canada as the

respective versions of the Licensed Programs

become available to HMRI.

15.

Save and except for

the provisions contained in the Proposal made

the 20th day of September, 1985 this Agreement

constitutes

the

complete

and

exclusive

statement of the agreement between HMRI and

PRISM and supercedes all other proposals oral

or written, and all other communication

between the parties related to the subject

matter

of

this

Agreement

and

all

representations, conditions, warranties or

collateral agreements express or implied,

statutory or otherwise with respect to the

Licensed Programs supplied hereunder other

than as contained hereunder.

16.

This agreement shall

be deemed to have been made in the Province of

British Columbia and shall be governed by and

construed in accordance with the laws of the

Province of British Columbia.

17.

This agreement shall

enure to the benefit of and be binding upon

the parties hereto, their respective heirs,

executors,

assigns.

administrators,

successors

and

IN WITNESS WHEREOF

the parties hereto have hereunto executed

these present as of the day and year first

above written.

- 32 -

This license agreement is unusual in that by means of Article

15 it specifically preserves the provisions of the Proposal from

the general exclusionary clause which perfects the written

agreement.

As a result the proposal constitutes a part of the

contract documents themselves. This document does much to set the

factual matrix in which this agreement was reached and casts

considerable light upon the "expectations" of the parties, it goes

further, however, in that in the unusual facts of this case it

takes much of the matrix of facts and elevates them to the level

of having contractual force.

Prism was at the time of the agreement a small two man

company engaged in the business of providing medical records

abstracting software to hospitals in British Columbia.

In that

province it had met with some modest success. It had not been

involved in the creation, prior to the present project, of such a

system for PC microcomputers. HMRI was a much larger organization

whose business operations centered on the operation of its

mainframe computer database. As a result it had within its ranks

a group of people whose expertise was in the area of computers and

database systems, albeit in relation to large mainframe computers.

As expressed in the proposal (paragraph 1) and other evidence

led at trial HMRI wished to see developed and distributed a P.C.

- 33 -

based abstracting system which would reduce the number of paper

abstracts it was receiving from hospitals in Canada. In part the

timing of the approach to Prism was spurred by the withdrawal of

the scanning equipment it had been using to process paper

abstracts.

IV

THE RELATIONSHIP BETWEEN PRISM AND HMRI

It is clear on the evidence and I specifically find that HMRI

was aware of the existing products Prism had available and of the

computer platforms on which they ran. They were fully aware of

Prism's experience and that of its principals and all parties

appreciated that this project was a major undertaking, in a new

area aimed at providing a low cost system to hospitals to enable

them to submit data electronically to HMRI. The objective of the

project is well set out in the last paragraph of part 1 of the

proposal which says:

The

"SuperMACS" system which will be based on the

"PC" type of micro-computer. The system is

proposal

is

to

include

a

MINIMAL

intended to satify(sic) the Medical Records

data collection and basic reporting needs of

members who currently complete, in the range

of, 6,000 to over 30,000 abstracts per year.

These members are situated in Canada and each

Province involved has unique requirements,

additionally in the case of Quebec, both

English and French must be considered.

It is equally obvious that the project was a major commitment

- 34 -

of the resources of Prism and that their interest arose not simply

from the remuneration flowing from the agreement but in other

benefits they foresaw. The opening paragraphs of part 15 of the

proposal make this clear:

15.

THE FUNDING

Prism is prepared to invest expertise, time,

effort, and material toward the successful

completion, marketing and sales of this

product. In the process we are also willing

to share some of the risk which is involved,

with HMRI. Prism will provide the services at

significantly less than normal rates and HMRI

will provide the up-front funding to start the

project.

As you know Prism has developed, installed,

and is supporting the "SuperMACS" and "MACS"

micro-computer based abstracting and reporting

systems in B.C. We are also now in the final

stages of completing the development of a DEC

VAX based Medical Records system (also for

B.C.) which will be installed at St. Paul's in

early October. We have approached others with

this package and we have plans to prepare it

for use in other provinces.

Because of our involvement in the development

and sale of our systems, and our "built-in"

expertise, we are confident that we can

satisfy your needs for a quality product

within the compressed schedule previously

described.

However, that same schedule also

requires that the principals of Prism devote

all of their efforts toward this project which

in turn may require that we hire staff to

assist us with our present client base.

Prism was going to provide the services at "significantly

less than normal rates. . ." and HMRI was going to provide ". . .

the up-front funding to start the project."

In fact the first

- 35 -

portion of this funding was the purchase by PRISM at HMRI's

expense of a microcomputer system to be used in the development of

the software.

In many respects the reality of the situation was that Prism

and HMRI embarked on a joint venture to develop a low cost PC

based medical abstracting system which would then be distributed

by HMRI and placed in hospitals across Canada.

The benefit

flowing to HMRI was the conversion of paper abstracts to

electronic media with all the error verification and time saving

that would flow from that conversion.

The benefit flowing to

Prism was primarily the market penetration they would achieve with

the use of HMRI's "clout" in the marketplace and the placing in

hospitals of a "minimal" version of their software which they

could then enhance or add features to at extra cost. These were

the commercial purposes of the transaction.

The joint nature of the project was evident throughout the

early months of the project, in many ways, from the early funding

provided by HMRI, the structure of the payments agreed upon, the

pattern of testing which developed between Prism and HMRI, and the

division of labour including the preparation of the user

documentation by HMRI, and the way in which both installation and

customer contact was handled.

Many of these areas will be

canvassed more fully later in these reasons.

- 36 -

Key to the relationship between the parties was the

acknowledgment by HMRI of Prism's ownership of the programs,

including ". . . translations, compilations, and partial copies

with modified or updated work. . .". The license granted to HMRI

was to "use the Licensed Program in its business operations in

Canada and to sub-license the Licensed Programs to HMRI's Canadian

clients" unless the agreement was terminated for non-compliance by

HMRI.

The scope of the license was further defined in Articles 5

and 6 of the License agreement and limitations on the use of

either the object or source code are imposed.

The heart of the

agreement is that HMRI could use the programs on its data

processing equipment in Canada and sublicense the programs to its

Canadian client hospitals for use on their data processing

equipment, while recognizing Prism's ownership of the software.

I am unable to accede to the submission that there was any

misconception on the part of HMRI or misrepresentation on the part

of Prism as to the latter's expertise and experience with

microcomputers in general and the PC in particular.

The

suggestion flies in the face of reality and the structure of the

arrangements made between the parties to the agreement, to say

nothing of the process HMRI went through in approaching Prism,

- 37 -

assessing their product and selecting them.

The process is reflected in the records of HMRI, particularly

the minutes of the Executive Committee, at the relevant times. At

their meeting of September 16, 1985 the executive committee had

before them for consideration a report entitled "Overview of

Scanner Elimination and Staff Recommendation.

This sixteen page

report is far from a cursory treatment of the decision facing

HMRI, it is a detailed consideration of the options, the

advantages to be derived by HMRI, the marketing prospects for the

system and a clear recommendation that the "ABC software" (Prism)

should be selected.

The report was prepared by J. Bickle, the Assistant Executive

Director. In attendance at the meeting and speaking in favour of

the staff recommendation was, among others, John Stenabaugh, the

Director of Data Processing for HMRI.

Mr. Stenabaugh among his

responsibilities at HMRI was in charge of programming and

maintenance activities of the programs running on the mainframe

computer used by HMRI. It is obvious from even a cursory reading

of these minutes and the report that the review was carried out

drawing upon the experience and expertise of various departments

at HMRI, these included the Liaison staff, the marketing staff,

and the people responsible for their computer operations.

- 38 -

Ms. Bickle on page 2 of her covering letter refers to the

then existing Prism software in these terms:

An

excellent

software package has been developed by the

"ABC" Company. HMRI Liaison staff is familiar

with the product and agree that it is superior

to all others.

HMRI. . .

The ABC Company has visited

The executive committee approved the proposal in principle

and authorized the staff to negotiate the necessary contracts.

Both Prism and HMRI were aware that the preparation or

conversion of the Prism software to run on a PC was a major

undertaking and one which, in all likelihood would take some time

and require some considerable effort and cooperation from both

Prism and HMRI. The first act of HMRI following the conclusion of

the agreement was to pay for Prism's acquisition of a PC to be

utilized to develop the software.

It was obvious that such an

acquisition reflected the "new" nature of the project and the fact

that Prism did not have the necessary hardware to begin the

development process.

It would also appear that both anticipated potential

difficulties in the project. This inference flows from the fact

that Article 13 of the License Agreement provides that Prism would

not be liable for damages "of any nature whatsoever" arising from

any breach or default arising from the subject matter of their

- 39 -

agreement.

I conclude on all the evidence that both Prism and HMRI

entered on this project with their eyes open and with the

knowledge that it was a complex project.

Each saw substantial

benefits ahead of them if it succeeded and, for a time at least,

each pursued those objectives.

The proposal and the licence agreement, in my view, clearly

disclose a disparate or unequal bargaining position between Prism

and HMRI. This emerges from the recognition within the documents

themselves of Prism's finite resources, HMRI's agreement to the

clause limiting Prism's liability for damages arising from the

project and the incorporation within Article 7 of the licence

agreement of a saving clause with respect to the delivery dates.

These provisions taken together with the overall evidence

concerning the relationship between the parties leads inexorably

to this conclusion.

This was not a case in which preliminary

discussions broke off before an agreement to pursue a joint

venture emerged. Prism was selected by HMRI, precisely because of

the fortuitous combination of its knowledge of medical abstracting

and computers.

The objects of the project were very much in

HMRI's interest because of its size and relationship to hospitals

in most of Canada and the volume of data it was collecting. The

withdrawal of the scanner service provided the impetus for the

- 40 -

project, but HMRI's desire to provide such a system to hospitals

across Canada was precisely the factor that gave value to the

project from Prism's perspective. When Prism's limited resources

created a difficulty HMRI used its greater resources to solve that

difficulty as they did most notably by paying (in advance of even

the execution of the agreements) for the development hardware.

It is notable that many of the plaintiff's submissions in

this action focus on what they assert to be a duty on the part of

HMRI to deal with Prism "in good faith" and the assertion that

they failed to do so.

As matters progressed the relationship between the two

changed. This change, the reasons for it and the legal effect of

these events are at the centre of the dispute between the parties.

Regrettably

the

resolution

of

these

issues

requires

a

determination of credibility issues involving the main witnesses

called by each side and a consideration of the effect, if any, of

what the plaintiff's suggest are:

(a)a failure of the defence to call

certain significant witnesses;

(b)a failure on the part of the defence

to cross-examine in significant and

contentious issues;

(c)a failure on the part of the defence

to

confront

the

plaintiff's

areas they

witnesses

in

subsequently adduced evidence on;

and

- 41 -

(d)the impact of defence counsel asking

leading

questions

in

areas

of

controversy.

The plaintiffs submit that these sins should result in the

drawing of inferences adverse to the defence in the case of (a)

and (b) and the rejection of defence evidence or, alternatively,

the determination that it can be afforded no weight in the case of

(c) and (d).

In approaching these issues one must begin with

a

consideration of the organization and management of HMRI itself.

V

HMRI'S ORGANIZATION AND MANAGEMENT

HMRI is governed by a Board of Directors elected by its

members. They are generally drawn from hospitals, provincial

hospital associations and provincial government health ministries.

At all material times until 1987 HMRI's chief operating officer

was Bob Colman who held the position of Executive Director. The

Assistant Executive Director was Jean Bickle.

HMRI had a number of committees which dealt with different

areas of the organization's operations. At the top was the Board

of Directors, below them there was the Executive Committee. Below

this level there were a series of specialized committee's which

- 42 -

included

the

Personnel

&

Finance

Committee,

the

Product

Development Committee, the Data Base Committee, the Strategic

Issues Committee, the Users Group Committee, the Liaison Committee

and the Savoir-Faire Committee.

Although HMRI was a non-profit organization, this fact did

not prevent, in the plaintiff's submission, it pursuing aggressive

expansion in the Canadian market.

In simplified terms the

plaintiffs submit that it was this expansionist drive which

characterized the organization throughout the material times and

which lay behind their decision to rewrite Prism Savoir-Faire and

cut Prism out of the market.

Much of the evidence on which the plaintiffs rely in support

of this position emerges from documents and, in particular, the

minutes of the various committees. The reasons why the evidence

emerges from these documents is, in part, due to the absence from

the witness box of certain individuals who played important roles

in the events which underlie this dispute. Most notable of these

are the absence of both the Executive Director, Bob Colman, and

the Assistant Executive Director, Jean Bickle.

Both of these

individuals were present at the initial meetings in August of 1985

and held the senior administrative positions throughout the events

with which this action is concerned.

Other members of the HMRI

team who played various roles, and were not called, included

- 43 -

Stephen Dreezer, Deanna Hickey, Lynne Abbott and Barbara McLean.

Mr. Colman's absence is perhaps understandable given the fact

that he appears, on the evidence, to have been the first casualty

of the present dispute.

The decision to abandon Prism Savoir-

Faire came to the attention of the Executive Committee and the

Board of Directors in March of 1987, nearly six months after Mr.

Imbert began his project of writing a Q-Pro-4 version.

The

firestorm which followed, as it emerges from the minutes and other

documents, consumed Mr. Colman and led to his dismissal and his

replacement with Mr. Yungblut, the current president.

What emerges from these documents is critical to an

understanding of the events which preceded the meeting of the

Executive Committee on March 1, 1987 and in understanding the

nature and management of HMRI at the material times.

It is of

substantial importance in the consideration of the issues in this

action that the position taken by the defendants in this action,

who did not produce certain significant witnesses, then seek to

explain the absence of critical commentary on Prism Savoir-Faire

in these minutes by suggesting they were trying to keep the

problems quiet. They don't stop at that, however, they go further

to say that the apparent positive reports contained within the

minutes are not to be believed, and that, in reality, there were

major problems with the software which necessitated the far

- 44 -

reaching decision to rewrite the software "in house" and

eventually to replace it.

The documentary evidence, particularly that which was created

after a dispute arose but before litigation began, must be

approached with caution, but where the defendant is a corporate

body which acts through its officers and employees, the acts and

knowledge of the company is to be found reflected in what those

individuals record in the course of their day to day activities on

behalf of the corporate body.

The minutes in this case were

created by individuals during the performance of their duties with

HMRI, they were in most cases circulated within that organization

and adopted at subsequent meetings, they were clearly in the

possession of HMRI and were both created and maintained in the

ordinary course of their business.

In Ares v. Venner [1970] S.C.R. 608, 14 D.L.R. (3d) 4, the

Supreme Court of Canada considered the effect to be given the

nurses' notes recorded in a set of hospital records.

In stating the opinion of the Court, Hall J. said, at page

16:

Hospital

records,

including

nurses'

notes,

made

contemporaneously by someone having a personal

knowledge of the matters then being recorded

and under a duty to make the entry or record

should be received in evidence as prima facie

- 45 -

proof of the facts stated therein.

This

should, in mo way, preclude a party wishing to

challenge the accuracy of the records or

entries from doing so.

Subsequent decisions have interpreted Ares v. Venner as

having a broader application than simply its consideration of

hospital records.

In Setak Computer Services Corp. v. Burroughs

Business Machines Ltd. (1977) 15 O.R. (2d) 750, 76 D.L.R. (3d)

641, Griffiths, J. specifically considered Ares v. Venner and its

application in

a

case where minutes of meetings between

representatives of the plaintiff and defendant and notes of

internal minutes of meetings of the plaintiffs' personnel were

tendered by the plaintiff. In ruling on their admissibility the

learned trial judge wrote at p. 755:

In my opinion, the common law exception applies only to

writings or records made by a person speaking

from personal observation or knowledge of the

facts recorded. In the Ares case, the nurses

were recording observations they had made of

the plaintiff as a patient and the entries

thereof describe matters within their personal

knowledge.

In my view, the minutes here do

not meet this requirement of the common law

exception; the authors of the minutes did not

have personal knowledge of all the facts

recorded. At most, it can be said that [the

authors of the minutes] had personal knowledge

that the statements attributable to those who

attended the meeting, which were duly recorded

in the minutes, were in fact made, and it is

on this narrow basis only that I would admit

the minutes under the common law as modified

in Ares v. Venner.

In the present case these minutes are tendered by the

- 46 -

plaintiff to show the knowledge of the defendant HMRI through its

various employees and officers and to rebut a specific defence

advanced by the defendants, namely that there were major flaws and

problems

with

the

plaintiffs'

software,

and

that

these

necessitated the decision by HMRI to rewrite the software. They

also go further to demonstrate actual knowledge and, by inference,

lack of knowledge of certain events by particular individuals at

particular times.

In my respectful view, they are receivable for those purposes

given the issues in this action, the absence of certain witnesses,

the nature of the defences advanced, and the way in which these

minutes were created, circulated, adopted and kept by HMRI. They

also, in my view, fall within the provisions and meet the criteria

of s. 48 of the Evidence Act, R.S.B.C. 1979, c.116 as I ruled

during the course of the trial.

At the Executive Committee meeting on March 1, 1987, the tone

of the minutes as they touch on the Savoir-Faire project change

dramatically.

After a review of "the highlights of the Savoir-Faire report'

the minutes record that:

After many and frequently heated discussions on various

topics from the report, it was moved by Mr.

Dreezer, seconded by Dr. Gellman and carried

- 47 -

by the other members present, that Mr.

Stenabaugh and Mr. Imbert were to meet with

Mr. A. Adams and an outside expert of his

choice, to deal with the Savoir-Faire issue.

In its report to the Board of Directors the next day, the

Executive Committee told the Board, apparently for the first time,

that there were problems with Savoir-Faire " . . . in both

hardware and software . . ." The tone of the meetings is apparent

from the recording in the minutes of the Executive Directors'

report:

Mr. Colman said that he was distressed by the

allegations of incompetence directed at him

and

the

staff

throughout

the

Executive

Committee meeting that had been held on Sunday

evening. He wished to assure members of the

board

that

HMRI's

staff

is

skilled,

enthusiastic and competent and that it had an

unusual depth of management support for a

small organization.

He added that with a

staff complement that had not exceeded thirty

nine at any time in the current year, eleven

had to be laid off, two reassigned to other

jobs, two replaced and five hired to maintain

the proper mix of talents required to retain

competency as the institute changes its

mandate and methods of operation.

It is clear and I specifically find that the Board of

Directors and the Executive Committee were unaware until these

meetings, of the decision to rewrite Savoir-Faire and that Mr.

Imbert had been engaged in the project for nearly six months. It

is equally clear that they were unaware of the breakdown of the

relationship with Prism or with the suggestion that there were

- 48 -

serious problems with the software.

An indication of how well the lines of communication were

working is revealed by the minutes of the Executive Committee from

April 6, 1987, more than a month later, where Dr. Patterson, the

Chairman of HMRI's Board of Directors, is reported to have said:

. . . that he had just heard that the HMRI Savoir-Faire

programs were written in Q-Pro-4 language and

not in DBASE-III.

The Executive Committee then received the report of Mr. A.

Adams on "the Savoir-Faire technical issue".

recorded reads:

The report as

Mr. Adams replied that the original program was a user

friendly nontransportable technology but that

he had only recently become aware that DBASE-

III was not being used for the HMRI rewrite

for which Q-Pro-4 had been selected. He added

that Q-Pro-4 was not generally supported:

that it was based on 'C' and that it had some

advantages such as its network interface.

The scope of the problem then existing at HMRI is reflected

in the fact that the Executive Committee set in motion an internal

review of HMRI's ". . . internal management and organizational

structure, and of the relationship between the board, its advisors

and management".

The difficulties were felt elsewhere in the

organization as well, as was reflected in the minutes of the

Personnel and Finance Committee meeting of April 28, 1987, which

note:

- 49 -

Mr. Coleman and Ms. Bickle commented on senior staff's

concern about the well being of the company

because

of

the

current

breakdown

in

communications between the Board, Executive

and HMRI staff.

If there was any doubt about the level of communication it is

laid to rest by the entry in the minutes of the Executive

Committee from May 11, 1987. The review of the previous minutes

led to the following entry:

Mr. Dreezer said that the minutes referring to the use

of the QPRO4 software at the bottom of Page 2

did not relfect[sic] the concern expressed by

the Executive Committee.

There had been

considerable discussion about the risk of

using

QPRO4

as

opposed to DB-III.

Furthermore, the reason for the staff's

decision to use the language should have been

communicated

Committee.

to

HMRI

CEO

and

Executive

Attached to the Minutes of the Board of Directors' Meeting of

May 12, 1987, is Mr. Colman's final report as Executive Director.

The report is revealing, in commenting on what he characterized

as four waves of change at HMRI he concludes:

What is common to all four is that each in combination

has contributed to stress and turmoil at HMRI.

To this trouble has been added the needless

aggravation of accepted methods for getting

things done being either forgotten, bypassed

or abrogated.

I recognize that the days of a stable environment at

HMRI are over, and that managing change is a

long, tough and continuous process that

consumes much energy. The job has been made

more difficult by a failure to communicate.

- 50 -

On the same date, Mr. Whitton, the outside expert, retained

to review the internal management and organization of HMRI

reported to the Executive Committee. The minutes as they record

Mr. Whitton's report record the following weaknesses:

a) A reactive management style

b) Ad hoc decision making

c) No sense of direction

d) Strong resistance to change

e)

A

defensiveness

criticism

unwillingness

to

accept

f) A lack of leadership

g) A lack of good management skills,

resulting in:

- No planning done by staff

- Poor management controls in place

- Poor communication with staff

h)

The advisors to the board and managers is

unclear and creates problems.(sic)

In reply to a question by the Secretary, Mr. Whitton

indicated that the weaknesses as identified

related to the performance of the Executive

Director and he further indicated that he had

not evaluated the other members of the Senior

Management

weaknesses.

Team

in

relation

to

these

Mr. Whitton went on to outline the qualifications for a new

CEO/President and concluded that they should include:

a)Strong executive interface skills

b)Business planning skills

c)A good understanding of technology

- 51 -

d)Good public relations skills, to assist in the selling

of the organization and its products.

e)Good project management skills

f)Good general management skills

g)Leadership qualities

He further stated that, in his opinion, there was not a

member of the management in HMRI currently who

exhibited these skills.

At the conclusion of the meeting the Executive Committee

unanimously accepted the Whitton report and recommended to the

Board of Directors that Mr. Colman's association with HMRI be

ended.

VI

HMRI'S MANAGEMENT OF THE SAVOIR-FAIRE PROJECT

When this project was initiated HMRI assigned responsibility

for the project at their end to Wendy King.

Ms. King had been

employed by HMRI since December 1983 as a liaison representative

where her duties involved providing training to hospitals in the

use of HMRI paper abstracts and interpretation of reports provided

by HMRI. In carrying out these duties she reported to Jean Bickle

who was, at that time, the Director of Marketing and Liaison

Services.

Ms. King's educational background included the

completion of grade 13 followed by two years at a Community

College resulting in her receiving a certificate from the Canadian

College of Health Record Administrators.

Her exposure to

- 52 -

computers during this period was limited to a course in data

processing at the college level.

During a period of employment

with Queensway Carlton Hospital in Ottawa she had some exposure to

the testing of a new version of medical abstracting software

provided by I.S.T. Health Comp.

When the Savoir-Faire project began she was given greatly

expanded responsibilities.

In her examination for discovery she

described it in this way:

I was to help with the marketing of the Prism

product.

I was to develop user manuals for

the Prism product. I was to develop training

sessions for the Prism product. I was to test

the product.

I was to be in charge of

distribution of the product. I was to be in

charge of ordering hardware for the product.

I was to be in charge of shipping the

hardware. I was to be in charge of testing

the hardware.

. . .

I developed -- I guess, I was in charge

of the first O.H.A. demonstrations and booking

the appointments.

releasing the software.

I

guess testing and

Within the scope of these many responsibilities she was

responsible for developing the testing requirements and acted as

the liaison between Prism and HMRI.

These new responsibilities

did not, however, replace her other duties but rather were added

to them. While undertaking the Savoir-Faire project she remained

responsible for liaison services to half the hospitals in Ontario

- 53 -

and a number of other duties. Of more importance to the present

action she was primarily responsible for installing new systems at

hospitals and taking complaints and problems and responding to

them. Regrettably she had neither the training or the experience

to do so.

Curiously, direct responsibility for the Savoir-Faire project

was kept away from the computer side of HMRI and controlled by the

liaison side.

although she was extremely busy she had help available to her when

she needed it. Despite her testimony there is little evidence

In her evidence at trial Ms. King testified that

that she was given technical support from within HMRI.

The dangers inherent in this arrangement were compounded by

HMRI's decision that Prism, with some minor exceptions, was not to

have direct contact with the hospitals.

This left Prism to act

and respond to problems relayed to them by Ms. King who, although

knowledgeable in the area of hospital records, lacked the

technical knowledge to adequately assess and relay the substance

of problems to Prism.

Much time was spent during the course of this trial

addressing the issue of responsibility for testing the software as

it was developed and as each succeeding version was delivered to

Ms. King. On this issue the parties were truly poles apart, Prism

- 54 -

submitted that HMRI took de facto responsibility for the testing

and did so both inadequately and incompetently. HMRI on the other

hand points to the agreements use of the term "beta testing" and

submits that it was only responsible for final "user" testing.

These positions both ignore the reality of the relationship

between the parties and serve to obscure the real issues in this

litigation. Prism was obligated by the terms of the agreement to

provide a product which met the agreed specifications.

In the

circumstances of this case both sides knew that the delivery

schedule would place a significant strain on Prism's limited

resources, as a result HMRI, in fact, conducted much more than the

usual "beta testing." Of perhaps more significance, and harkening

back to the passage quoted earlier from Ms. King's evidence, the

decision to release the software to HMRI's customers following

testing was made by HMRI. This gave to HMRI the ultimate control

of not just the testing they would carry out but when it would be

released to the hospitals.

On the whole of the evidence, it is clear that there were

failures and inadequacies in the testing procedures on both sides

and that the problems which occurred flowed in part from each

side.

What in the end caused the biggest problem in the development

of the Prism software was not the testing procedure itself but the

- 55 -

lack of technical expertise on HMRI's end of the project (Ms.

King) and HMRI's decision to have Prism's contact with the user

sites funneled through Ms. King.

The result of this was to

seriously affect the quality of the information from the field as

it was relayed to Prism.

Ms. King, in struggling with her many responsibilities and

the sheer volume of work sought assistance where she could find

it.

President Computers and Network Data Systems, local suppliers of

hardware with whom HMRI had existing ties. Ms. King during the

This included, on the evidence, outside "experts" such as

course of her evidence showed repeatedly an uncertainty and

hesitation on technical matters which demonstrated in the most

graphic terms her lack of technical expertise.

I have little

doubt that during the initial phases of this project she gave her

best to it, however, her lack of technical background, and the

sheer volume of her work caused her to seek assistance where she

could find it and her decisions were frequently based not on a

rational analysis of problems but on the basis of opinions she

solicited as she could find them.

I specifically find that she

resorted to advice she obtained from individuals at President

Computers at times rather than contacting Prism.

Ms. King in her approach to her decision making on this

project epitomized Mr. Whitton's criticism of Mr. Colman's "ad hoc

- 56 -

decision making." I find that there was at HMRI at the material

time no effective management control of this project and that the

"sense of direction" was provided by Ms. King's uncertainty and

the direction in which the wind was blowing.

In their principal submissions the plaintiff's submit that

HMRI, in essence, utilized the strength of its position and its

control of the project to expropriate Prism's work and pursue it's

own goals.

suggestion that these goals were considered and consciously

pursued. In my respectful view, the evidence does not support

such a suggestion.

Implicitly within those submissions lies the

I find that the criticism leveled by Mr. Whitton describe not

just Mr. Colman but HMRI's whole management at the relevant

times, and in particular their management of this project. The

failure to effectively communicate affected all levels of HMRI's

management as it related to this project and effectively left Ms.

King struggling to cope with a project for which she lacked the

technical background.

This failure to communicate was nowhere more evident than in

the one sided correspondence flowing from Prism to HMRI. Mr. Fisk

valiantly attempted to explain HMRI's consistent inability to

reply in writing to pointed correspondence from Prism by

- 57 -

suggesting that there was no obligation on HMRI's part to reply

and that much of their communication was by telephone.

It is

necessary to return to this issue later in these reasons, for now

suffice it to say that the pattern of correspondence, or rather,

the lack of correspondence from HMRI is yet another indication of

the lack of management and control at their end of this project.

VII THE CREDIBILITY OF WENDY KING AND STEPHEN IMBERT AND THE

"DECISION" TO REWRITE THE SOFTWARE

Credibility is an important issue in this action and, in

particular, it is central to the determination of why HMRI

embarked on the writing of the Q-PRO-4 version of Savoir-Faire.

At the center of this credibility issue is the evidence of Wendy

King and Stephen Imbert. I have already touched on the evidence

of Ms. King and her uncertainty and hesitation during the course

of her evidence. Those terms do not adequately describe the level

of her discomfort during her testimony particularly during cross-

examination. Ms. King as the examination grew pointed frequently

paused for lengthy periods before responding to questions, these

pauses, her inability to answer many questions concerning her

activities, and her emotional upset during her evidence give rise

to considerable concern about her evidence.

It became increasingly apparent during the course of her

evidence that her grasp of technical matters was superficial at

best and was the reason for at least some of the frequent gaps in

- 58 -

her evidence.

The demeanour of a witness is not the sole factor to be

considered in gauging a witnesses' credibility. As O'Halloran,

J.A. said in Faryna v. Chorny [1952] 2 D.L.R. 354 at p. 359:

The credibility of interested witnesses,

particularly in cases of conflict of evidence,

cannot be gauged solely by the test of whether

the personal demeanour of the particular

witness carried conviction of the truth. The

test must reasonably subject his story to an

examination of its consistency with the

probabilities that surround the currently

existing conditions. In short, the real test

of the truth of the story of a witness in such

a

case

must

be

its

harmony

with

the

preponderance of the probabilities which a

practical and informed person would readily

recognize as reasonable in that place and in

those conditions.

satisfactorily appraise the testimony of

quick-minded, experienced and confident

Only thus can a Court

witnesses, and of those shrewd persons adept

in the half-lie and of long and successful

experience in combining skilful exaggeration

with partial suppression of the truth. Again

a witness may testify what he sincerely

believes to be true, but he may be quite

honestly mistaken. For a trial Judge to say

"I believe him because I judge him to be

telling the truth", is to come to a conclusion

on consideration of only half the problem. In

truth it may easily be self-direction of a

dangerous kind.

The evidence of Wendy King and Stephen Imbert must be

assessed particularly within the context of the events surrounding

the writing of the Q-PRO-4 version of Savoir-Faire. This involves

the consideration of a number of issues arising from that

- 59 -

evidence.

A.

STEPHEN IMBERT

Mr. Imbert joined HMRI in September 1986 as a Senior

Programmer/Analyst. At the time he was hired Mr. Imbert's job

description included as a major component the maintenance of Prism

Savoir-Faire. He had obtained a bachelor's degree in economics

and over the next seven years before joining HMRI he had focussed

increasingly on computers. During this period he took a four

month course at the Honeywell Institute which involved three

months of programming and one month of systems analysis. During

this course a portion of his studies involved the BASIC language.

Most of Mr. Imbert's programming experience prior to joining

HMRI involved Q-PRO-4 and it was in that language that he was most

comfortable working.

It was during this initial description of his background that

the first conflict in Mr. Imbert's testimony emerges.

The

discrepancy may be on a minor point but it is a direct

contradiction of two pieces of sworn testimony.

At trial when asked about the use of computers during his

employment with the University of Calgary during 1979 and 1980 he

testified:

"AI did not do any programming in that

- 60 -

used the

job.

I

university

used it for material

computer in my functions at the

job.

I

procurement, that type of thing.

used the university VAX machine."

I

During the course of his examination for discovery Mr. Imbert

was also asked about that job and his use of computers, his

evidence at that time was:

"QWhat job did you obtain at the

University of Calgary in the fall

of 1979?

AThat was -- the name of the job, I was

called a technical assistant.

QTo whom?

ATo -- to the department of physics.

QAnd for what period of time did you work

at the University of Calgary in

that capacity?

AUntil July of the next year.

Q1980?

AH'mmm-h'mmm.

QBriefly, what were your job duties as a

technical assistant?

AVery loosely defined, I would help co-

ordinate physics experiments, build

physics experiments, that type of

thing.

QOkay. Did you use computers?

ANo.

QAnd you didn't do any programming?

ANo."

- 61 -

From this apparently minor discrepancy in sworn testimony

what would become an all too familiar pattern emerged.

In the

questions which followed Mr. Imbert's discovery evidence being put

to him the following exchange took place:

QDo

you recall being asked those

questions and giving those answers?

AI don't -- I don't dispute that I was

asked those questions. I don't

specifically recall them, no, but I

don't

dispute

that

this

is

accurate.

QWere your answers true?

AThe context that I read into this was,

was

I

doing

--

I

working

with

computers, was

working as

a

programmer, and that's the context

that I answered in.

of computers at

Now, the use

University of

Calgary was limited to, as I said,

acting as a user.

QWell, you see, my confusion is that at

discovery in answer to the question

"Did

you

use

computers,"

you

answered "No," and the, "And did

you do -- "And you didn't do any

programming?"

Answer, "no."

At

trial, where in discovery your

answer was "No" to "Did you use

computers," your answer to the

question, "Tell me whether you made

use of computers," you go on to

explain you didn't do programming

but, yes, you did use them?

AYes, I did use the terminal that was in

our office, yes.

QSo that your discovery evidence was not

true; is that fair?

AIf you take it literally, yes, okay?

- 62 -

What I read -- during discovery, I

read into it was he asking me did I

-- was I in the computer field, was

I using computers.

QThank you.

AMy use of computers at the University of

Calgary was very minor.

Three things emerge from this evidence, first a direct

contradiction in Mr. Imbert's sworn evidence, secondly, an

explanation from Mr. Imbert for the contradiction which is

centered on his suggestion that he understood the question

differently than its plain wording, and, thirdly an intelligence

and quickness which was used by Mr. Imbert to manipulate both the

process of the trial and counsel. As cross-examination continued

the number of contradictions and their relative importance

steadily mounted.

As this occurred Mr. Imbert's hostility to

counsel and the court process became more and more apparent.

What emerged from the whole of Mr. Imbert's evidence was the

picture of an intelligent young man who was prepared to change and

shade his evidence when he perceived it to be in his best interest

and who sought to explain away the changes in his testimony by

reinterpreting the questions, asking questions of his own and

generally seeking to focus attention away from the original

question. In short Mr. Imbert was precisely the kind of witness

referred to by O'Halloran, J.A., in Faryna v. Chorny.

- 63 -

Mr. Imbert's capacity for changing history, and his evidence,

emerges more clearly when he testified about what was initially

expected of him in his employment with HMRI. He testified that he

expected to be performing functions in three specific areas, (a)

technical support for HMRI's liaison department in the area of

microcomputers, (b) software support in the sense of superficial

maintenance and annual edit updates on Prism Savoir-Faire, and (c)

some work in COBOL on HMRI's mainframe.

He specifically denied

having understood from his initial meetings that one of his main

functions would be dealing with the hospitals.

Mr. Imbert was confronted with three passages from his

discovery,

once

again

the

passage

which

follows

those

confrontations is revealing:

QWere those answers true?

AYes, if I am allowed to explain the

context?

QPlease do.

AOkay, we are talking about -- during my

job interview, there was concern

about

my

capacity

to

perform

Now,

certain liaison functions.

that's what I was describing in my

three points.

One of the points

was that I was to provide technical

support in that -- I'm not saying

that I -- at the time I wasn't sure

whether I was going to be the guy

picking up the phone and phoning

the hospitals, but the function

here was to provide liaison support

for the hospitals. It was software

- 64 -

support, functionality.

stand by my answers here.

And

I

QBut you see my difficulty, Mr. Imbert,

is that at discovery the impression

I read into those questions and

answers is you are dealing directly

with customers, with people at

hospitals, whereas in your evidence

in this courtroom, it has been:

I

dealt with HMRI staff who in turn

deal with customers, and I also had

these other two areas, maintaining

the Prism software and doing some

programming on the mainframe.

AI worked in concert with HMRI's staff to

provide liaison services to the

hospitals, yes.

QAll right.

We are talking about your

expectations though going into it.

AThat was my expectations in a sense that

my functionality -- at the job

interview here, the concern was

they

wanted

somebody

who

was

technically -- had the technical

background but they wanted somebody

who could speak English.

And I'm

not being -- what I mean by that,

somebody

who

can

convey

the

And

knowledge to a user, okay.

that's the gist of what I am trying

-- that I was getting at.

There

was a big focus in the -- at the

employment agency, that's what the

screening was all about.

Of more importance was Mr. Imbert's evidence concerning his

first few weeks of employment at HMRI. In his direct evidence Mr.

Imbert testified that during his first week there Wendy King was

out of town and so he spent the time being oriented by Lynne

Abbott and John Stenabaugh.

This testimony was consistent with

- 65 -

that already given by Ms. King but, unfortunately, it again

contradicts Mr. Imbert's sworn testimony on discovery.

evidence he testified that he was shown the Prism software by

Wendy King ". . . within my first week of employment." When

In that

confronted with these answers his initial response is brutally

accurate:

QAnd were those answers true?

AIt was -- as true as my recollection was

at the time.

Mr. Imbert's recovery, however, is quick and in his answer to

the next question the sands again begin to shift:

QAnd does your recollection improve with

the passing of time, Mr. Imbert?

AMy recollection improves when you are

preparing for a trial and you're

going through -- going through

material. You are talking a matter

of days here, okay. I think what I

said during the trial was that the

-- yes, during the first week I

spent some time running executable

version of the Prism product. Now,

Wendy King, my recollection right

now is that she was away -- I know

that she was away the first few

days. Now, we are talking down --

you are asking me to be definite on

a day that's five and half years

ago.

following Monday.

it could have been Friday, but my

understanding -- my best

I believe Wendy came in the

Now, whether --

guesstimate, the best recollection

that I have was that she came back

the following week.

- 66 -

Mr. Imbert's previous answer summarizes his evidence

perfectly. This is a witness who was prepared to testify, without

qualification or hesitation, and then change that evidence as his

need or recollection subsequently dictated.

Mr. Imbert, in his

testimony, relied upon his technical knowledge and his wits to

obfuscate, shift and avoid difficult questions seeking in his

explanations to focus attention away from difficulties he

perceived within the evidence.

When his concentration slipped his grasp of the evidence did

as well. At the end of his evidence, in response to a question

directed to him by the Court he returned to his previous evidence:

QAm I to understand that your entire

exposure to the Prism software

during that time period was the

one-hour approximate demonstration

or walk through or user use that

you described as having with Wendy

King?

ANo, it is not.

QWould you please tell me what exposure

you did have?

AI had exposure to the product, as

testified, that in the first week I

spent two or three hours as a user

using the product. I was -- I was

demoed the product by Wendy King,

shown how it worked -- excuse me,

are you talking direct exposure as

opposed to --

Mr. Imbert was

a

programmer with limited experience

- 67 -

programming in the BASIC language. During the first week of his

employment, Wendy King, the co-ordinator of the Savior-Faire

project was away from the office. She returned, it would appear

on the Monday of his second week with HMRI. Mr. Imbert testified

that he reached certain conclusions with respect to the Prism

software. In his direct evidence he expressed it in this way:

AI expressed an opinion that the -- given

the functionality of what this

product did, there was no reason --

there was -- it was -- it was not a

very complicated product to write.

I suggested that there were better

languages out there at the time

that could handle this type of

development.

I suggested that I

could put together a prototype in

Q-Pro-4.

QDid you have -- did you express an

opinion with respect to whether the

product which HMRI then had was in

a

condition where it could be

maintained?

AYes, I did.

I expressed a strong

opinion that it was not in a state

to

be

delivered

that

to

it

HMRI

was

for

not

maintenance,

finished development, and it was

not what was portrayed to me in my

job interview. I was told that it

was going to be a stable product

that required superficial changes

on a fiscal-year basis, and this is

not

understanding was that not all the

provincial versions were even

developed at this point.

what

was

happening.

My

These opinions were expressed to Lynne Abbott and John

Stenabaugh:

- 68 -

AIt was within the -- around the second,

third week of my employment.

I'd

say at the beginning of my third

week of employment.

The factual base on which Mr. Imbert formed these opinions

included what he testified were:

(a) an inordinate number of support calls

(b)a trend towards the software becoming "buggier" not

"cleaner"

(c)a lack of "revision control on the source code"

(d)a slow response time particularly with the month end

procedure

(e)the use of a DOS extender (Golden Bow), a type of product

with which he had no experience, and

(f)a concern about problems with the system's indices.

It is always important to place significant events fully in

their context, and having set out Mr. Imbert's concerns about the

Prism product it is important to recognize that these "strong"

opinions were apparently expressed at the end of his second or at

the beginning of his third week at HMRI. As already indicated Ms.

King was away for the first week and returned only during his

second week on the job. As important as his expressed reasons for

forming his opinions are the things he did not do during the few

short days before he recommended to HMRI that he should put

together a "prototype in Q-Pro-4."

- 69 -

Mr. Imbert was, as previously stated, a programmer with

limited experience in BASIC who was most comfortable programming

in Q-PRO-4. He did not speak to or have any contact with Prism in

the persons of either Mr. Jeffery or Mr. Trueit.

He neither

participated in nor watched any testing of the Prism software.

His total exposure to the Prism software was working with it as a

"user" for a brief period. He in no way sought from its authors

understanding or assurance with respect to the software.

Within a day or two of making his recommendations Mr. Imbert

began work on the Q-PRO-4 version of Savoir-Faire. This began as

early as October 2, 1986 according to him.

The approval process Mr. Imbert described occurred in three

stages. He was originally told by Mr. Stenabaugh and Ms. Abbott,

neither of whom had direct responsibility for the Savoir-Faire

project, to "go-ahead and see what you can do".

At some point

prior to the meeting of the Savoir-Faire Committee on October 23,

1986 he demonstrated his program to Stenabaugh and Abbott and at

that point they gave him what he described as the "unofficial go-

ahead."

The final stage of the process he described was the

approval of the Savoir-Faire Committee on October 23, 1986. The

minutes record the following:

S. Imbert reported that the program is written

in Quick Basic.

He stated that there are

other database manager languages which would

be more suitable and make the size of the

program much more manageable.

- 70 -

Review of the HMRI - PRISM contract must be

done in order to avoid legal complications and

implications if HMRI rewrites the program to

make more supportable. Also, the question of

"royalties" must be answered in the event of

rewriting programs. Mr. Colman will look into

this matter.

S. Imbert feels that he can rewrite the whole

program in approximately 2 months (i.e. a

single terminal Ontario basic system by the

end of December).

It was recommended that a formal work plan be

done regarding the rebuilding of the software

taking into account the overall architecture

of the product.

Product extension will become easier if the

software is written in a database manager

language.

It has been decided to write the programs in-

house for Alberta, New Brunswick and ICD-9 CM.

Mr. Colman will be reviewing the contract

with PRISM and talk to PRISM about the turning

over of software changes to HMRI staff.

What the minutes record is a decision to approve the

"rewriting" of the software, or later, the "rebuilding" of it.

Perhaps the only things more amazing than the speed with which

Imbert reached his "decision" that the Prism software should be

replaced are the limited basis of knowledge on which he reached

that conclusion, the ease with which he received approval and the

people who gave it, John Stenabaugh and Lynne Abbott, neither of

whom were responsible for the project.

It is also important to keep in mind the events that

- 71 -

transpired at the Executive Committee meeting on March 1, 1987 and

in the following meeting of the Board of Directors and the state

of knowledge of those individuals of what had gone on since Mr.

Imbert was hired.

The evidence leads irresistibly to the conclusion and I

specifically find that Mr. Imbert joined an organization which at

that time had major problems in its management and organization.

He was hired to undertake the maintenance and support of a Basic

program when his major experience over the previous two years had

been in the Q-PRO-4 language. Mr. Imbert was a person of forceful

personality and aggressive by nature.

He was working directly

with Wendy King, the Savoir-Faire project co-ordinator who had

neither the technical expertise nor the force of personality of

Mr. Imbert.

The organization of the project at HMRI lacked

leadership, proper planning, proper management controls and proper

communication lines.

Mr. Imbert began to impose upon HMRI the direction and

planning it lacked. The force of his personality and the strength

of his convictions met not just little opposition it appears to

have met no opposition at all.

In essence Mr. Imbert began

The project was

recasting his job into what he wanted to do.

recast into one in which he began to "rewrite" the software in the

language with which he was most familiar.

- 72 -

Not only was he turned loose to do what he essentially wanted

to do he was doing so in a language in which only he, of all the

people at HMRI, was familiar, and in a language not generally

supported.

It is in this context that the subsequent events must be

viewed.

Mr. Imbert is a witness whose credibility is, in my

respectful view, highly suspect.

proceedings, but he is also a person who has other interests in

the outcome of this litigation. The defendant HMRI paid Mr.

He is a defendant in these

Imbert $800 per day for his time in preparation for and giving

evidence. In addition he has since left HMRI, and together with

Ms. King who has also subsequently left HMRI is engaged in the

development and sale of computer software in the health care

field.

Mr.

Imbert

in

his

demeanour

and

in

the

numerous

contradictions within his sworn testimony gave ample reason to

find that he is a witness who was prepared to shape, alter and

shade his sworn evidence to suit his own purposes.

B.

WENDY KING

- 73 -

I have already dealt with Wendy King during the course

of these reasons.

She and Stephen Imbert became the principal

architects of the Q-PRO-4 version of Savoir-Faire.

The evidence

supports the inference that with the arrival of Stephen Imbert she

abandoned her role as co-ordinator of the Prism Savoir-Faire

project

and

indeed

Prism

Savoir-Faire

itself

and

became

effectively Imbert's assistant as he pursued the Q-PRO-4 rewrite.

When he left HMRI Mr. Imbert reactivated a proprietorship by

the name of Cactus Computers which he had originally set up in

September 1984, that proprietorship was incorporated as Cactus

Health Systems Limited after he left HMRI in January of 1989. In

October 1989, five months after she left HMRI Wendy King joined

Cactus as its second full time employee.

During the preceding

five months she had worked at least part time for Cactus. At the

date of this trial Cactus had only the two full time employees,

Stephen Imbert and Wendy King.

Not surprisingly the products offered by Cactus are in the

health care field and directly related to the manipulation and

utilization of the data collected by the Savoir-Faire system. As

Wendy King put it in her evidence:

We offer a medical information retrieval

system, which is a data base manager for

hospitals.

It allows a hospital to, using

their input that is produced, that a hospital

would send to HMRI, according to the tape

specifications input, allows them to create a

data base -- when you input it, it creates a

- 74 -

data base.

In other portions of these reasons I will deal further with

the evidence of Ms. King, in particular her version of events

surrounding the arrival at HMRI of the computer system intended

for the University Hospital in Saskatoon.

At this point it is sufficient to say that Ms. King's

interests in the outcome of this proceeding are very much tied to

those of Mr. Imbert.

She was also paid by HMRI to attend and

testify, in her case $500 per day. It is primarily her activities

and those of Mr. Imbert which are at the center of controversy in

this action.

In view of their present employment and the

marketing area of Cactus coupled with the apparent interest in

this litigation they have much to lose in the event of adverse

findings in this litigation.

For these reasons coupled with the demeanour of each on the

witness stand and the numerous contradictions within their

respective testimony I am wholly unable to accept either as a

disinterested witness attempting to tell the truth.

In fact I

conclude that each attempted as best they could to obscure the

truth and offer explanations which shaded or altered events and

advanced their particular interests. Their evidence is in my view

unreliable.

- 75 -

C.

THE "DECISION" TO REWRITE

The defendants seek to justify the decision to pursue

the Q-PRO-4 rewrite in a number of ways but central to their

position is the thesis that as of October 1, 1986 Prism Savoir-

Faire was not a stable, maintainable product and that the Version

86.6 release 2 delivered to HMRI on September 28, 1986 was

"deficient and inoperative" in many respects.

HMRI submits that

when Imbert was initially authorized to write Savoir-Faire in Q-

PRO-4 it was only undertaken as a contingency plan, as a backup so

to speak, in case Prism was unable to solve the problems with the

Prism product.

A review of the surrounding circumstances is complicated by

the management difficulties within HMRI and the breakdown in

communications which has already been touched upon. It is obvious

from the evidence that the Board of Directors and the Executive

Committee as groups, as opposed to certain individuals present at

the meetings of these committees were unaware of the decision to

rewrite the software or of the scope of the difficulties which are

said to justify that decision until months after the project had

been first begun.

As a result some of the apparent conduct and

actions of those committees must be viewed in that light.

On September 22, 1986 there was a meeting between Prism and

- 76 -

senior people from HMRI who were in Vancouver to attend a Board

Meeting on the previous day.

In addition to the fifteen

individual members present for the Board Meeting, Mr. Colman, Ms.

Bickle and Deanna Hickey were present for the Board Meeting.

The timing of these meetings is of some significance.

Mr.

Imbert had been hired and had assumed his duties, Prism had

delivered a release of the final version of its software, and the

date for the end of Prism's obligation under the contract to

maintain the software, October 1, 1986, was slightly over a week

away. HMRI had just concluded an agreement with CPHA to take over

the hospital sites in Canada which sent data to that organization,

a process which would require additional versions of software to

be produced.

Mr. Imbert had either been given the initial go ahead to

begin rewriting the software in Q-PRO-4, or was about to be given

it. On HMRI's version of events this decision was reached because

of the ongoing serious difficulties with the Prism program and as

a contingency in case Prism was unable to correct the problems.

Both Mr. Jeffery and Mr. Trueit testified at some length as

to what occurred at these meetings but the defence called no

evidence from any witness present at these meetings.

I am left

with the plaintiff's evidence and that contained within the

- 77 -

documents.

There is within the evidence before me not the faintest

suggestion that HMRI took Messrs Jeffery and Trueit to task over

problems in the software they had delivered to HMRI. There is no

indication that there are major problems to be corrected let alone

any suggestion that unless they were corrected other solutions

would have to be found.

Colman, Ms. Bickle, Mr. Adams (an advisor to the board) and a

number of other Board members. Aside from the issue of "multi-

user locking" which Prism was investigating there were no

expressions of concern. The discussions were centered on the

future course of dealings between Prism and HMRI and the

atmosphere was cordial. Among the topics discussed were the

At the meeting, from HMRI, were Mr.

takeover of the CPHA provinces and the preparation of the software

necessary to collect the ICD-9 CM codes necessary for that

project. During the course of the meeting Mr. Colman, according

to the evidence of Mr. Trueit, gave tentative approval for Prism

to prepare versions of the software for Alberta and New Brunswick.

Colman indicated that he would get approval within HMRI and

confirm it.

A review of the minutes of the various meetings held during

the trip to Vancouver create a very different impression of the

state of the Prism software to that suggested now by HMRI. There

- 78 -

is a reference in the minutes of the Director's meeting on

September 21, 1986 to a "hold" on installations of the software

and the fact that this adversely affected revenues. The board was

advised, however, that the software was being ". . . well received

by recipient hospitals."

At its meeting on the same day the Executive Committee was

advised that there had been a suspension of installation over the

summer caused by "minor program problems" which had been

discovered at various installation sites. The minutes also record

that ". . . seventeen machines were now installed and the

remaining thirteen installations were scheduled to take place

before the end of November." There is found within the minutes no

expression of concern about the software.

On September 28, 1986 Prism forwarded two letters arising

from the meeting in Vancouver.

One was directed to Mr. A. J.

These letters outline the

Adams and the other to Bob Colman.

nature of the discussions from Prism's perspective but also deal

with a series of specific issues. The letter to Mr. Colman reads

in part:

We are including with this letter, a tape

containing Savoir-Faire Version 86-6.

This

tape includes the source code for the Licensed

Programs, which now includes several utilities

and sub-routines. These Licensed Programs are

detailed in revised "Schedule B" which is also

enclosed.

Will you please sign, date and

return one copy of the Schedule for our

records. This package also includes a listing

- 79 -

for "EISAM" the sub-routine which controls all

of the ISAM file handling within Savoir-Faire.

The tape of Version 86-6 includes all of the

changes requested by Wendy and we now, barring

some unforeseen bug, consider that this

portion of our contract is complete.

You have made it clear, via the above noted

discussions, that Prism will not be involved

in the software maintenance of Savoir-Faire

after September 31(sic), 1986, and that our

sevices(sic) will not be required to provide

that support. This is disappointing since we

felt

that

the

cost

of

our

support,

approximately $1,500 to $1,800 per month would

have been more cost effective for you, but

this option in accord with our Agreement.

For the record, we would like to remind you -

and we would ask that you advise others within

HMRI - that the License Agreement restricts

the manner in which HMRI may make the code

accessible

to

non-HMRI

personnel.

The

Agreement also provides that modifications are

the property of Prism and we would ask that a

copy of all source (including source in

progress) and the object of Savoir-Faire be

sent to us at least once per month beginning

November 1, 1986. Details for this can be can

be(sic) worked out over the next month.

It is EXTREMELY important that if Prism is to

provide changes to Savoir-Faire, and if HMRI

personnel are also modifying the software,

that some co-ordination and change control

procedure be established, implemented and

maintained.

requirement

The importance of this control

cannot be over-emphasized.

Perhaps this can be established while John

Jeffery is in Toronto during the week of

October 13.

With reference to ICD9-CM's.

Our Agreement

with you provides that Prism will supply each

of the remaining Province/Territory versions

of Savoir-Faire to HMRI on order and at costs

as are detailed in the Agreement.

In our

meeting of September 22, Prism offered HMRI

the CM's at no additional charge, provided

that we receive your order for both New

- 80 -

Brunswick and Alberta at the same time. This

offer is good until December 1, 1986.

We

understand that we have tentative approval for

this order but would appreciate something in

writing, along with the specifications and

requirements involved.

At the time this letter was written to Mr. Colman Prism

remained completely unaware of any major concern about its

software for the simple reason that HMRI had not expressed any

such concern.

It is implicit in the letter that Prism has been led to

understand that Mr. Imbert has been hired to maintain the Prism

Savoir-Faire package after September 30, 1986 and that they are

both reminding HMRI of their contract obligations with respect to

the source and object code and seeking to formalize the exchange

of code after HMRI undertakes maintenance.

It is equally implicit that they have no knowledge that HMRI

has embarked on a "contingency plan" to rewrite the program in Q-

PRO-4 or any other language.

Whether version 86.6 release 2 was deficient in any respect

or not, the parties agreed that it was shipped to HMRI with this

letter to Bob Colman on Sunday September 28, 1986. It could not

have been tested or have formed the basis for any opinion Mr.

Imbert expressed concerning the software on September 29 or 30 at

- 81 -

the start of his third week with HMRI.

After some testing by HMRI 86.6 release 3 was forwarded by

Mr. Jeffery on October 9, 1986.

This release included what Mr.

Jeffery described was a "considerably diminished" number of change

requests but did include a sub-routine which had become corrupt at

Prism's facilities in Vancouver.

The problem with this routine

prevented the operation of the software in multi-user mode.

It is clear that Prism was made aware of this difficulty and

that it was the subject of a number of telephone calls between

John Jeffery who was at HMRI's offices and Wendy King who was

trying to install the software at a hospital in Saskatchewan.

VIII THE TRAINING SESSION - OCTOBER 1986

In mid-October 1986 Mr. Jeffery travelled to HMRI's offices

in Ontario for the purpose of conducting a training session on the

internal structure of Savoir-Faire. The purpose of the course was

to assist HMRI in assuming the responsibility of maintaining the

software.

Mr. Jeffery had prepared a course outline and the intention

was to spend four days going through the material and training

HMRI's new programmer in the internal structure of the program.

Mr. Jeffery had sent in advance the compilers, the manuals and the

- 82 -

course outline. In attendance at the course were Mr. Stenabaugh,

Mr. Imbert and Ms. Abbott.

Mr. Jeffery testified that the course did not go at all as he

had planned.

The attitude of those in attendance from HMRI was

hostile and before the course even began Mr. Stenabaugh took Mr.

Jeffery aside and told him that they would not be following his

course outline.

What then followed was an abbreviated course

interrupted by telephone calls from Wendy King who was in

Saskatchewan having difficulty with an installation.

During the course of what remained of this training session

Mr. Jeffery was directed by those in attendance to focus on

particular areas of the program in which they were interested. By

Thursday morning those in attendance had run out of questions and

Ms. Bickle approached Mr. Jeffery and advised him to leave as

"people were not happy" with him.

Mr. Jeffery described the

entire event as being extremely uncomfortable.

A measure of how unsettling the experience was for him and of

the level of his concern can be found in the fact that before he

left to return to Vancouver he returned to HMRI's offices and

placed within each of the screen images a copyright notice which

would not be visible on the screens as they displayed but would be

to a programmer accessing them.

Mr. Jeffery was also concerned

- 83 -

enough that he did not leave at HMRI the source code for the

general use subroutines.

On his return to Vancouver Mr. Jeffery discussed his concerns

with Mr. Trueit and it was his belief, although he was not

certain, that he sent the code for these routines with the

University Hospital computer when it was shipped to HMRI on

October 21, 1986.

What then was the purpose of this training course and HMRI's

participation in it? There is no evidence that HMRI ever actually

began a program of software support or maintenance of the Prism

Savoir-Faire package. Indeed the evidence of Mr. Imbert, who was,

after all, the person hired to undertake those responsibilities is

quite the opposite.

Both in his discovery evidence and in his

evidence at trial Mr. Imbert indicated that by the time Mr.

Jeffery arrived in Toronto to give this course he had no interest

whatsoever in the Prism product or the course because he had

already been instructed to rewrite the program in Q-PRO-4. It was

his opinion that all existing Prism sites would be replaced with

Q-PRO-4 as soon as it was ready.

Mr. Imbert went so far as to testify that he was annoyed at

having to attend the course and only did so because he was given

what he described as "marching orders" to do so.

In addition

- 84 -

Imbert testified that no one to his knowledge told Mr. Jeffery

that they had begun to rewrite the software in Q-PRO-4.

He

certainly didn't because just prior to the course starting Mr.

Stenabaugh took him aside and specifically instructed him to say

nothing about the rewrite.

Mr. Stenabaugh in his evidence agreed with Imbert that Prism

and Jeffery were not told about the rewrite but went on to suggest

a reason for not telling them:

AI think there was a conscious effort

initially not to tell them.

QAnd I suggest that's because you knew

that it was wrong?

ANo. I suggest to you that the reason in

my mind was that at that time were

still wanting to get a complete and

working version of, whatever we

were up to, 86.6, for the provinces

for which it was designed, and the

intent was to use it there; and we

were concerned that -- at least I

was concerned -- I can't speak for

the others -- I was concerned that

if we said we were writing in

another

language,

program

--

another

lead to

it might

difficulties and we may not get the

completed version of 86.6 with the

bugs removed, which is what we

really wanted to get, from my

perspective.

This evidence might have some force if there was anything to

suggest that HMRI was actually trying to accomplish this goal.

Even then it is directly contrary to Imbert's evidence and opinion

- 85 -

that from the time he was given the go-ahead the Prism product was

intended to be replaced.

Mr. Stenabaugh's evidence can best be put in perspective by

considering it in the context of the arguments advanced on behalf

of the defendants at this trial. That position is that the Prism

product was defective and unstable and incapable of being

maintained at that point in time and, hence, as a contingency plan

in case these problems weren't fixed HMRI elected to pursue

secretly the rewrite. The difficulty with this position is that

it would appear logical that Prism should be put on notice of the

extent of the problems and concerns and told to fix them or else.

This was never done. It hardly appears to be, if accurate, a set

of circumstances in which HMRI would logically exercise its option

under the contract to take over the maintenance of the software,

in its unstable form, while quietly going about the rewrite.

The out of pocket expenses for this project during the first

year exceeded $157,000 according to Mr. Colman's report to the

Board of Directors.

If the results of those expenditures had

yielded the delivery of a product so defective that it did not

work one would expect HMRI to notify Prism in no uncertain terms,

not to set about quietly and secretly "rewriting" the software at

their own expense.

- 86 -

This version of events is so out of keeping with logic or the

harmony of the surrounding circumstances as to be not just

improbable, but, in my respectful view, wildly improbable.

I reject the explanation offered by Mr. Stenabaugh.

In my

view, the course was staged and arranged for the express purpose

of assisting Mr. Imbert and to advance his understanding of the

program and the writing of the Q-PRO-4 version.

Mr. Jeffery's

instincts and concerns were not only justified but they

unconsciously identified the actual events which were going on

behind the scenes.

HMRI through the employees present at this course set out to

deceive Prism and Jeffery for the specific purpose of advancing

the Q-PRO-4 project.

The subsequent focus of the defendants submissions on Mr.

Jeffery's telephone discussions with Ms. King and his refusal to

leave behind the source code for the general use subroutines are

attempts to divert attention from the real purpose of the training

course. In terms of the general use subroutines, the short answer

to that criticism is twofold; firstly HMRI did not intend to

actually maintain the software and secondly, these general use

subroutines were not listed nor included within the list of

Licensed Programs contained within Schedule B of the license

- 87 -

agreement and proposal. As a result HMRI's contractual rights did

not extend to them and Prism was not obligated, in any event to

provide that code.

The fact remains, however, that there is no evidence that

HMRI changed its intended course of action as a result of Prism

withholding that code, particularly in light of Mr. Imbert's

complete lack of interest in the program.

IX

THE INSTALLATIONS IN SASKATCHEWAN IN OCTOBER 1986

Much attention was focused during this trial on the failure

of the attempt by Wendy King to install a multi-user system at St.

Paul's Hospital in Saskatoon in October 1986. It was the

difficulties with this installation attempt that led to the

telephone calls from Ms. King to Mr. Jeffery during the training

course in Toronto.

The trip was intended to complete three

installations. Two of these, Meadow Lake and Battleford Union,

were single user installations which were completed without

difficulty and ran continuously until they were replaced with the

Q-PRO-4 version in 1988.

What is not explained on the evidence in any satisfactory

fashion is how Ms. King, the person responsible for testing and,

apparently, deciding when a version was ready for release could

have with her in Saskatchewan to install a multi-user site a

- 88 -

version of the software which could not be run in multi-user mode

because of a corrupt subroutine.

The point here is not how the

routine had become corrupt but how Ms. King came to take a version

of the software which had obviously not been tested by her in

multi-user mode to Saskatchewan in the first place. No reasonable

explanation is offered.

X

VERSION 86.6 RELEASE 4

Prism had received an order for a Multi-user system from the

University Hospital in Saskatoon in the fall of 1986.

When Mr.

Jeffery returned from the training course in Toronto he

immediately prepared a new release of Version 86.6 correcting the

problem with the corrupt sub-routine.

This new "release" which

may have included some other modifications was shipped to HMRI

after having been installed on the hard disk of the computer

intended for the University Hospital.

With the exception of a single request for a change to one

program during the month of November, Prism received no change

requests of the sort usually arising from testing until the end of

December 1986 nearly two months later. The defendants submit that

this time period is not unreasonable ". . . particularly

considering the problems with the previous release of Version

86.6." The difficulty with this submission is that it offers no

explanation for the delay, particularly in light of Ms. King's

- 89 -

evidence that it was her practice to carry out the tests of a new

version in a narrow time frame consisting of one or two days.

Version 86.6 release 4 arrived at HMRI on October 23, 1986

yet the Liaison Committee minutes of December 3, 1986 note that

"Saskatchewan, B.C., Newfoundland, P.E.I. programs have to be

tested."

86.6.

Ms. King agreed that this entry referred to Version

On the same day that release 4 arrived at HMRI a meeting of

the Savoir-Faire committee attended by virtually all of the senior

staff of HMRI gave Mr. Imbert the formal approval to rewrite

Savoir-Faire.

It goes without saying that there had been no

testing of release 4 at the time of this approval.

The minutes contain a very specific entry which Mr. Imbert

disputed. The entry reads:

S. Imbert feels that he can rewrite the whole

program in approximately 2 months (i.e. a

single terminal Ontario basic system by the

end of December)

Mr. Imbert testified that he complained about this entry when

he saw the minutes. The entry he insisted was inaccurate. He was

not developing an Ontario system at all but one for Alberta.

When directed to an entry in subsequent minutes of a meeting

- 90 -

which he attended in which previous minutes were approved he gave

this evidence:

AYes.

Your Honour -- My Lord, the --

this puts -- the record-keeping or

the minute-taking practices of HMRI

were casual, to say the least,

okay. Now, I did not -- I did not

raise it at the meeting because I

didn't feel it was necessary.

were small organization.

We

I

a

didn't feel that I was going to be

held

accountable

for

what

was

written down in the minutes in the

previous meeting.

THE COURT: Go ahead, Mr. Chaster.

MR. CHASTER: Thank you, My Lord.

QSo you didn't subsequently ask that

anybody correct the minutes?

ANo.

I had very little experience, by

the way, in sitting in committees.

This was a new process for me as

well.

It is perhaps time to put the proposition bluntly, I do not

believe Mr. Imbert in this particular portion of his evidence.

I

do not accept the proposition that, even given the management

difficulties of HMRI, that virtually the entire senior staff of

HMRI together with the committee members present could so

misinterpret the actual plan approved so as to record it as

yielding a single terminal Ontario system within two months when

what they actually discussed was the production of a prototype for

the Alberta system.

Once again it is necessary to leave the minutiae and look at

- 91 -

the broader picture. Mr. Imbert has already conceded that when he

received the initial go-ahead he saw all installations being

replaced with his Q-PRO-4 version. I find that the entry in the

minutes accurately reflects the discussion and that Mr. Imbert's

denial was simply calculated to serve his perceived interests at

the expense of the truth.

With the formal approval of the Q-PRO-4 project, Stephen

Imbert and Wendy King became even more heavily involved in the

rewrite of the program.

It was this project and Ms. King's

involvement in it which resulted in the delay in testing release

4. Her time commitment to the project and the way in which it was

approached is evident from a portion of her evidence in which she

described using the Prism program to explain to Mr. Imbert the

functions:

QDid you define -- in preparation of

these

screens

for

the

Q-PRO-4

version, was it you or Mr. Imbert

who defined the data that was to go

in?

AIt was me.

I went through the fields

field by field, screen by screen,

and instructed Mr. Imbert which

were

appropriate

entries,

which

were inappropriate entries, what

error messages should be applied to

them,

what

cross-field

edit

messages should be applied to the

fields.

On the whole of the evidence I find that Ms. King and HMRI's

- 92 -

priorities had changed and that the priority was to complete the

Q-PRO-4 rewrite not to test or even obtain a final version of

Prism Savoir-Faire.

With the formal approval of the rewrite HMRI began to take a

different approach to its relations with Prism.

Mr. Colman was

Savoir-Faire

directed following resolutions passed by the

Committee to write letters to Prism concerning the use of Golden

Bow (a DOS extender) and the failed installation at St. Paul's

Hospital in Saskatoon.

The letters which are dated the same date as the Committee

Meeting october 23, 1986 mark a distinct change in tone.

The

first of these letters issued instructions that Golden Bow was not

to be used on any hardware supplied to Savoir-Faire Customers and

concludes:

I recognize that I do not have a watertight

case against "Golden Bow", but I recognize

that those sites to which it has been supplied

have problems that have not appeared in other

installations.

Please do not waste your time or mine in the

defence of this software package. It may well

do what you say it can do, but I just do not

wish it complicating our installation schedule

and confounding our users.

The technical basis upon which this direction was issued can

be found within the minutes of the meeting of October 23, 1986 the

- 93 -

entry reads:

J.

Stenabaugh

talked

to

R.

Belsey

as

recommended at the last meeting about Golden

Bow.

Mr. Belsey had not heard of this

product, but felt it was not necessary and

should not be done.

On the basis of an opinion obtained from an individual who

was not only unfamiliar with Golden Bow but had never heard of it.

HMRI issued its direction and closed off discussion, this was

indeed a new approach from HMRI.

Of the three problem sites

referred to in Mr. Colman's letter one of them, Women's College

Hospital, had never been brought to Prism's attention prior to

this correspondence.

Mr. Colman's second letter recounts a

version of the events at St. Paul's and is similar in tone.

On November 9, 1986 Prism responded to the letters from Mr.

Colman giving their version of the events.

wrote:

In part Mr. Jeffery

On the night of October 12th while testing

changes on a future version of SuperMACS.

I

discovered the above noted problem in our

development system. HAVING HEARD NOTHING FROM

HMRI to indicate to us that you might also

have the problem, we naturally assumed that it

was local TO OUR DEVELOPMENT SYSTEM and

affected only our current work. Consequently

HMRI were not advised.

ANY TEST by HMRI of Abstracting in multiuser

mode would have immediately produced the error

we

are

referring

to.

There

are

no

circumstances whereby this fault could have

gone undetected had any of your staff tested

the system fully, BEFORE IT WAS AUTHORIZED BY

HMRI FOR RELEASE.

- 94 -

When the problem was reported to me from St.

Paul's, I recognized it immediately and told

your staff member what had caused it. ONLY AT

THIS POINT did I realize that the problem

existed in earlier versions of the software

which had obviously NOT BEEN FULLY TESTED by

HMRI. At this time I also told your staff in

Toronto of the problem and advised them of the

circumstances.

Prism delivered a corrected version of 86-6,

with the University Hospital machine shipment,

the following Wednesday.

This shipment also

included the demonstration version of the

Report-Writer in time for the OHA Convention.

A phone call from you to clarify this matter

would have been appreciated by us and would

have been more in keeping with the cordial

relations which we have previously established

between ourselves. This call could also have

removed any doubts on your part, or at the

very

least

provided

you

with

correct

information which could have provided you the

opportunity of further research to obtain the

facts.

There is no apparent response from Mr. Colman to this

correspondence.

In the meantime the "corrected" version of 86-6

was already in the possession of HMRI, but apparently untested.

XI

THE UNIVERSITY HOSPITAL COMPUTER

On September 18, 1986 the University Hospital in Saskatoon

placed an order for a three user Savoir-Faire system.

The

hardware from the installation was provided by Prism and shipped

by prior arrangement to HMRI who was to carry out the

installation.

This installation was never carried out and the

- 95 -

computer system was eventually returned to Prism in late January

1987 after the order was cancelled.

Prior to shipping this hardware to HMRI on October 21, 1986

Mr. Jeffery installed on the computer's hard disk the source code,

subroutine libraries, compilers, the complete corrected Savoir-

Faire system as well as Prism's Report Writer for testing by HMRI.

Mr. Trueit testified that HMRI at no time gave him an

explanation for why the hardware was not delivered.

On January

22, 1987 Mr. Trueit wrote to Mr. Colman, in the last paragraph of

his letter he enquires about their outstanding invoice for $13,200

for the University Hospital Hardware which had been sent to HMRI

at their request.

Shortly after this letter was forwarded HMRI

returned the hardware to Prism.

Mr. Jeffery and Mr. Trueit both gave extensive evidence

concerning this order.

In many respects this evidence was

Despite this the defence led

untouched in cross-examination.

evidence as to the events which had not been put to the

plaintiff's witnesses while they were on the witness stand. One

example of this was Wendy King's evidence that she had advised Mr.

Jeffery by telephone that University Hospital was wavering on

their order. No such suggestion was put to Mr. Jeffery during his

cross-examination and as a result he had no opportunity to respond

- 96 -

to such a suggestion or place any evidence he had with respect to

such a suggestion before the court. Fundamental fairness dictates

that in such circumstances the subsequent evidence can be afforded

little weight, any other approach would afford major advantages to

those who practice "trial by ambush."

In their final argument the defendants respond to the

plaintiff's suggestion that the University Hospital machine was

used in the development of the Q-PRO-4 system by taking the

position that this issue was not raised either in the pleadings or

in the Statement of Facts and Issues filed by the parties. They

go on to suggest that:

Such an allegation, if proved, might give rise

to damages for rental of the computer for the

period when HMRI had it. However, such a head

of damages is not pleaded. It is not relevant

to the issues as defined in this action.

With the greatest of respect to the argument presented, the

factual issue does arise on the evidence and it arose both from

evidence led by the plaintiffs and responded to by the defence,

without objection.

The plaintiffs in this action, among their

This

other claims, seek both punitive and exemplary damages.

evidence is both relevant to the general factual issues and to

those claims.

Mr. Trueit testified that as of the date he sent his letter

- 97 -

on January 22, 1987 he had not received any notice of cancellation

of this order, the notification and return of the equipment

occurred after his letter was sent to HMRI. Mr. Jeffery testified

that he spoke to Wendy King on a good many occasions about the

University Hospital equipment.

According to him he had good

reason to do so. The corrected version of 86.6 had been shipped

on the hard disk of that computer system and Mr. Jeffery was

anxious to ensure that the software was functioning correctly. It

was Mr. Jeffery's evidence that he spoke to Wendy King regularly

and her responses were essentially that everything was going fine

but that she hadn't completed the testing.

When it was shipped by Prism the software installed on this

computer included and utilized the Golden Bow file extender, and

the hard disk was formatted as a single "C" drive. When it was

returned three months later it was in a very different condition.

Mr. Jeffery described it in this way:

AThe hard disk had been reformatted to a

Drive C and a Drive D, Golden Bow

was no longer on it, and the Drive

C had been wiped to nulls, the

whole of the Drive C had been set

to nulls.

QMeaning . . .

AThe -- it was apparent to me that the

Norton Utilities WIPEDISK had been

used to wipe the drive.

QWhat about Drive D?

AD was all right,

I

could create

directories in D, and I could move

- 98 -

data to Drive D.

QWas there anything on Drive D?

ANo.

I earlier understood that there

might have been something on D, but

in retrospect, thinking about it,

there wasn't anything on Drive D.

It was not used.

Not only was Mr. Jeffery not cross-examined on this evidence.

Ms. King did not dispute the details but, in her evidence, sought

to explain them.

Ms. King testified that when the equipment arrived she made a

backup of the software and installed it on her own machine for

testing. She then tried to get the University Hospital computer

running in multi-user mode. In this she was unsuccessful and she

was only able to get it running in single user mode. In chief she

described the events in these words:

QDid you do anything with respect to the

University Hospital machine other

than that which you have described

to the present?

AYes, I did.

QWhat did you do?

AI had made the backup of the program.

When I couldn't get it to run as a

multi, formatted the hard disk and

reinstalled the software as I had

backed it up to try to get it to

run.

I can't recall whether I

checked the slave boards before or

after that step.

At some point shortly thereafter,

- 99 -

when I couldn't get it to run as a

multi, I formatted the hard disk

again.

After I had checked with

the slave boards and checked that

they

settings

were --

were

the

dip

switch

that

appropriate,

they were seated properly and all

of the rest of the boards were

seated properly, I attempted to

install the software one more time,

and to the best of my knowledge, we

left it there for a day or two as

at that time. I recall being told,

when scheduling was to be arranged

for University Hospital, I was told

that

University

Hospital

was

wavering and was not sure as to

whether or not they would go ahead

with the system.

QSo what did you do as result of that?

AI didn't worry about setting it up as a

multi at that time. I had packed

it up in a box. I formatted the

hard disk again.

Ms. King in her evidence had difficulty from time to time,

but nowhere were the difficulties more apparent than in this area

of her evidence.

examination in this area her uncertainty, hesitation and lack of

technical expertise reached its zenith. Prior to packing the

She was shaky in chief but during her cross-

computer back into its boxes she described formatting the hard

disk with what she described as ". . . a regular DOS format to

create two drives."

In this way she explained Mr. Jeffery's

evidence of finding the hard drive formatted into two logical

drives.

- 100 -

When cross-examined on specific details of the events after

the receipt of the University Hospital equipment she was unable to

provide answers and frequently launched into a repetition of the

recital of events previously quoted.

paused for long periods of time before attempting an answer or

simply trailed off to nothing. Once again an example is

Even more frequently she

warranted, to place this answer in its proper context it must be

remembered that Ms. King was the co-ordinator for the Prism

Savoir-Faire project and responsible both for testing at HMRI's

end and for developing the testing requirements. She also had the

responsibility and the power to determine when a version of the

software would be released to the hospitals.

She had just

recently returned from the failed multi-user installation in

Saskatchewan and at the Savoir-Faire Committee meeting of October

23, 1986 had apparently reported that the final version of the

Prism software "does not work." The problem appears to have been

when you attempted to install the system to operate as a multi-

user system but when she was asked what the specific problems she

encountered with the University Hospital computer her response

was:

AI can't recall if I did that first --

like, if I brought it up -- I can't

recall if I did that first or not,

if I tried to bring it up as a

multi, and then I do know that I

formatted the hard disk, restored

from the backup that I had, tried

to

problems with it.

what the exact nature of the

problems were. I do recall taking

get

it

running,

had

some

I can't recall

- 101 -

the slave board out and checking

the dip switch settings on the

slave board.

QAnd checking the seating. You told us

about that in Chief.

What I'm

after is the specific symptoms of

the problem.

You can't recall; is

that your evidence?

AI'm afraid I have to say generally -- I

couldn't get it up. I can't --

As her cross examination progressed Ms. King became steadily

more uncomfortable, at times she seemed to forget the evidence she

had given shortly before. Although she testified earlier that she

had in fact partitioned the hard disk to create the "C" and "D"

drives she appeared later to forget that she had done so.

When asked to explain why the last thing she did before

packing up the computer was to format the "C" drive she testified:

QCan you explain to His Lordship why the

last thing you did with the machine

was to format the C drive?

AI think it was to, from what I recall,

just not go through the trouble of

reloading the software again when I

had heard that the hospital was not

interested or not -- was wavering

about Savoir-Faire. I . . .

QMs. King --

AI don't recall anything specific or

unusual about it.

THE COURT:

I'm sorry, Ms. King, I don't

understand that answer.

AI don't recall --

- 102 -

THE COURT: Just a moment, please. Let me try

and explain to you my problem with that

answer. You said you formatted it because you

-- as I understood Mr. Chaster's question to

you, it was your evidence so far, if I recall

it correctly, is that the last thing that you

did when you packed it away was format the

hard disk drive again?

A

Yes.

THE COURT: And what you just said to me was

you did that because you didn't want to go

through the trouble of reloading the software?

AI'm sorry, that's an assumption on my

part today.

I have been thinking

about it for the last couple of

days trying to recall University,

what

machine.

hard disk the first time.

happened

I recall formatting the

To the

with

University

best of my recollection, I recall

formatting it a second time.

making an assumption that

I'm

I

possibly did it because I couldn't

get the multi terminal up and just

deleted what I had put on there,

and it's an assumption that I'm

making now after having thought

about it. I'm certainly not trying

to mislead you in any way.

THE COURT:

I'm sorry, I'm just trying to

understand what your evidence is about that,

and I don't understand it.

AI recall with the University machine

doing the format the first time, I

recall restoring the backup to try

and get it going as a multi with

the Drive C and D, I recall having

problems.

can't honestly say why I did the

format. I can make an assumption

I recall a format.

I

now as to a possible reason why I

did it, but I'm having trouble

recalling exactly -- I do know that

I did pack it up when I heard that

University was wavering.

- 103 -

Ms. King testified that this computer was not used at all

after it was packed up within a week or two of its arrival and,

that only Prism software was loaded on the machine. She offered

essentially no explanation for their retaining the hardware for

three months before returning it to Prism.

Mr. Imbert did not testify concerning the University Hospital

computer during the course of his examination in chief but during

his cross-examination he did testify that:

I know that that hardware wasn't used for

any in-house purposes.

This evidence was given by Mr. Imbert despite the fact that

he had already testified that from the end of October, until

sometime in January of 1987 virtually all of his work on

developing the Q-PRO-4 version was done at his home on a computer

he had there, and that he only occasionally went into HMRI's

office.

Coincidentally this period corresponded almost exactly

with the period that HMRI had the University Hospital equipment in

its possession.

According to Mr. Imbert he at no time did any work on this

computer and did not touch it.

- 104 -

When the University Hospital computer was returned to Prism

the hard disk had been formatted (and partitioned) to form both a

"C" and a "D" logical drive on the single installed hard disk.

Mr. Jeffery found the entire "C" drive had been formatted and set

to "nulls" whereas the "D" drive although it contained no data had

not been reset to "nulls".

Mr. Jeffery concluded that a utility like Wipedisk had been

used to wipe all data from drive "C".

Gordon Van Toen was the expert witness tendered by the

defence with expertise in computer programming, the PC DOS and MS

DOS operating systems and microcomputers generally.

The evidence is that there are two types of formats which can

be done in DOS:

high level format leaves intact on the hard disk a boot record, a

file allocation table and an empty root directory. A low level

a high level format or a low level format.

A

format if performed would wipe the hard disk completely clean and,

in the process eliminate any partitioning of the hard disk into

logical drives. The condition of the hard disk when returned to

Prism cannot be explained by either process, nor by the evidence

given by Ms. King.

The wipedisk utility contained within the

Norton Utilities package, however, does have the ability to

overwrite all data contained on a specific logical drive resetting

- 105 -

the values to "nulls" and leaving the partitioning of the hard

disk into logical drives intact.

During his trip to Toronto to give the training course in

October of 1986 Mr. Jeffery was introduced to the Norton Utilities

software by Mr. Imbert.

It was a package which Mr. Jeffery had

never seen before but it was one with which Mr. Imbert was both

familiar and had available to him.

In a complex web of evidence occupying in excess of half a

day of evidence and over 30 pages of transcript Mr. Jeffery

analyzed a series of "electronic" documents produced by the

defendants during the course of this litigation.

The documents

analyzed were tape backups of the contents of hard disk drives of

computers used at the material times at HMRI's offices. The five

tape backups were identified in Exhibit T21-14 as follows:

Document

Description as Description as

Listed Labelled

D182

System

60MB tape "test":

TEST

Wendy's

Backup of hard (New Ver Run) 2l/9/86,

disk - 01/23/8730/10/86

D226

60MB tape "backup

University system

Backup

Sask.

University/

system/Sask.

86.6 ..."

86.6 not final 86.6

- report writer

on it

Hmri

(Ex46 Prism V.

W. King June

- 106 -

22, 1989

D227

(S.C.)

60MB tape "86.686.6 New index - JJ's

- New indexed -seen (University

J's seen-"

mahcine) executable

only

V.

-

installed -

Estevan

Hmri

Plains

(ex 47 Prism v.

W. King June 22

S.C.)

1989

D252

20MB tape "Wendy's

27/3/87" 03/27/87

Feb 11 Wendy's

complete

1

6

(WX71A

1989)

D253

WK

October

20MB tape without

No label called

(EX72A WK Prism

Hmri October 6

label

"Anything tape"

v.

1989)

The tape backup identified as D226 was labelled as a backup

of the University Hospital hard disk on November 3, 1986. The

files on that backup indicate that, prior to November 3, 1986 (the

date of the backup), Prism Savoir-Faire was last used on October

23, 1986 in multi-user mode and that although the computer itself

was used on October 31, 1986 Prism Savoir-Faire was not run on

that date.

This backup is significant in yet another way, in that it

shows that on the date of this backup the University Hospital

- 107 -

Computer was still configured with a single "C" drive and that

Golden Bow was still present on the system. This is contrary to

the evidence by Ms. King.

The defendants in their written submissions describe the

plaintiff's

analysis

of

these

tape

backups

variously

as

"confused", "absurd" and "stretched." I do not propose to trace

within these reasons the intricate analysis carried out by Mr.

Jeffery.

each of the tape backups and followed their movement from backup

to backup. Underlying this analysis is the process by which

In his analysis Mr. Jeffery has examined the files on

computers record within their file directories the date and time

at which a particular file was created or modified.

Certain files were readily identifiable by Mr. Jeffery as

having been provided to HMRI with the University Hospital

computer. In part he was able to do so because of his intimate

knowledge of many of these files and the fact that in most cases

he created them.

Certain of his conclusions are beyond dispute.

The tape

backup D182 was made on or about January 23, 1987. The contents

of the hard disk backed up on this tape contains in a directory

(SRC) the source code for Prism Savoir-Faire 86.6; in a separate

directory it has the executable program code for Prism Savoir-

- 108 -

Faire together with the necessary compilers and linkers. On the

same hard disk of the same computer is found the QNE4 directory.

This directory is a Q-PRO-4 development directory containing,

among other things, Q-PRO-4 source files. An examination of the

directory shows virtually no activity after January 15, 1987. Mr.

Jeffery concluded that on or about that date the system was backed

up and moved to another computer.

I accept Mr. Jeffery's evidence and analysis in this respect.

On the balance of probabilities the tape backup D182 is a backup

of the University Hospital computer performed on January 23, 1987

before the data on its hard disk was removed in preparation for

the return of the hardware to Prism.

A good many events were

occurring at this time in January of 1987, but I am wholly unable

to accept Wendy King's evidence as to the formatting of the hard

drive of the University Hospital computer before it's return.

This evidence though quite precise is demonstrably untrue. No DOS

format either high level or low level produced the result observed

by Mr. Jeffery. If D182 is a backup from the University Hospital

machine, and I find on the evidence that it is, then Ms. King and

Mr. Imbert did indeed set about the development of the Q-PRO-4

version of Savoir-Faire on Prism's own computer while they kept it

at HMRI's offices or, more probably, Mr. Imbert's home between

October 1986 and late January 1987. More significantly, in their

removal of data from the computer, the use of wipedisk or a

- 109 -

similar utility on drive "C" and in the evidence of the formatting

of the hard drive Ms. King and Mr. Imbert set out to conceal their

activities and eventually to deceive the court.

Wendy King's

recollections were too precise to be a mistaken product of faulty

recollection, the reasons for her demeanour on the witness stand

is readily explainable in this context.

The preponderance of the probabilities in this case supports

these conclusions, no other reason was advanced for HMRI's

retention of $13,000 worth of computer equipment for a period of

three months and no rational explanation was offered for the

deletion from Prism's computer of their own software (whether it

worked or not) prior to returning it to them, the explanation

offered by Ms. King was nothing short of fanciful and contrived.

XII THE DECISION TO REPLACE PRISM SAVOIR-FAIRE

The defence maintains in this action that the eventual

decision to replace the Prism software was a staged one in which

the initial preparation of the Q-PRO-4 version began only as a

contingency plan.

This position must be assessed in the light of other events

which were ongoing at HMRI at the same time.

One of these

developments was the agreement announced on September 15, 1986 and

completed the preceding week whereby HMRI agreed to take over the

- 110 -

Canadian Hospitals then submitting data to CPHA.

CPHA was an

organization based in Ann Arbour, Michigan which at the time

collected data from hospitals in Alberta and New Brunswick

together with a few sites in British Columbia and Nova Scotia.

Alberta, in particular, was viewed by HMRI as a major

potential market and the takeover was viewed as securing HMRI's

position by creating a national health care data base.

This agreement required two important changes to the

abstracting software, firstly new provincial versions were

required for Alberta and New Brunswick, and secondly, the software

would have to be modified to accept ICD-9 CM codes.

With the

pending takeover, which was to start in Fiscal 1987 HMRI's staff

which was already very busy took on additional work both in the

liaison department, whose members had to assume responsibility for

Alberta and New Brunswick, and on the system side where the

mainframe programs had to be changed to accept abstracts in the

ICD-9 CM coding.

To this workload was added the fact that under the existing

contract with HMRI Prism was to provide software maintenance until

October 1, 1986 after which HMRI had the option of assuming

responsibility itself or contracting separately with Prism.

- 111 -

Throughout the summer and early fall of 1986 Prism wrote to

HMRI concerning the need for a decision by HMRI.

Mr. Trueit's

letters of June 17 and July 17 drew no apparent response and in a

letter dated August 10, 1986 he writes:

On other matters - having not had a

response from you with regard to the questions

raised in our letters of June 17 and July 17

(I assume you are waiting for the meeting in

Vancouver). . .

It is not until sometime in mid-September of 1986 that Prism

was

responsibilities.

appears to correspond with the timing of the decision to hire Mr.

Imbert. There is little doubt that Prism is aware of this

told

of

HMRI's

decision

to

assume

maintenance

This date is not entirely certain but it

decision by September 28, 1986 at which time Mr. Trueit confirms

by correspondence his understanding of that decision.

There is no question, on the evidence, that Prism and HMRI

were in frequent contact by telephone. The partial records placed

in evidence indicate no less than 278 such calls during the period

from January to July 1986.

What is more puzzling, given the

positions taken by HMRI in this litigation, is the one sided

nature of the written correspondence. Between January 4, 1986 and

late July 1986 twelve letters directed by Prism to individuals at

HMRI went unanswered.

- 112 -

The defence in its written submissions suggests that "Mr.

Trueit's long rambling letters, which are not in answer to any

request from HMRI. . . are suspect as being possibly merely a way

for Mr. Trueit to put his own shade of meaning upon events."

The difficulty with this submission is that it fails to

address the fact that these letters were written prior to the

breakdown of the relationship between the parties and reflect a

view of events, at the time of those events, which was not

challenged by HMRI, at least by way of contemporaneous written

responses.

HMRI in the defence of this action takes the position that

Prism's software was unstable and defective to the point that it

could not be maintained in the field and eventually had to be

replaced.

Yet not once in the critical period did any senior

member of HMRI, or indeed anyone from HMRI, put pen to paper to

record those concerns or demand that the problems be fixed.

Despite the evidence led at trial that the software was

unstable and could not be maintained unless it was fixed, HMRI

gave notice to Prism that it would undertake to maintain the

software after October 1, 1986. Despite the fact that HMRI hired

Mr. Imbert to maintain the Prism software and advised Prism that

they had done so, he never took a single step to actually maintain

- 113 -

the software but began instead to rewrite it.

On the evidence before me it was HMRI not Prism who were

making critical decisions based on versions of events which they

did not in any way reduce to writing.

In order to pursue the

position they take in this action HMRI says initially that the

correspondence Prism sent to them (which remained unanswered) is

"suspect" and "shaded" the real situation, while maintaining that

HMRI did not "request" the letters and was not "required" to

answer them.

They then go on to assert that what is actually

recorded in the records of HMRI should not be accepted either

because HMRI was concerned about letting the hospitals know about

the problems they were actually encountering.

I am somewhat concerned at the prospect of accepting the

suggestion that Prism's correspondence is suspect when it comes

from an organization which put virtually none of its present

concerns in writing, embarked on a secret rewrite of the program

and consciously planned to withhold information from its

contractual partner and instructed its employees to do so.

The position advanced by the defendants is little short of

preposterous.

It is not just out of harmony with the

preponderance of probabilities it is completely out of accord with

any form of reasonable and prudent business practice, particularly

- 114 -

when one reviews the correspondence which went unanswered and the

assertions contained within it.

One example is found within Mr. Trueit's correspondence of

July 17, 1986 to Mr. Colman:

We feel that we can supply a superior product,

one that will work now and in the future, with

the current software and with the future

enhancements.

Our hardware has served very

well, with no reported problems - although we

have seen several "quirks" in the IBM hardware

and in other systems. The software has also

stood

up

to

the

test,

large

clients,

Scarborough for example at zero errors and the

smaller users are doing equally well. If we

can work and plan together we can ensure that

these users remain satisfied and that we can

provide them with useful new products without

the possibility of a duplication of effort.

These assertions followed the more detailed report contained

within his letter of June 17, 1986, and went completely

unchallenged.

Despite the nature of the defence advanced in this action the

tone emerging from these letters from Prism's side is of a project

being successfully brought to

a

conclusion.

The letters

frequently note individual difficulties and what has been done or

is being done to correct them, what is notably absent is any

indication from HMRI, in writing, that these were significant

problems, which required positive and immediate action by Prism.

Mr. Stenabaugh was the only senior member of the staff at

- 115 -

HMRI who testified during the course of this trial. His evidence

about discussions between HMRI's senior staff and Prism concerning

problems with the software was as follows:

QNow, you say that people were reporting

back to Prism.

Do you mean Wendy

King talking to Prism about the

test results, that sort of thing?

AYes, and problem issues.

QAre you suggesting that anybody else at

HMRI,

and

specifically

senior

management staff, were talking to

Prism during the summer of 1986

about

software?

problems

with

the

Prism

AI don't know if other senior staff did.

I know that I didn't.

When the question became more specific the answer did not

appear to change:

The question, as I understood it, Mr.

Stenabaugh, was: Did anybody at

HMRI contact Prism and tell them

that the customers or users were

not satisfied and wouldn't remain

so unless the problems were fixed?

AI

don't

know

if

anybody

did

specifically.

QCertainly you did not?

I did not.

A

HMRI not only did not express major concern about the

software in writing, there is no evidence they did so verbally.

In fact HMRI was doing the exact opposite. In the HMRI Newsletter

- 116 -

which was sent to hospitals across Canada the reports were not

just positive they were glowing.

In the October 1986 Newsletter, published at a time when

according to HMRI the "contingencies" required the software to be

rewritten, HMRI included a full page under the title "SAVOIR-FAIRE

UPDATE". It is only necessary to quote two portions of the report

to reflect the nature of what HMRI was conveying to the hospitals

who constituted both its client base and its market:

Savoir-Faire has proven to be an innovative

and responsive medical record electronic

abstracting system.

Savoir-Faire is easy to

use, does ensure the accuracy of data

submissions and increases the productivity and

efficiency of the health record department

staff.

. . .

9.

Comments from Users of Savoir-Faire:

-"everything is going very

smoothly

with

no major problems"

-"staff is very enthusiastic

and

satisfied"

-"staff really like it and it

beats

working

with

a

pencil"

-"very user-friendly"

-"quick responses when calling

HMRI with problems or

questions"

-"coders adjusted like `ducks

to water'"

-

"great!"

-"presented

Report

the

at

Mortality

rounds

extremely well received"

and

it

was

- 117 -

Even allowing leeway for the fact these newsletters could be

considered "promotional literature" this is a pretty positive

portrayal of software which they now claim was critically flawed.

The discussions between Prism and HMRI during September of

1986 focussed on the future directions of their relationship

including enhancements to the software, add-on products and the

development of the ICD-9 CM versions necessitated by the agreement

to take over the CPHA sites. Although HMRI may not have written

correspondence to confirm these discussions their internal

committee minutes continue to document the nature of the

relationship.

The minutes of the Savoir-Faire Committee meeting

of September 8, 1986 record that:

4. SOFTWARE STATUS

a.Multiple Terminal Status - Scarborough

General is up and working. Toronto

Western Hospital has also been

installed and it is

terminal as well.

a

multi-

b.Provinces Completed & Tested - By the

end of September, all of the other

provinces will be tested (Nfld.,

Sask., P.E.I., N.S. and B.C.).

This version of the software is

86.6.

Once these have been tested

and are `bug-free', all of the

hospitals will receive this version

of the software.

c.All of the hospitals are running on

Version 86.4 with no problems.

Plummer Memorial is having problems

with their month-end, however, it

is

undetermined

at

this

time

- 118 -

whether or not it is a "software"

problem or "user" problem.

They

are sending in their data and it

will be tested here.

(emphasis added)

6.

INSTALLATIONS TO DATE AND SCHEDULED

All hospitals ordering Savoir-Faire have

been scheduled with the exception

of Norfolk General and this will

probably

be

scheduled

for

the

middle of November.

A. Adam recommended that:

a more aggressive marketing

strategy be developed

immediately. This should

include an evaluation of

the market.

10. ENHANCEMENTS

Mr. Colman is meeting with Prism next

week and will be discussing with

them

the

enhancements

to

the

It

software to include ICD-9 CM.

was recommended that:

J. Bickle and C. Colman review

Prism's Report Writer to

get costing and ideas.

The new microcomputer staff member will

probably do any other enhancements,

(e.g., what the market wants).

These reports, discussions and decisions as recorded are

directly contrary to the position advanced by the defendants, yet

this meeting took place one week before Mr. Imbert began his

employment with HMRI and only three weeks before he expressed his

opinion that the software should be replaced.

- 119 -

Returning for a moment to HMRI's decision to maintain the

Prism software after October 1, 1986, the significance of this

lies not in the amount of notice given to Prism but in the fact

that they exercised that option and communicated it to Prism.

With the greatest of respect to the submissions made on behalf of

the defendants it is inconceivable that such a decision would have

been made if the software was as bad as HMRI now claims.

Once again this issue is addressed in Mr. Stenabaugh's

evidence:

Mr. Stenabaugh?

Can you just tell me

what steps HMRI took from September

until

January to develop the

capability of maintaining the Prism

software?

AThe key thing we did was the hiring of

Mr. Imbert.

intention

That was -- his

originally

was

to

maintain the Prism software.

THE COURT:

Well, let's deal with the time

What, if

after Mr. Imbert was hired and he's doing

something completely different.

anything, were you doing to maintain both

versions of the software?

AWell,

up

--

we

never

did

take

We

responsibility for maintaining the

Prism version of the software.

still hadn't received an acceptable

version.

Prism's actions, and inaction during this period must be

measured and viewed in the light of a concerted effort by HMRI to

- 120 -

keep them in the dark as to what was actually happening at HMRI.

This is demonstrated most notably by the instructions issued by

Mr. Stenabaugh at the time of the training session in October 1986

that Mr. Jeffery was not to be told about the rewrite.

Perhaps most significantly the conduct of HMRI can be

measured in light of the answers of John Stenabaugh quoted above

and the fact that the training session was held at all, if his

evidence is true. Stenabaugh was the most senior person present

at the course. In his testimony HMRI, despite the notice given,

never did take responsibility for maintaining the software, yet he

insisted Imbert attend and directed that Jeffery not be told the

"real" state of affairs.

I am driven to conclude that from the moment that Imbert was

given the go-ahead the intention was to replace the Prism

software.

It may have taken months for the full import of the

decisions taken to have filtered up to the Board of Directors

level but there is little doubt that this was the way in which the

project proceeded. It is equally clear that it was the impact of

these decisions and the failure to communicate them which led

directly to the firing of the Executive Director, Mr. Colman.

Mr. Imbert's evidence on this point is clear and the

documentary evidence supports that opinion.

In the Savoir-Faire

- 121 -

Committee meeting of October 23, 1986 which gave the rewrite

"formal approval" there are recorded discussions about Marketing

Strategy and what is recorded as

a

"feasible 'start-up'

timetable." In the Liaison Committee minutes of November 11, 1986

it is recorded that:

Wendy and Steve are re-writing the Savoir-

Faire program for Ontario. It should be ready

by February on March of 1987.

This entry corresponds with that contained within the October

23, 1986 minutes themselves which record:

S. Imbert feels that he can rewrite the whole

program in approximately 2 months (ie. a

single terminal Ontario basic system by the

end of December).

The timing may be different but the intention recorded, an

Ontario version, remains the same.

The documents forming EXHIBIT T13-64 reflect a plan aimed at

replacing the Ontario system.

As I have already indicated on the whole of the evidence I

find that, from the inception, the plan was to replace the Prism

software.

It is neither probable nor likely that the people

involved in the various committee meetings to which I have

referred so misunderstood what was going on as to record an

entirely different intention than that actually discussed.

- 122 -

On January 27, 1987 Mr. Colman wrote to Prism, in this letter

he wrote:

I have asked Wendy to let you have the Alberta

specs but you should remember that in early

November, I advised you that HMRI had hired

its own staff as it wished to develop its own

Alberta and New Brunswick programs.

Originally,

I

had

anticipated

that

our

programmer would develop additional provincial

versions of your program, but subsequently we

chose to have him write a new set of programs

because of the difficulty in obtaining a

working copy of 86-6, the unacceptable - to

our users - length of time to run the month

end programs and the difficulty of maintaining

non standard ISAM.

This was the first indication to Prism that HMRI intended to

replace the Prism software and the reasons given were the only

ones given for the decision. Even at this late date there is no

mention of the fact that the "new set of programs" is being

written in a different language.

XIII THE ONTARIO GRANT

On March 6, 1987 as the relationship between Prism and HMRI

was entering its final throes the Ontario Ministry of Health

issued a news release announcing that "The Hospital Medical

Records Institute will receive a $2 million grant to install its

patient discharge data processing system in the provinces

hospitals. . ." Mr. Stenabaugh testified that HMRI learned about

- 123 -

the grant only a week before it was announced.

To properly assess the role, if any, this grant played in the

unravelling of the relationship between Prism and HMRI it is

necessary to go back and examine the historical relationship

between the Ministry of Health and HMRI. This relationship dated

back to the 1970's when the Ministry operated HMRI directly.

Initially HMRI operated its mainframe database on a mainframe

computer which belonged to the Ministry of Health. A measure of

the relationship can be found in the apparent fact that the

proposed contract with Prism was reviewed by the Ministry which

submitted its comments to the Personnel and Finance Committee

before the contract was approved and executed.

The relationship between the two included the presence of a

senior official from the Ministry of Health on HMRI's Board of

Directors.

Dreezer.

At the relevant times that official was Stephen

Once again the documents are revealing. As early as November

1986 the minutes of the Personnel and Finance Committee record Mr.

Dreezer suggesting a more active marketing strategy which would

include ". . . in particular. . . that HMRI submit to the Ontario

Ministry of Health a proposal to bring small hospitals in Ontario

- 124 -

He felt it would be to HMRI's

on the Savoir-Faire system.

advantage to have as much control as possible over the medical

records input."

On December 1, 1986 the Board of Directors is told:

The Ontario Ministry of Health has asked for a

proposal for putting Savoir-Faire into 25

hospitals in the 3,000 to 6,000 abstracts per

year size bracket, as an aggressive marketing

plan is being put together so that 70 to 75

implementations can be handled in 1987-88.

On the same date the Product Development Committee minutes

record the following report:

4.

UPDATE ON SAVOIR-FAIRE

Twenty-five

installations

have

been

completed and there are eight more

to be done. Single terminals

installations are working fine but

there are some problems with multi-

terminal installations.

There

are

approximately

forty

new

installations expected.

Projected - Alberta 15

New Brunswick15

Ontario 12 - 20

The Ontario Ministry of Health wishes to

encourage small Ontario hospitals

to use Savoir-Faire.

They will

assistance to

provide financial

HMRI in order to complete these

installations.

A

Savoir-Faire User Group was held

November 14, 1986.

It was

lots

a

successful

day

with

of

positive feedback.

The group felt

a Report Writer would be useful in

- 125 -

the future.

On October 3, 1986 a Liaison Committee meeting is told that

"The Ontario Ministry of Health will provide funding to the small

hospitals for twenty five additional installations." At a similar

meeting on January 13, 1987 the minutes note that "The funding

from the Ontario Ministry of Health for Savoir-Faire is on hold."

On March 11, 1987 the following entry appears in the minutes

of the Liaison Committee:

"Good news! It is official that the Ontario

Ministry of Health will provide funding to

install Savoir-Faire (find enclosed News

release).

The Ontario Hospitals will be

advised in the March Newsletter. The details

will be released by HMRI and the Ontario

Ministry of Health in the future."

Mr. Stenabaugh in his evidence steadfastly denied that he and

the staff of HMRI had known for months that "some sort of

government grant was highly likely to assist small hospitals in

Ontario in computerizing."

proposition:

Instead he agreed with the following

QIs it your evidence then that his

Ontario grant came out of the blue

like manna from heaven?

A

Yes.

During the course of

a

lengthy cross-examination Mr.

- 126 -

Stenabaugh maintained that the various entries between November

1986 and January 1987 had nothing to do with the grant announced

on March 6, 1987 but related to an earlier proposal put together

by Mr. Colman, Ms. Bickle and Mr. Stenabaugh of which nothing had

come. Mr. Stenabaugh could not recall any details of the proposal

he apparently jointly prepared.

Ms. Bickle's evidence on discovery appears to differ from

that of Mr. Stenabaugh:

The

consideration of the ministry was the small

hospitals. By the time the final grant was

grant

was

then

--

the

initial

delivered to HMRI, it was not for small

hospitals but was for all hospitals in Ontario

who were still using paper abstracts for

submission to HMRI. So the initial focus was

a small hospital. By the time they got up to

the figure of $2-million, it was to cover any

hospital that was on paper.

The final step on this issue came when HMRI opposed Prism's

application under the Freedom of Information Legislation in

Ontario and prevented the disclosure of documents concerning the

grant in the possession of the Ministry of Health.

The distinction drawn by Mr. Stenabaugh seems artificial in

light of the comments recorded in the records at HMRI. Whether or

not the initial entries relate to a proposal in a different form

than that of the final grant it is probable, in my view, that HMRI

was aware at least by November of 1986 that government funding of

- 127 -

some kind was likely forthcoming to assist in placing Savoir-Faire

in Ontario hospitals. I find that the likelihood of such funding

being available was one part of the backdrop to the events of late

1986 and the early months of 1987.

In so far as Mr. Stenabaugh suggests that the grant

announcement came as a surprise to HMRI I reject that evidence as

being out of keeping with the whole of the evidence. Finally, on

this point I would, in any event, draw an inference adverse to

HMRI arising from their opposition to the disclosure of the

documents related to the grant.

Absent some reasonable

explanation which was certainly not put before me I must, and do,

conclude that the documents would have disclosed HMRI's knowledge

of and participation in the process.

XIII SOME FACTUAL ISSUES

In approaching the issue of liability in this action and, in

particular, the remaining factual issues central to that

determination it is necessary to sift from the mass of evidence

that which bears directly on the liability issues or the factual

matrix in which they arise.

It is conceded that HMRI had the desire to expand its client

base and to establish it's position as Canada's national medical

records database. This may well have been an objective that was

- 128 -

being pursued by HMRI during the relevant time period but it does

little more than assist in an understanding of the events.

It is clear from the whole of the evidence that in a

discussion of what were characterized as "strategic issues" in

early 1986 HMRI elected a "Strongly Pro-Active" approach which

held the view that "the mission of HMRI should be re-defined to

place greater emphasis on data use, appropriate data analysis

tools, publications, training, and the provision of an independent

forum for debate and action on relevant issues."

That HMRI wished to pursue growth and a broadening of its

base seems beyond dispute.

Of perhaps more significance in the

resolution of the remaining factual issues is HMRI's relationship

with President Computers and Network Data Systems (NDS). Mr. Bill

Moore who served as an advisor to and sometimes as a member of

HMRI's Board of Directors appears to have been the source of the

close and friendly relationship.

President was the vendor HMRI recommended initially for

computer hardware in eastern Canada.

It is clear from the

evidence of Wendy King that when she experienced technical

difficulties in carrying out her duties she would consult Mr.

James of President Computers and to a lesser extent Mr. Craig.

- 129 -

One of the most perplexing problems encountered with Prism

Savoir-Faire was the difficulties encountered at multi-user

installations.

President supplied both the slave boards and the

ATNX software utilized in multi-user installations.

Mr. Moore had been a member of HMRI's Board of Directors for

a period of four years ending in 1980. In May of 1987 he returned

as a member of the Personnel and Finance Committee.

During the

intervening seven years he remained in contact with HMRI and

assisted in organizing the supply of hardware for the Savoir-Faire

project through President Computers.

The connection between

President and NDS lay in the personal relationship between the

heads of the two companys, Mr. Moore and Mr. Craig.

These reasons have already touched on Ms. King's weakness in

technical matters.

This weakness, I find, led her to seek

technical assistance where she could find it and at times that led

her to consult President in the persons of Mr. James and, on

occasion, Mr. Craig.

There is nothing inherently wrong with

seeking assistance in areas outside your own technical expertise

but when this outside assistance, was combined with a lack of

proper communication with Prism and HMRI's decision that Prism was

not to communicate with the hospital sites directly the potential

for difficulty emerges.

- 130 -

I find on the balance of probability that Ms. King consulted

President with respect to both hardware and software problems

during the material times. If there could be any doubt concerning

this it is laid to rest by Ms. King's correspondence of September

22, 1986 in which she writes to Mr. Craig:

I would like to again thank both you and James

for all your assistance during out Montreal

installations.

James was most helpful in solving both

hardware and software "problems."

I truly

appreciate the extra effort President has

shown.

It is of some passing interest that Mr. Imbert who had

relatively little to do with the Prism software because of his

almost immediate involvement in the Q-PRO-4 rewrite spoke to Mr.

James to find out what ATNX was and what its purpose was. It is

also notable that until he was "directed" by Stenabaugh to attend

the training session in October he had no contact with Prism or

either of its principals despite the fact he was hired ostensibly

to maintain the software they had written.

XV

LIABILITY

Two other major factual issues remain for consideration

and these are central to the underlying issue of liability, they

are, firstly, an assessment of the problems encountered with the

Prism Savoir-Faire program and its installations and, secondly, a

determination of whether or not the defendants in their

- 131 -

preparation of the Q-PRO-4 version copied the Prism software in

any way and if so the extent of such copying and the legal effect

it.

An approach to these two central issues involves not simply a

consideration of the specific evidence led with respect to them

and the submissions made on these issues but a drawing together of

the numerous other factual issues raised during this lengthy trial

to establish the surrounding circumstances in which these issues

must be viewed and understood.

Many of these issues have already been examined within these

reasons, but it is necessary to draw those various conclusions

together and to weigh them and the evidence as a whole in

resolving these critical factual issues. In doing so I am mindful

of the credibility issues raised and the forceful submissions made

on those issues by Mr. Chaster.

In Blatch v. Archer (1774) 1 Cowp. 64 Lord Mansfield had

occasion to address similar issues. At that time he expressed the

principle in these words:

It is certainly a maxim that all evidence

is to be weighed according to the proof which

it was in the power of one side to have

produced, and in the power of the other to

have contradicted.

- 132 -

In these eloquent words Lord Mansfield touches on two other

difficulties that arise in the present action, the failure to

produce evidence (or to allow it to be produced) on the one hand

and the failure to cross-examine a witness prior to leading

evidence on which he might have been able to testify on the other.

A great many things have changed in the field of evidence

since 1774, one of these is certainly the rules and procedure

which permit one side to call as a witness an adverse party who

can then be treated as "hostile" and cross-examined.

The

existence of these rules and procedures has removed some of the

force of the principles formerly applied when evidence was not

produced by a party, but it has not eliminated them, nor has it

resulted in a situation where the court may no longer draw an

adverse inference in appropriate circumstances.

The court must,

however, approach with caution a request that an adverse inference

be drawn and examine both the surrounding circumstances and the

applicability of the procedure envisioned by Rule 40(17) to

40(22).

One of the key provisions of this procedure is the

availability of the adverse witness to the other side for cross-

examination on the issues raised.

I am painfully aware of the length of time spent at the trial

- 133 -

of this action, and of the time that has passed before the

delivery of these reasons.

The issues raised were complex and

difficult, and in all actions some decisions must be made as to

what evidence is necessary for the determination of the issues

raised.

In the present case the absence of certain important

witnesses from the defence side, most notably Robert Colman,

Jeanne Bickle, Lynne Abbott and Barbara McLean, left a significant

gap in the evidence.

Mr. Colman's absence may be explainable in terms of the way

in which he left the employ of the defendant institute, but the

fact that the plaintiff's had a witness statement from him does

not require them to place him in the witness stand at the trial of

this action. Indeed, such a course of action given the breadth of

the issues, the seriousness of the credibility issues, and the

potential animosity of Mr. Colman may have been foolhardy in the

extreme.

The absence of these witnesses must be measured not simply

against the evidence they could have given in crucial areas, but

also in light of the positions taken by the defence in this action

and in relation to it.

Among these I consider HMRI's repeated insistence that the

internal

minutes

produced

are

in

significant

particulars

- 134 -

inaccurate and untrue, HMRI's opposition to the production of

documents concerning the Ontario grant, and their payment of

substantial sums of money for the attendance at trial of a named

defendant to be significant factors.

In my view, this is an action in which the circumstances

warrant the drawing of inferences adverse to the defendants.

I

have already touched on some of these specific areas and I will

within these reasons detail other inferences which may be drawn

from the evidence, or its absence.

A.

PROBLEMS WITH PRISM SAVOIR-FAIRE

The main thrust of the defence in this action is that

the Prism software was so flawed as to be a continuing and

unacceptable problem to both HMRI and its client hospitals. These

flaws justified, according to the defence, not only the decision

to rewrite the program but the decision to replace the software

entirely.

Curiously the defendant's approach to the legal issues, as

opposed to the factual ones, takes a slightly different tack. In

their written argument the defendant's assert that:

385. In defendants' submission, the problems

with Prism software are only relevant if the

Court should find that the Q-Pro-4 software is

not a modification or updated work as set out

in T6 AGREEMENT 8.

If HMRI Q-Pro-4 Savoir-

Faire is a modification or updated work under

T6 AGREEMENT 8, then HMRI says that it had the

- 135 -

right to write the Q-Pro-4 Savoir-Faire

whether or not Prism's programs had problems.

With the greatest of respect to this submission it misses the

point and represents a continuing theme in the submissions of the

defendants.

The task of the court in a civil trial in this

country is to first determine the facts and then to apply the law

to those facts. This simplistic statement serves to obscure the

real difficulties which emerge from the practical application of

that process, but, in my respectful view, it is necessary to keep

that two step process firmly in mind particularly in complex cases

where the issues are usually interrelated.

It is not either proper or desireable to exclude material

evidence from a resolution of the factual issues on the basis that

it is only "relevant" to a particular legal issue.

The first

phase of any decision is the resolution of the factual issues and

in examining individual pieces of evidence they must be placed

within the factual matrix in which they occurred.

In the present case the plaintiff's assert that their

software worked not just adequately but very well and that the

difficulties encountered were either normal problems anticipated

during the development of software with this degree of complexity,

or problems caused by HMRI's activities and not brought to Prism's

attention.

- 136 -

This is the central factual issue underlying the entire

action, either the software worked adequately or it didn't, if it

did then the credibility issues relied upon by Prism take on a

whole new meaning as do the explanations offered to this court by

the defendants.

The starting point in this analysis is the letter written by

Mr. Colman to Prism on January 27, 1987. In this letter Prism was

advised, in writing, for the first time that the Prism software

was being replaced. The decision to write a "new set of programs"

is said to be based on three difficulties:

(1)the difficulty in obtaining a working

copy of 86-6,

(2)the unacceptable - to our users -

length of time to run the month end

programs, and

(3)the difficulty of maintaining non

standard ISAM.

No other reasons for this decision were given to Prism prior

to the commencement of the present action. In the defence of the

action a host of other difficulties were put forward for

consideration.

i)

WIDE DISTRIBUTION

The first of these issues as set out in the Statement of

- 137 -

Facts and Issues was the allegation that:

. . . Prism had no experience with systems

which had been installed on more than five or

six sites, it did not anticipate problems

which

occur

when

a

system

is

widely

distributed to microcomputers users. . .

This issue can be dealt with quickly. No evidence was placed

before the court in support of this allegation and there is no

evidence to support the proposition that this created problems

which were not corrected.

ii) DESIGN SHORTCOMINGS

The defendants assert as a general proposition that some of

the individual problems arose as a direct result of Prism's

failure to take reasonable steps in its programming and to build

in a level of "robustness" which would tolerate the activities of

users having different degrees of experience and expertise in the

use of microcomputers.

In support of this general proposition they point to a number

of problems with the Prism software. One of these was a "problem"

which was advanced by the defendants in several different ways as

being both a source of difficulties and unnecessary to the Savoir-

Faire project as envisioned by HMRI.

The Prism software by its

underlying design provided for large data files. When a month end

was run with Prism Savoir-Faire the abstracts sent to HMRI were

- 138 -

added to a cumulative History file which, in the larger hospitals,

became quite large.

The size of this History file in larger

hospitals is what led to the use by Prism of the file extender

Golden Bow. This software was a means whereby the internal limit

of DOS with respect to the maximum size of files (32 MB) could be

circumvented to allow larger files.

HMRI submits that this design which permitted large files not

just to record the History but also the hospital specific data

contained within a single Ancillary file led to a series of

problems (a)

index problems, (b)

the use of Golden Bow itself

and (c) the problem with long month end procedures.

The index problems referred to occur when some portion of the

data contained within a file is damaged. When this occurs in most

cases the damaged file becomes inaccessible.

Prism recommended

that when this occurred the data be restored from the last tape

backup which would result in the loss of only the data entered

since the last backup was made.

When Scarborough General Hospital,

a

multi-user site,

encountered such a problem Mr. Jeffery was contacted and he wrote

a specific utility NEWINDEX which was provided to them to re-index

their data file. This was not intended to be used generally but

rather had been necessitated by the fact that Scarborough had not

- 139 -

been backing up their data.

The program, however, once created

was used to restore data in other situations.

There were initially a significant number of "error 63"

messages recorded.

Most, if not all, of these on the evidence

related to multi-user sites, and will be dealt with later within

these reasons.

The evidence falls far short of demonstrating that the "index

problem" was one that extended outside the multi-user sites or

that it represented a significant and ongoing problem with single

user installations.

In the Statement of Facts and Issues the issue concerning

Golden Bow is set out in p.44:

HMRI contends that Goldenbow[sic] gave rise to

a

number of problems, that its use in

association with the other software forming

Prism Savoir-Faire was not adequately tested

by Prism and that MICROSOFT, which supplied MS

DOS and QUICK BASIC, would not support these

products if Goldenbow was used with them.

The difficulty with this assertion is that there was no

evidence led to indicate that any specific problem arose from the

use of Golden Bow or from any aspect of Prism's testing of the

product.

- 140 -

There was, in addition, no evidence that MICROSOFT would not

support its products. Mr. Van Toen in his evidence in fact agreed

that MS DOS was in fact designed with "hooks" specifically for the

purpose of utilizing such products.

HMRI's case against Golden Bow, at least in so far as it was

presented at trial was far from "water tight", it was in fact non-

existent.

HMRI's real concern about Golden Bow and the History file was

that it was unnecessary for the abstracting system which HMRI

required. In designing the history file the way they did, Prism

was not fulfilling the requirements of HMRI's abstracting system,

but providing a capacity and a capability that was designed to

promote Prism's purposes. This added some costs to the system for

a hospital acquiring it and, according to HMRI, added a further

level of complexity to the system which ". . . made maintenance

and tracking of bugs more difficult. . . for no reason necessary

for the HMRI installation."

This complaint summarizes in pristine terms the nature of

many of HMRI's complaints as presented at trial.

There is no

contractual term which restricts the design of the software in

this way, and, in reality, it was this design which from Prism's

point of view gave the product value to hospitals and to Prism in

- 141 -

that, over and above the requirements of HMRI, it allowed the

hospitals to create, maintain and manipulate its own in-house

database.

It was this feature which provided Prism, potentially with

the market for its add on products and enhancements.

As has

already been indicated this was the value of the agreement to

Prism which was specifically contemplated within the contractual

documents.

When two parties negotiate and enter into a contractual

relationship they often see for themselves different benefits

flowing from the objects of the contract.

So long as these

benefits can be reached without breaching the terms of the

agreement, in most circumstances there can be no complaint.

The fact is that, absent a contractual term or some breach of

other terms, HMRI had no right to require Prism to abandon the

value they saw within the agreement reached.

Yet this is

precisely the thrust of HMRI's submissions in this aspect of the

case.

In their submissions on this point the defendant's rely on

the evidence of Mr. Bardswich and his expressed opinion that

hospitals do not want reporting capability and therefore do not

- 142 -

need the history file provided in Prism's program design.

Mr. Bardswich stood virtually alone on this issue in the

evidence presented at trial. The plaintiff's called a series of

witnesses whose experience lay in the area of medical records,

these included Geraldine Melnechuk, Daniel Neibert, Leslie

Carlson, Harold Schmidt and Angela Breland.

Each of these

witnesses had experience with the software produced by Prism at

hospitals located variously in Alberta, Toronto and various

locations in British Columbia. Each in their own way testified as

to the value of accessing their history files and utilizing the

various reports this made available to their particular hospital.

It is of interest as well that both Ms. King and Mr. Imbert

through Cactus now make their living by providing the means for

hospitals to manipulate historical data to generate reports.

The balance of the evidence lies firmly against Mr.

Bardswich's opinion and I specifically find that the history files

are of value to the hospitals as is the ability to manipulate that

data and generate reports.

Finally despite their present position HMRI continued to

promote in its newsletters the Savoir-Faire program by underlining

the existence of the history file and by advising client hospitals

of their intention to offer enhancements capable of generating

- 143 -

"specialized reports".

- 144 -

iii) THE CHOICE OF PROGRAMMING LANGUAGE

Once again this issue was defined in the Statement of Facts

and Issues p. 42:

42. Prism Savoir Faire was written in QUICK

BASIC 1.0 and its updates 1.01 and 1.02. HMRI

contends that Prism should not have used QUICK

BASIC because it was a completely new language

and had severe limitations on the size of the

programs it could produce.

Further, HMRI

contends that Prism was unable to learn how to

"chain" together in QUICK BASIC 1.0 and

updates, resulting in programming limitations.

Prism contends that QUICK BASIC was not

completely new and that the difficulties with

QUICK BASIC, if any, were not insurmountable.

Prism says further that it made full

disclosure to HMRI of the fact that it was

using QUICK BASIC and HMRI acquiesced.

It is useful to begin consideration of this issue by

reviewing s. 2 of the original proposal executed by HMRI on

January 29, 1986. It provides as follows:

At the request of HMRI the software will

be written in BASIC as appropriate for use on

'PC' type micro-computer equipment which uses

the MS-DOS operating system.

It is apparent that HMRI knew about the use of QUICK BASIC

since sometime in the fall of 1985 and there is no evidence of any

complaint concerning its use.

On this issue as is the case with other similar complaints

the defendants position and the evidence tendered does not

identify particular problems and relate them to the particular

- 145 -

criticism. Instead the evidence on which the defendants rely is

contained within the expert report and evidence of Gordon Van

Toen.

The nature of the evidence relied upon by the defendants in

support of this criticism can best be appreciated by reviewing the

following portion of Mr. Van Toen's report:

The QuickBASIC manual indicates that

version 1.0 was MicroSoft's first BASIC

compiler for the PC/MS-DOS operating system.

There are some obvious restrictions documented

in the manual concerning the size of programs

that can be produced (only 64 Kilobytes) and

some of the language constructs which can be

used (no use of COMMON in CHAINed programs in

the executable mode). Such restrictions mean

require the programmer to "code around" them,

to write more complex code in order to

accomplish the same functions.

This, of

course, takes extra development time and makes

for difficult maintenance.

The manual itself does not document all

the commands (there is nothing documented for

the OPEN statement for example) and suggestion

is made to "see the reference manual that came

with your BASIC interpreter for a complete

reference to the functions and statements that

work identically in both interpreted and

compiled programs" (page 103, D-424).

In

other words, the manual is not complete in

itself.

Using two manuals, one of which is

not complete and the other of which is not

100% relevant, adds un-necessary complexity to

the programming task.

Such complexity will

result in a greater effort being required and

poorer code.

The decision to use the initial version

of a compiler for major development is always

inadvisable because bugs may be expected. The

presence of such bugs makes programming,

testing and maintenance much more difficult

because the programmer has to take time to

- 146 -

determine which errors are his and which are

caused by the compiler.

In

the

case

of

QuickBASIC

it

is

significant to note that the READ.ME file for

Version 1.02 lists 43 compiler errors which

were in the prior version. Most, not all, are

identified as being corrected in Version 1.02

Systems developed using the previous versions

would have contained the bugs these errors

introduced.

discernable

Failures would occur for no

reason and effort would be

expended in tracking down the true causes.

While these comments may be of passing interest they are of

little if any assistance in demonstrating that the choice of

language was inappropriate or that this choice resulted in any

problems of significance to this action.

The other difficulty with the passage quoted from Mr. Van

Toen's report and its utility in the consideration of the issues

to be resolved in this action is that it does not mention the fact

that the compiler used with Microsoft Quick Basic was common to

previous Microsoft versions including Microsoft Basic, IBM Basic

and other versions going back some years, and that the evidence

indicates that the compiler in use was Microsoft BASIC Compiler

version 5.6.

Mr. Van Toen in his cross-examination agreed that the

compiler had been around for a number of years but testified that

with the move to different hardware some problems might be

- 147 -

expected.

What flows from this is that the type of general observations

contained in Mr. Van Toen's report are, with the greatest of

respect simplistic, lacking in real analysis and are provided with

no link to any actual problems which occurred.

are of no assistance.

In the end they

Mr. Van Toen in his evidence did, however, comment on Prism's

use of subroutines written in Assembler. This was another

criticism of the Prism software raised in the defence of this

action, by HMRI.

It is significant in considering this particular criticism

that attached to the minutes of the first Savoir-Faire Committee

meeting of March 17, 1986, held nearly a month before the first

installation, there is a document entitled Software which includes

under the second section the following:

Programs

-programs are modular

-written in compiled BASIC

-hospitals

only

receive

executable code

-compiler

QUICKBASIC

-

MICROSOFT

-one module is written in

ASSEMBLER

-structured as called routines

from main program

(emphasis added)

- 148 -

There is no record of any complaint about the use of

assembler for some subroutines before the dispute between the

parties arose.

Mr. Van Toen testified that the BASIC language allows for

"calls" to assembler subroutines because some things are easier to

do in that language, he went on to agree that the use of such

programming techniques was accepted programming practice.

iv) THE MULTI USER FAILURES

On the evidence it is clear that some serious problems were

encountered with the operation of multi-user sites at which Prism

Savoir-Faire was installed.

To place this problem in perspective two things must be

remembered, firstly that there were a total of six Prism Savoir-

Faire multi-user sites installed in Ontario and Saskatchewan, and

secondly, that these installations were at larger hospitals.

Regrettably, to resolve the issue arising from these failures

it is necessary to examine in some detail the way in which the

Prism system sought to provide multi-user capability.

Prism's solution to providing multi-user function was to

provide two components and a particular process.

It used slave

- 149 -

boards and software known as ATNX. The difficulty to be resolved

was providing a means by which more than one user could access

data files at the same time. ATNX provides a means of allocating

disk space by dividing it among more than one user; when ATNX

allocates disk space to one user it is not accessible to other

users.

When data is added in that space by one user it is not

accessible to other users and an attempt to access it by another

user causes the system to "hang up" or "crash".

The solution to this problem recommended by the manufacturers

of ATNX and adopted by Prism was to "pre-extend" or "pre-allocate"

the existing data files prior to invoking ATNX. In essence this

solution created a series of blank or dummy records in the common

area prior to ATNX being invoked.

The number of blank records

created depended on the monthly volume of abstracts submitted by

the particular hospital. When ATNX is then invoked the remaining

disk space is divided between the users leaving the existing data

file and the blank or dummy records in the common area accessible

to all users. When one of the users then created a new record it

was written into the pre-extension area and replaced one of the

blank records.

The specific means by which this process took place is

important on the facts of this case.

When a Prism Savoir-Faire

system was started DOS searches the root directory for a batch

- 150 -

file AUTOEXEC.BAT which contains a series of commands and executes

it.

In the Prism system the AUTOEXEC.BAT file does a number of

things then invokes a file named CHECKISAM. This file determines

whether or not the system is a multi-user system and if it is it

determines from the control file the number of slaves and the size

of the hospital (ie. the number of abstracts per month) and pre-

extends the data files to allow space for new records to be

entered. It then writes the necessary commands to a file called

STARTDAY.BAT which it then runs invoking ATNX, changing to the

HMRI directory and starting Savoir-Faire.

The placement of these files is important to the operation of

the system. AUTOEXEC.BAT is located in the root directory where

DOS looks for it when it starts, STARTDAY.BAT is located in the

HMRI directory and when that command is issued DOS searches its

tree directory starting with the root directory until it locates

the file STARTDAY.BAT in the HMRI directory.

One other process provided by Prism is also important. One

of the first functions carried out by CHECKISAM was to check what

was called a "dirty disk flag."

This "flag" is checked to

determine whether the program was properly shutdown when it was

last run. If it wasn't then a warning message "WARNING, WARNING,

WARNING" was displayed and the user was directed to restore data

from their last backup.

This security system was intended to

- 151 -

protect the program from data files which may have been damaged by

an incorrect shutdown procedure, and to provide an opportunity to

restore the data.

Mr. Jeffery testified that in preparation for trial he

recognized that many of the difficulties encountered with the

multi-user system were likely attributable to the fact that

someone other than Prism had altered the AUTOEXEC.BAT file and

placed a STARTDAY.BAT file in the root directory.

Mr. Jeffery

testified that someone had changed the AUTOEXEC.BAT file by

placing the command invoking ATNX ahead of the pre-extension

routine. The effect of this was to allocate the diskspace prior

to pre-extension and thus prevent Prism's "solution" from taking

effect.

Mr. Jeffery found instances in which STARTDAY.BAT had

been placed in the root directory, because of the way in which DOS

searches initially in the root directory any STARTDAY.BAT file

placed there would be executed rather than the one created in the

HMRI directory by CHECKISAM.

These changes, particularly the invoking of ATNX before the

pre-extension was carried out, would create the very situation the

Prism system was intended to avoid, and lead to a situation where

the computer would "hang" or "freeze".

users frequently turned their computers off then on again to

restart the program. When they did this they had not properly

When this occurred the

- 152 -

shutdown the program and as a result the "WARNING, WARNING,

WARNING" message was displayed.

Rather than restoring data as recommended, the error logs

introduced in evidence clearly demonstrate that HMRI instructed

users to use Control "C" or Control Break to bypass the warning

screen.

This procedure which was apparently routinely given by

liaison staff at HMRI not only served to bypass the security

feature but created a potential for damaged or corrupted indices

in the hospital's data files.

Mr. Jeffery reviewed the error logs, his testimony was:

QIn your evidence, in very broad terms,

can it be said that there is a

categorization of the types of

problems we see in the problem logs

based on your review?

AYes.

When I did my review, I hadn't

realized what was causing the

problem, but that review indicated

to me that the preponderance of the

problems came from stuck screens,

and looking at it now, in not

restoring, and, in fact, I think --

I

noted

that,

they

were

not

restoring in my initial review.

Q

Yes?

AThat would indicate to me that the

preponderance of the errors that

occurred,

especially

in

the

Scarborough General Hospital, were

due to the incorrect procedures

with regard to ATNX, either the

setup, the pre-extension, and very

definitely the recovery.

- 153 -

QWhat do you mean by the recovery?

AThe Control C-ing out of the Warning,

Warning, Warning message instead of

restoring.

Each of these would

magnify the problems of the other.

If there was more than one of

those

situations

present,

the

problems would be magnified; and

the not correcting the problem in

the first place as it occurred, in

other words restoring, caused the

system to just get worse and worse

and worse.

QCan you say, based on your review of all

problem logs, can you give an

approximate breakdown between what

relates to singles and what relates

to multi-user installations?

AThere are proportionately very, very few

error logs reporting that come from

single-user systems.

It is once again necessary to turn to the submissions made by

the defendants with respect to both Pre-extension and the use of

ATNX and the way in which it was invoked:

417. As has been discussed above (DFARG377)

ATNX is a stand alone package for controlling

slave machines, and can be used without

Savoir-Faire.

ATNX was supplied by the

It is

company supplying the hardware.

submitted that it would be reasonable to

expect that it can or will be loaded onto the

system by someone other than Prism.

As the

ATNX Manual says that the ATNX command can be

loaded into the AUTOEXEC.BAT, it is reasonable

to expect that someone will do this.

418. It

is

submitted

that

it

is

poor

programming by Prism to have a sequence such

that ATNX must be only loaded through Savoir-

Faire, or else pre-extension will not occur

properly. There is no warning to HMRI or to

users that ATNX must not be loaded before

- 154 -

invoking Savoir-Faire.

419. Thus, it is submitted that the appearance

of an ATNX command in AUTOEXEC.BAT is

something which Prism should have foreseen,

and should have arranged its program so that

it would work, even if ATNX had been loaded or

if a loading command for ATNX was in the

AUTOEXEC.BAT.

Alternatively, Prism should

have given a warning that AUTOEXEC should not

be loaded before Savoir-Faire.

The defendants in this portion of their submissions suggest

that Prism is guilty of poor programming practices and that they

should have anticipated changes by other individuals involved in

the process.

This portion of the argument demonstrates the danger of

isolating individual events from their surrounding circumstances.

Once again it is necessary to leave the minutiae and take the

broader view. What would a reasonable man in Prism's shoes have

foreseen? Prism provided the software to HMRI for testing. HMRI

made the decision as to when the testing was sufficient and when a

particular version of the software would be released.

When a system was sold to a hospital the hardware was

delivered to that site and the software was installed and training

arranged by HMRI. Included within the software provided by Prism

to HMRI was the AUTOEXEC.BAT file which they expected to be used.

- 155 -

These systems were sold to enable hospitals to use Savoir-

Faire, they were not in any sense general use computers, although

I have no doubt that in some cases hospitals may have wished to

use other software on them.

The fact remains what was expected

was that HMRI would test the software and report any problems to

Prism. They would then install the software provided by Prism and

test the installation, again reporting any problems to Prism.

It was completely reasonable in my view for Prism to assume

that their software would be installed as delivered to HMRI and

given the specific use for which these systems were sold and

HMRI's involvement in the installations that no changes would be

made.

It is always possible to review the programming in retrospect

and suggest improvements or ways in which situations, not

anticipated originally, could have been prevented. This case is,

in many respects, proof positive that this is so, however, in my

respectful view the majority of difficulties which occurred with

the multi-user sites were caused, on the balance of probabilities

by the changes detailed in Mr. Jeffery's evidence. I accept his

testimony that those changes were not made by Prism and conclude

that these difficulties occurred as a result of the activities of

HMRI and changes they made to the systems or allowed others to do.

- 156 -

These difficulties, in the end arose from the very thing

noted earlier in these reasons, Ms. King's lack of technical

expertise and were compounded by her inability to adequately

communicate the details of problems to Prism as they occurred and

HMRI's decision to limit Prism's access to the user sites.

It was this combination of circumstances, all of which arose

from decisions directly within the control of HMRI which gave rise

to the problems with the multi-user sites and then compounded the

problem and made the diagnosis and correction of the problems more

difficult.

It is clear from the evidence that direct access to

problem sites would have allowed better information to have been

provided to the programmer and more chance of success in the

correction of such problems.

I have touched on the difficulty in communication which arose

between Wendy King and Prism, and indeed between HMRI and Prism

generally.

This problem manifested itself in a number of ways

most obviously in the one sided nature of the written

correspondence. Nowhere, however, did the lines of communication

cause a greater potential for problems than those between Wendy

King and John Jeffery. HMRI made a series of critical decisions

concerning the completion of this project, (a) they decided that

they would through their liaison representatives and not their

- 157 -

systems department carry out the testing which they undertook and

determine when a particular version would be released for sale to

its client hospitals, (b) during that process they communicated

test results and change requests to Prism for action, (c) they

placed Wendy King in overall charge of the project and of liaison

contact with Prism, presumably with full knowledge of her level of

technical expertise, (d) they made the decision that all contact

between Prism and the individual hospital sites would be routed

through HMRI and with only minor exceptions this was done, and (e)

they left Wendy King, in particular, with numerous other duties

and responsibilities, in part at least because of serious staff

shortages at HMRI.

These decisions critically affected both the timely and

accurate flow of information to Prism during the development stage

of this project.

Both by inactivity and by positive actions taken with the

hospitals which were not communicated to Prism (such as the

instructions to bypass the security represented by the "Warning,

Warning, Warning" screen), Wendy King in no small measure was

responsible for creating many of the problems the defendants now

rely on to demonstrate the inadequacy of the software and by her

failure to communicate adequately with Prism, their apparent

failure to respond to those problems.

- 158 -

v)

THE "WARNING, WARNING, WARNING" MESSAGE

The defendants submit that this message screen which, on

its face, warns the user that the data files may be corrupt and

prompts the user to restore the data from their last backup was

poorly designed.

This screen and warning is displayed whenever

the user shuts off the system without going though the exit

procedure, whether or not there is any problems with the data

files.

The heart of Mr. Van Toen's criticism of this procedure is

that there is no direct relationship between the warning message

and actual damage to the data files. Mr. Van Toen testified that

there were at least two methods known in 1986 which could have

been used to test the integrity of the data.

In the first of

these a "dirty data flag" could have been utilized to signal the

opening and closing of each data file, in the second the software

would be designed to maintain a count of the records in a file and

to test that count when the file is reopened.

Either of these

methods would link the warning more directly to the actual

integrity of the data files and constitute better programming

practice.

This criticism is both accurate and valid.

As a test for

- 159 -

data integrity the programming left much to be desired.

The

plaintiffs, however, submit that this was not the major purpose of

the warning. That purpose was to ensure that a user went through

a proper shutdown sequence each time which would minimize the

likelihood of the data files being damaged and avoid the display

of the warning message when the system was restarted. Rather than

forcing users into a pattern of shutting down the system properly,

HMRI's advice to use control "C" to bypass the screen compounded

the potential problem.

The fact remains that this particular criticism must not be

viewed, in any event, in isolation but rather in the context of

the problems as they occurred.

While the criticism levelled by

Mr. Van Toen in this respect is valid, it too narrowly views the

purpose of the programming. In so far as it seeks to address the

specific problems which occurred it does not adequately address

the root cause of the data problems which was the changes made to

the system which were previously discussed.

vi) HARDWARE PROBLEMS

Under this category there were two specific complaints,

the first of these was the use Prism made of the memory space

contained in the Hercules Graphics Card and the allegation that

this ". . . caused problems with the operation of Savoir-Faire"

and "limited the usefulness of Prism's software."

- 160 -

Not only was there no evidence of actual problems created by

this use of the Graphics card or of any limited usefulness of the

software, there was only one incident touched on in the evidence

at all.

This was a failed installation at Woodstock Hospital

sometime after April 10, 1986. HMRI attempted to install Savoir-

Faire at that time on an IBM personal computer which was not

equipped with a Hercules Graphics Card, despite the fact that HMRI

was aware from at least November 1985 that the hardware

requirements laid down by Prism required the presence of a

Hercules Graphics Card.

presented in the evidence.

No other complaint or problem was

The second hardware complaint related to an allegation that

in some cases Prism supplied hard disks which were rated by

manufacturers for Modulated Frequency Method (MFM) and reformatted

them to run with Run Length Limited (RLL) disk controllers.

specific problems related to this process were identified despite

a lengthy cross-examination of Mr. Jeffery. Once again the

No

defendants rely on Mr. Van Toen's evidence that speaks of the

"possibility of disk failure and loss of data" and goes on to

suggest that "In case of disk failure, if the manufacturer knew of

the use, . . .[he] could reasonably refuse to support or warranty

the disk."

- 161 -

This evidence is not helpful, in the present case the

evidentiary base does not establish that any hard disks were

"originally sold as MFM controlled" nor has a single problem been

identified as flowing from such a process in any event.

vii) MONTH END PROCEDURE

HMRI consistently complained about the length of time

Prism Savoir-Faire took to run its month-end procedures. This in

turn arose from complaints from hospitals concerning the process.

The underlying reason for the length of time this process

took was the design of the program to incorporate a history file.

The Prism system maintained abstracts not yet sent to HMRI in a

"day file" which during the month end procedure was merged with or

added to the history file.

This process took a significant length of time but it

produced a history file which could be utilized by hospitals to

analyze their data and to generate reports.

I have already concluded that such a capability, on the

evidence, was of value to hospitals and it was obviously the

critical component providing Prism with the future value and

opportunity it had bargained for. Prism suggested that hospitals

should be instructed to run their month end procedures overnight.

- 162 -

In my respectful view this was not an unreasonable suggestion

given the balance of interest involved.

viii) CONCLUSIONS

It is necessary to resolve the apparent conflict between

what is recorded in the records of HMRI and the complaints

concerning the Prism software and the positions taken by HMRI at

this trial. It is grossly simplifying things, however, when entry

after entry in the records of HMRI indicates that the software was

working with the exception of "minor problems" and the troublesome

difficulty encountered with the multi-user sites, it is difficult

to accept the positions taken in the defence of this action.

Mr. Stenabaugh in his evidence sought to address this problem

in his evidence in chief:

QI'm sure that a question will be raised

in these proceedings as to what

problems are -- with the Savoir-

Faire committee -- sorry, with the

Savoir-Faire system, are discussed

at that committee.

Is there a

reason why problems would not be

discussed at that committee?

AThere was staff deliberately trying to,

I would say, I guess "downplay"

perhaps is the right word, the

severity of some of the problems

because we felt at the time that

staff

were

looking

into

these

things and they were dealing with

them or trying to deal with them.

We were expecting the problems to

be resolved by Prism, and we didn't

want to cause undue concern on the

committee

because

the

committee

- 163 -

because

members,

of

their

association at the national level

and the fact that the Savoir-Faire

committee

minutes

went

to

the

directors as well, then if we

identified severe problems, then

that might have been taken out of

context

and

people

across

the

country would be aware that there

were problems which we felt -- we

thought were going to be addressed

and resolved.

So in a sense we

didn't want to cause undue alarm

and upset the potential marketing.

QWhy was the fact that the minutes that

went to the directors a concern?

AThe directors were from across the

country and representing various

hospitals and organizations, and

that would be -- they would be in a

position

to

influence

decisions

elsewhere in terms of what they

might do with regard to Savoir-

Faire.

This concern is perhaps understandable, although it is

difficult to understand how "downplaying" the severity of the

problems translates into reports such as ". . . all of the

hospitals are running on Version 86.4 with no problems." (Savoir-

Faire Committee Minutes, September 8, 1986); "All hospitals are

running on 86.4 with no problems." (Product Development Committee

Minutes, September 21, 1986); "Ms. Bickle reported that HMRI had

no difficulties when installing single terminal systems but had

encountered problems with multiple terminal systems for larger

hospitals" (Personnel and Finance Committee Minutes, November 25,

1986); "25 installations have been completed and there are 8 more

- 164 -

to be done. Single terminals(sic) installations are working fine

but there are some problems with multi-terminal installations",

(Product Development Committee Minutes, December 1, 1986);

"Ontario and Quebec Installations will be done using 86.4 until

such time that 86.6 is fixed and has proven to be better than

86.4. 86.4 is working well in Ontario and Quebec." (Savoir-Faire

Committee Minutes, February 18, 1987); and as Robert Colman

reported in his report on May 12, 1987:

Although

terminal

Ontario,

teething

installations

British Columbia,

problems

with

installed

Saskatchewan,

single

now

in

Quebec, P.E.I. and Newfoundland were soon

overcome, no orders from multi-terminal

installations were accepted after last June,

and four of the five initial multi-terminal

installations were reduced to single terminal

systems early this year as multi-terminal

systems performance deteriorated in spite of

our best efforts to improve it.

Toronto

Western Hospital's multi-terminal system - our

sole

remaining

multi-terminal

system -

continues to operate to that hospital's

satisfaction!

(attached

Directors

12, 1987)

to

Board

Minutes,

of

May

All of these entries with the exception of the first one

occur after Mr. Imbert was hired and all but the first two occur

after he has begun the Q-PRO-4 rewrite.

These are not by any

means all of the entries contained within the records but they are

representative of what was being recorded.

It is clear, in my view, on the whole of the evidence that

- 165 -

Version 86.4 was a successful version of the software and that it

functioned

adequately

with

the

exception

of

difficulties

encountered in multi-user installations.

Of more significance is the fact that these entries do not

stand alone. Inexplicably absent is any letter from HMRI advising

Prism that there are major problems and that they must be fixed.

Even Mr. Colman's obstreperous letters of October 23, 1986 written

the day formal approval is given to embark on the rewrite do not

contain any such suggestion.

HMRI continues to publish glowing

reports, including user endorsements in its newsletters and the

test results on version 86.6 slow to a trickle or stop entirely

for nearly two months.

There is no evidence that HMRI made any significant effort to

obtain a final working copy of 86.6 free of the major errors they

now allege, in fact, the whole of the evidence indicates that

their entire efforts were focussed on the writing of the Q-PRO-4

version of Savoir-Faire at the expense of any significant testing

of the Prism product.

In pursuing this project they not only

failed to keep Prism informed of their requirements and problems

with Prism Savoir-Faire but actively and calculatedly set out to

deceive Prism and keep from them their activities.

There were clearly problems with at least two releases of

- 166 -

Version 86.6.

fields.

In the first Mr. Jeffery had failed to test two

There is no evidence that there were any difficulties

with the second release which was shipped on September 28, 1986

and, in any event, it certainly could not have been tested before

Mr. Imbert expressed his opinion with respect to the software on

September 29th or 30th.

Release

3

contained the corrupt subroutine previously

referred to.

What is significant about the entire course of events once

they are reviewed in the light of the whole of the evidence is

that before the problems with 86.6 release 3 became apparent and

indeed after the success of version 86.4 Mr. Imbert is already

rewriting the software.

Version 86.5 was forwarded, I find, by

Prism on July 22, 1986 and never tested by HMRI.

What happened at HMRI was what I have already described,

namely that Mr. Imbert had come upon the scene and having very

quickly assessed the situation had taken control of the project

and made it his own, he quite literally swept Wendy King off her

feet and enrolled her effectively as his assistant on his Q-PRO-4

project.

The secrecy of his rewrite project, the absence of any

- 167 -

written demands to Prism, the positive reports contained within

HMRI's records and newsletters, the failure to inform the

Executive Committee or the Board of Directors of any "major

problems" or indeed Imbert's project at all and the evidence as a

whole all lead to one conclusion, the present complaints and

criticisms

of

the

Prism

program

are

afterthoughts

and

retrospective analysis which seek to justify decisions made for

quite different reasons.

I reject the defence's suggestion that there were major flaws

in the Prism product which Prism failed to correct, there were

certainly problems during the development as both parties

anticipated, but nothing which justified the decision they suggest

to write the Q-PRO-4 version as a contingency plan.

I

also reject the suggestion that HMRI did so and

specifically find that, from its inception Imbert and Stenabaugh

intended to replace the Prism software with their own product.

The secrecy surrounding the project was designed to deceive Prism

and to allow them to develop the Q-PRO-4 version while continuing

to utilize the fruits of their contract with Prism. The fact that

the veil of secrecy extended to include the Executive Committee

and the Board of Directors speaks volumes about the conduct of

Stenabaugh, Imbert and King.

- 168 -

The management difficulties at HMRI were such that no overall

corporate purpose sanctioned by the senior committees may have

emerged but the project was certainly sanctioned by Mr. Stenabaugh

and by Mr. Colman.

B.

COPYING

The two broad issues arising under this head are whether

Stephen Imbert in writing the Q-PRO-4 version of the software made

use of Prism's source code and associated materials and whether if

Imbert copied, generally the structure, sequence and organization

of Prism Savoir-Faire a claim of copyright infringement is

maintainable.

While HMRI disputes Prism's claim to any

proprietary rights in the screens, the "look and feel," and the

"user interface" of Savoir-Faire it acknowledges Prism's ownership

in the source and object code of the software.

HMRI also maintains its denial that Imbert copied the Prism

software but submits that, in any event, the Q-PRO-4 version is a

modification or merger of the licensed programs into an "updated

work" within the meaning of Article 8 of the Licence Agreement.

i)

DIRECT USE AND COPYING OF THE PRISM SOURCE CODE

This first issue is framed by the parties in the

Statement of Facts and Issues in p. 68:

68.

Prism says that its

source code and associated materials were

available to Imbert and were used by him in

- 169 -

the preparation of QPRO 4 Savoir Faire in

breach of the contract and Prism's copyright.

HMRI admits that Imbert could have had access

to the Prism source code if he had wished, but

says that he did not in fact use it except to

obtain one public domain formula.

In any

event, HMRI says that it is legally irrelevant

whether Imbert had access to or used Prism's

source code, as any use would have been

authorized under Article 8 of the Licence

Agreement.

In beginning the consideration of this aspect of this action

two things must be recognized at the outset, first that both

programs are designed in the broadest sense to do the same thing

namely collect the data required by HMRI, format it in a manner

specified by them and transmit it for entry into HMRI's mainframe

computer, secondly, that Q-PRO-4 is classified as a "fourth

generation language" whereas BASIC is

a

"third generation

procedural language." The specific significance of this

distinction will be touched on later in these reasons but at this

point it is sufficient to say that the use of a fourth generation

language (4GL) will, itself, introduce some differences into the

programs because of the inherent structural differences between

third and fourth generation computer languages.

(1) THE AVAILABILITY OF PRISM'S SOURCE CODE

The defence in this action admits that Mr. Imbert had

access to Prism's source code and that he looked at it on one

occasion to obtain what he described as a "public domain formula

- 170 -

for a check digit." The defence has also admitted that Prism and

Q-PRO-4 source code were on the same computer at the same time

although they maintained this was only on Wendy King's computer at

HMRI's offices.

Mr. Imbert had the Prism executable software installed on the

computer he had at his home on which he did the primary

development work for the Q-PRO-4 software.

He denied, however,

that he had on that computer Prism's source code.

I have already concluded earlier in these reasons that the

University Hospital Computer was used by Mr. Imbert and Ms. King

in the development of the Q-PRO-4 software and indicated my

findings about their credibility. It is clear from the evidence

of John Jeffery which I have accepted that the University Hospital

Computer had installed on its hard disk as late as January 23,

1987 (the date of the type backup D182) the source code and the

executable code for Prism Savoir-Faire as well as the Q-PRO-4

development system.

On the balance of probabilities I find that this was the

computer Mr. Imbert used at his home to develop the Q-PRO-4

software and that, contrary to his sworn evidence, he had

installed on it the Prism source code for Version 86.6.

- 171 -

In reaching these conclusions on the whole of the evidence I

am also mindful of yet another series of variations in Mr.

Imbert's evidence between his discovery and that given at trial.

His sudden recollection at trial that the computer he had at home

had a President label on the keyboard with the not so subtle

implication that it must therefor have been a President computer

was tailored to walk the narrow line between his sworn testimony

on discovery where he testified that the computer at his home had

no name on it and he didn't know what make it was and the

perceived danger of the finding that the computer used was the

University Hospital machine. The fact that his initial answer at

trial that there was a President label on the keyboard came in

answer to a question in chief about the size of the hard disk is

also revealing.

His cross-examination in this area was an

intriguing mix of skilled cross-examination and artful and

intelligent dodging mixed, in my view, with careful misdirection.

His suggestion that HMRI provided him with a computer to be

used at home in the development of this software flies in the face

of the equipment shortages which were an ongoing problem at HMRI

and more significantly his own written complaints about that

situation. On March 16, 1987 he wrote to Mr. Alex Adams and at

one point where he addressed the required resources he recorded

the following:

The hardware at HMRI is woefully inadequate.

I put in a request for two microcomputers last

October and had to wait until March for

- 172 -

delivery of one.

For the last three weeks,

Wendy (who is writing a manual on a PC) and I

have had to share one machine (less than 6

weeks before deadline). As outlined above, to

convert data from the Prism version to our own

will require extensive conversion of very

large data files.

In the recent Scarborough

General fiasco I was simply unable to do

anything to their history file because I did

not have a machine big enough to work on it.

It is not just difficult to reconcile this with Mr. Imbert's

evidence it is impossible, even considering Mr. Imbert's

suggestion in cross-examination that in his letter he was

stretching the truth. I have no doubt at all that Mr. Imbert was

stretching the truth in his testimony.

If he had requested two

microcomputers in October 1986 (at the time the University

hospital computer arrived) and had to wait until March for the

delivery of one, where did the computer he was using at home come

from?

The answer is, in my view, obvious, Mr. Imbert had

Prism's computer intended for the University

appropriated

Hospital, and contrary to his evidence he did have Prism source

code on the computer on which he did the development. Once again

Mr. Imbert was undone and contradicted by a document he had either

forgotten or was unaware of at the time he embarked on his

"story".

(2) MR. JEFFERY'S "FINGERPRINTS"

Mr. Jeffery during the course of his evidence testified

about what he described as a series of "fingerprints" which he

- 173 -

identified during lengthy and detailed examination of versions of

the two programs.

These demonstrate, according to Mr. Jeffery

actual copying of the Prism source code.

Mr. Jeffery began his evidence by referring to a series of

pages which were printed segments of Prism source code which had

been annotated by hand.

Some of these handwritten annotations

were in Mr. Jeffery's handwriting while others were not.

The defendants submit that only a very few of the total

number of pages, 24 pages out of some 6 to 8 inches of paper had

been annotated and that many of these were in Mr. Jeffery's

handwriting or had been done by those in attendance at the

training course Mr. Jeffery gave in October 1986. What underlies

this submission is that the entries were innocently made for

legitimate purposes during the course of a training session

designed to assist HMRI's take over of maintenance of the Prism

software.

I have already concluded earlier in these reasons that this

course was staged and arranged for the express purpose of

assisting Mr. Imbert in his understanding of the Prism program and

to advance his writing of the Q-PRO-4 version. In light of that

finding it is of some significance that one of the areas of source

code specifically annotated by hand (Exhibit T21-2 page 7) is the

- 174 -

portion of the subroutine which validates the "check digit on the

Ontario Health Insurance Plan number."

The defendants in their

defence of this action and Mr. Imbert during his cross-examination

both admitted that Mr. Imbert had obtained this particular formula

from the Prism source code.

This, in my respectful view, adds further support to the

finding I have already made as to the real purpose of the

"training course."

The next portion of Mr. Jeffery's testimony dealt with the

documents admitted as Exhibit T21-3.

These two pages had been

produced by the defendants and in the original form were a

computer printout on fanfold paper, of portions of Prism's source

code relating to Screen 3 of the Prism programs. The printout was

made in two stages the first is dated Friday September 19, 1986

and the second page Sunday, September 21, 1986. The printout was

made by means of the Norton Utility LP COM by someone familiar

with that utility. Mr. Imbert was both familiar with the product

and had it available to him when he arrived at HMRI. I conclude

that these printouts were made by Mr. Imbert from the Prism source

code shortly after his arrival at HMRI.

Despite the inherent differences between third and fourth

generation computer languages when one descends below the level of

- 175 -

function which is characteristic of 4GL's the code at the

procedural level is similar.

On this both experts, Professor

Dexter, for the plaintiff's, and Mr. Van Toen, for the defendant's

agreed with the opinion expressed by Mr. Jeffery that Q-PRO-4 and

Basic are comparable at the level of procedural code.

Mr. Jeffery testified that when he set out to demonstrate

that Mr. Imbert had actually copied the Prism software his task

was made difficult not just by the differences between the

computer languages but also because of the "transparency" of the

Prism software. In this he was referring to the fact that much of

the functionality of the Prism software can be learned by studying

the screens.

The approach taken by Mr. Jeffery is best taken directly from

his words:

QNow, Mr. Jeffery, before we go into the

details, could you please describe

in a general fashion for the Court

your approach to trying to identify

what we are calling fingerprints of

code which appears in both Q-PRO-4

and Prism systems?

AIt created quite a problem for me. The

Prism programs and the SUPERMACS

and Savoir-Faire were not designed

with protection in mind, they were

designed

with

functionality

in

mind, and consequently they are

fairly open, and by utilizing the

screens and the prompts and error

messages as they exist within the

Prism system, the underlying code

can be written in another language

- 176 -

fairly easily. In fact, in effect,

the Prism programs are transparent

in that the functionality can be

obtained very easily by looking at

the screens.

What I had to

look for to determine and to show

the Court that access had been made

to our source code, I had to look

for anomalies, I had to look for

data that is shown on the screen

that is not stored in the record or

is stored in a different manner. I

had to show errors that might have

occurred in the Prism system that

reproduced themselves in the -- or

appeared to be reproduced in the Q-

PRO-4

essentially for those odd things,

not the overall code, simply

code.

I

had

to

look

because the defence is that they

copied the screens and the user

interface, namely the messages and

error messages and prompts. For me

to find those differences was quite

a battle, but we did find some, and

we would like to show them to the

Court today.

- 177 -

(a) UNKNOWN BIRTHDATE

At the heart of Mr. Jeffery's evidence on this topic is

the location during his examination of the Prism source code of a

coding error he made during the writing of Prism Savoir-Faire.

Whenever a known birthdate for a patient is entered into the Prism

system, it is validated against certain allowed values.

If the

person has entered a date in the future, ie. after the date on

which the data is being entered an error message is displayed.

This is not surprising in light of the simple fact that it is

unlikely that any patient being treated at the hospital has not

yet been born! If the full birthdate of the patient is not known

a different routine is entered by indicating on one of the data

entry screens that the full birthdate is unknown. The coder can

then enter a partial birthdate, eg. month and year.

As a result of a programming error at this point the coder

can enter a partial birthdate in the future.

In his writing of

this portion of the program Mr. Jeffery, in error set the

validation parameters as 1800 to 2099 which would allow the entry

of any year within that range as the year of birth.

The Q-PRO-4 version of Savoir-Faire follows exactly the same

procedure for checking partial birthdates and will allow the entry

of an unknown birthdate in the future in this section of the

program.

The validation parameters in the Q-Pro-4 program are,

- 178 -

however, different. They allow the entry of a year from 1830 to

2000 as the date of birth.

This would appear to be a considerable coincidence even given

the change in the range of years, if this evidence stood alone.

It does not, in his cross-examination on this matter Mr. Imbert

was asked whether or not it was a "bug" his reply was:

No, I don't consider it a bug. The area, what

you're talking about is, when you -- there's a

field on the abstract called birth date

unknown.

You know, if you say -- there's a

field on -- in Savoir-Faire called birth date

unknown. If you say, yes, you don't know what

the birth date is, it will let you put in an

approximation of what you think it is.

For

instance, if you know the year, you put the

year in. If you know the year and the month,

you put the year and the month in.

Mr. Imbert's cross-examination on this point got more

specific, however, his answers did not:

QSo, Mr. Imbert, is it your evidence that

-- take if from me the Prism system

is the same -- that if both the

Prism and the Q-Pro-4 system will

allow a birth date in the future if

and only if it is an unknown birth

date, i.e., not a complete birth

date, that that's a coincidence

that the two systems are the same?

ANo, because generally date-checking as a

programming practice is a pretty

common thing.

Now,

I

haven't

looked at the Prism source code,

but I imagine that Prism has a --

some kind of procedure that checks

out dates because the way -- the

way you program in 3GL, generally

is you would write a sub program

- 179 -

that would check a complete date,

okay? Now, date-checking is pretty

simple. You check the year -- now,

checking the year is -- you can't

really check the year.

The year

can be anything. You can check the

month, it has to be one to twelve,

and you can check the day, which

asks that -- you know, you put in a

table for the maximum amount of

days for each month.

So they are

checkable. But I put to you that

typically that -- when you are

using date-checking, you would use

this -- you would reuse the same

procedure.

Now, in this --

in this case, you've got a strange

situation because you say birth

dates unknown, so you're letting

the -- by definition, the birth

date is not going to be complete,

so you can't run it through a

procedure that's going to do a

completeness check on it because by

definition it's going to be an

incomplete

check

--

incomplete

date. The point being is that the

way -- the way that was developed

in my program was that the only

thing I checked was -- the way I

checked it, and

I

have looked

through it, is the logic in that

area of the program was for -- I

parsed out the date into year,

month, and day.

could do a check on the month,

whether it's reasonable, it's

Now, if the -- I

either one to twelve, I could do a

check on the day, but I couldn't do

a check on the year, then I did an

off-the-wall type of thing, sort of

a reasonableness check, which was -

- I think I based it on --

Q1830 to year 2000.

AYeah, okay. And that's the sum total of

the checking that I did on it.

Now, admittedly, there are problems

- 180 -

with it, I'm not testifying that I

ever wrote a procedure to make it

airtight, but the philosophy was

that if you -- if the coder says,

"I don't know what the date is,"

I'm not going to do a lot of

checking.

Now,

I

would

like to point out too that it's a

very messy area of the code, and

once again, I haven't seen a logic

flow chart to show how the coding

is

the

same

between

the

two

systems.

QDo you dispute that it's the same?

AI dispute that it is the same, yes.

Mr. Imbert's "explanation" is no explanation at all. There

is no reason advanced why the year of birth logically could not be

tested against the current year to insure that it at least is not

in the future, yet this was not done.

Instead it is tested

against a range of years as did the Prism system thus allowing in

error a future year of birth.

In their written submissions the defendants say in part that:

. . . it is submitted that it is a reasonable

expedient merely to check against a fixed

range of years, including a few years into the

future so that programming will not have to be

amended as time goes on.

If Mr. Imbert had

copied Prism's work, there would have been no

reason for Mr. Imbert to use a different range

of dates than those of Mr. Jeffery.

One explanation for this change does logically arise and that

- 181 -

is that someone copying a program they didn't entirely understand

wished to make superficial changes to give their work the

"appearance" of being different without the reality of difference.

In other words a deliberate attempt to conceal the source of

their "work".

(b) MONTH END : PROCEDURE SCAN

The second of Mr. Jeffery's "fingerprints" is the way in

which the month end procedure checks the medical procedure

information to determine if it has to be converted for submission

to the HMRI mainframe. When Mr. Jeffery wrote the original MACS

software for the OSI computer he wrote the program to test the

complete procedure line which included the date, the procedure

code, the doctor code and each of the other entries to determine

if that line was blank.

converted for submission to HMRI.

testified unnecessary and time consuming.

If the line was blank it was not then

This was, as Mr. Jeffery

It was only necessary

to check the procedure code itself to determine if the line was

blank. This would have been a much shorter and more logical test.

In this specific area of the program, the Q-PRO-4 version

also checks the complete line. Mr. Imbert's explanation:

. . . you have to realize, at the time, this

was a very complicated area of the abstract,

and the safest way, was to check for the whole

line.

- 182 -

The contrast can be seen in the way in which Prism Savoir-

Faire handles therapeutic abortion procedures.

In this case the

Prism program scans only the procedure code instead of the whole

line to determine if it needs to enter the abortion portion of the

screen and exits upon encountering either a blank procedure code

or an abortion procedure. Similarly the Q-PRO-4 program in this

part of the program does exactly the same thing checking only the

procedure code instead of the whole line as it does in the other

portion of the program. Mr. Imbert's explanation is that this is

a coincidence.

In this case as well Mr. Jeffery identified on page 12 of

exhibit T21-2 the handwritten note "Abortion Codes" in Mr.

Imbert's handwriting opposite that portion of the printed source

code.

The defence submits that Mr. Imbert's reason for checking the

entire line in one case and only the procedure code in the other

is reasonable in that in the month end procedure it is necessary

to insure that all information is forwarded to HMRI whereas in the

other case the test is only to see if an abortion procedure has

taken place.

(c) MONTH END : OPERATING ROOM DEATHS

In this procedure Prism Savoir-Faire records information

- 183 -

concerning a surgical procedure which results in death.

HMRI

requires the time of death to be recorded in one of four mutually

exclusive categories. These categories are (1) in the operating

room, (2) less than 48 hours after the operation (3) between 48

hours and 10 days post operative, and (4) more than 10 days. On

the paper abstract (Form 780) this data was recorded by marking in

one of four "mark sense" fields.

Prism's solution to this data

requirement was to record the information by entering a single one

of four alpha characters each character represented one of the

four mutually exclusive fields.

The second step is to reformat this data to meet HMRI's data

specifications which required it to appear as four separate mark

sense fields when submitted with one of the four completed.

Prism's solution to this was to use a "mask" technique which took

the single alpha character and converted it to four fields with

the appropriate one marked. The same mask technique was used by

Prism to handle the type of death fields.

Although this masking technique cannot be seen on any screen,

the Q-PRO-4 version uses the same "mask" technique.

The defendants submit that this particular technique is one

of a limited number of options which are dictated by HMRI's

specification of the way in which data must be submitted to their

- 184 -

mainframe. What it demonstrates according to the defence is that

the detail of the HMRI specification dictates a limited number of

options as to how the programmer will handle creating the

necessary output.

(d) ICD-9 PREFIXES

In the case of this particular "fingerprint" there are a

number of ramifications which go well outside the scope of simply

a question of whether Mr. Imbert copied some portion of Prism's

program and, contrary to his original sworn testimony and the

position originally taken by the defence, resorted to Prism's

source code to do so. The other issues touched on are again Mr.

Imbert's credibility and the extent to which he was prepared to go

to conceal his activities and then when caught to shade or change

evidence to minimize or eliminate the damage.

The ICD-9 codes are an international classification of

disease listing published by the World Health Organization. Prism

in adapting the codes for its medical abstracting software

truncated these codes to fit them within a 59 character field. On

the evidence it is clear that both Prism and HMRI worked on

creating the eventual list of truncations used.

When Mr. Trueit designed the original record layout for the

ICD-9 file he chose to use prefixes to identify specific groupings

- 185 -

of codes. He did this because in the Prism system all the codes

were stored within a single file. The lead prefixes he utilized

were "D" for diagnoses, "P" for procedures, and for contributing

factors which already had a V prefix Mr. Trueit for some reason

added a V creating a prefix "VV".

The external cause codes

already had an "E" prefix and, in this case Mr. Trueit did not add

another but left them with their original "E" prefix.

The Q-PRO-4 program copied this use of prefixes exactly

despite the fact it maintained the procedure codes in a separate

file and, as a result, did not need to utilize a differentiating

prefix.

During the course of his discovery evidence Mr. Imbert

described obtaining the ICD-9 codes from Lynne Abbott on a nine

track tape. The sequence of questions is again revealing:

2925 QYes.

And do you recall

whether or not the Prism

software

containing various ICD-9

codes?

had file

a

AI

do.

Yes.

They did have a

file

containing

ICD-9

codes.

2926

QAnd

did

it also have a file or

the portion of the

software that contained

descriptions of diagnoses

in the English language?

AYes,

it

- 186 -

did.

2927

QAnd

did

you adopt those diagnoses

numbers and descriptions

of

diagnoses

in

the

English language in this

version of your software?

AOkay.

I

-- from the best of my

knowledge, I received a

9-track tape from HMRI as

input

for

the

ICD-9

codes.

Where this 9-

track tape came from, it

was

labelled,

it

was

labelled as long -- long

codes, it had on it.

That's all I remembered.

It was given to me by

Lynne Abbott.

2928

2929

QAn

8-

track tape?

tape.

AA 9-track

QI'm

sorry.

Was

it

your

understanding that that

9-track

tape

contained

code numbers?

AYes.

2930

QAnd

did

it

also

contain

the

descriptions

in

English language of the

diagnoses

relating

to

code

those

particular

numbers?

AYes,

it

did.

- 187 -

When these questions were put to Mr. Imbert the following

exchange took place at trial:

Mr.

Imbert,

were

you

asked

those

questions and did you give those

answers?

AYes, I did.

QWere those answers true?

AThose answers were true. I did -- I did

get

a

tape from the mainframe

department containing ICD-9 codes

and English language descriptions,

yes.

QWas your answer yesterday in examination

that you used ICD-9 codes supplied

by Prism also true?

AYes, I did -- I used them in the demo

version of the Alberta product.

QIs it your evidence than that in

discovery,

you

were

being

economical with the truth?

AI don't think I was being economical

with the truth. As I said, and to

best of my knowledge, that's what I

remember.

I remember attempting

getting them from HMRI.

At trial Mr. Imbert admitted that he had used the ICD-9 codes

from the Prism system in the Alberta demonstration version of Q-

PRO-4 Savoir-Faire.

In fact he gave this evidence during his

examination in chief despite the position taken by the defence at

this trial as set forth in p. 68 of the Statement of Facts and

Issues:

HMRI admits that Imbert could have had access

to the Prism Source code if he had wished, but

- 188 -

says that he did not in fact use it except to

obtain one public domain formula.

Why the apparent change of heart? During the course of his

evidence at trial, some two months before Mr. Imbert's appearance

in the witness stand, Mr. Jeffery identified and placed in

evidence a print out of a Q-PRO-4 program ICD9IN.PIC which he had

retrieved by undeleting files contained on a diskette produced by

the defendants.

The sole purpose of this program, which was written by Mr.

Imbert, was to open and read the sequential basic file ANCILL.MRA

which contained Prism's ICD-9 codes and copy them out into an

indexed Q-PRO-4 file.

The fact that this program had been discovered and placed in

evidence was, in my view, known to Mr. Imbert when he took the

stand.

He gave no credible explanation for the apparent

What

difference in his evidence at discovery and at trial.

emerges from this is the conclusion that Mr. Imbert is capable of

telling the truth, but he will do so when it suits his purposes

and he will do so when the circumstances are such as to leave him

no alternative. I have no faith that he will do so in any other

circumstances.

The issue raised by this particular evidence is much more

- 189 -

serious than any debate about whether or not Prism's work with the

ICD-9 codes represented sufficient originality of expression to

justify the extension of copyright protection with respect to

them.

I have already given the conclusions I have reached with

This evidence also

respect to Mr. Imbert's credibility.

demonstrates that Mr. Imbert and through him HMRI were prepared

during the course of their rewrite of the software to take

positive steps to conceal their activity and to attempt to hide or

destroy evidence. This was so in the case of the formatting and

wiping of the hard disk drive on the University Hospital computer

and it was equally so with the failure to produce evidence and

"documents" in this action as evidenced by their failure to

knowingly disclose the existence of ICD9IN.PIC.

The way in which Mr. Imbert gave his evidence on discovery

was the furthest thing from honest and forthright testimony; it

was evidence given by an individual who was comfortable that his

steps had been covered and his answers were calculated to deceive

rather than to reveal. This evidence, in the view I take of this

action, has far reaching implications for the case advanced by the

defendants.

The ICD-9 codes which were utilized by the Savoir-Faire

systems numbered approximately 10,000.

more than half or some 5,000 codes.

Of these Prism truncated

Of these HMRI changed some

- 190 -

200 of Prism's codes.

Prism.

The majority of this work was that of

Describing this work as "truncation" serves to focus

attention on what was removed rather than on what remained after

the truncation. The object of this exercise was twofold, firstly

to fit the codes within the 59 character field provided in the

program and secondly to leave a truncation that was identifiable

to persons using the system. The question is not whether Prism by

its efforts acquired any rights with respect to the ICD-9 codes

themselves, but whether in their selection of truncations and

changes to the codes to fit them within the field provided they

provided elements of originality and expression. In my view these

efforts are capable of satisfying the necessary tests.

(e) PATIENT SERVICE CONFLICTS WITH SEX

In this case Mr. Jeffery again found a programming error

he had made in his coding of the Prism software. When data was

entered into the Prism abstracting system cross field edits tested

the validity of the data.

One of these edits was supposed to

ensure that a medical procedure that was specific to patients of a

particular sex, the patients sex was appropriate. As an example

the system was supposed to check to ensure that where obstetrical

services were performed the patient was not a male.

Through a programming error, although this particular error

message was included in Prism's REEDIT program, it could never

- 191 -

appear on a Prism Savoir-Faire screen while the program was in

operation. This error message did not appear in any user manual

or in any other documentation for Prism Savoir-Faire yet it

appeared at the end of a list of error messages for the Alberta

version of Q-PRO-4 Savoir-Faire dated November 25, 1986.

This

particular error message was then apparently eliminated as it

never appeared in a production version of the Q-PRO-4 programs.

The defence submits that this particular error message was

required for Alberta and that it was, according to Ms. King taken

from her abstracting manual and given to Mr. Imbert. What these

submissions fail to address is the fact that the error message

delivered by Ms. King to Mr. Imbert, if this occurred, read "PAT

SERVE.OBS, BUT SEX MALE", and it does not explain how the early

version of the Q-PRO-4 software came to have an error message

which reads "SEX CONFLICTS WITH PATIENT SERVICES" the identical

wording chosen by Mr. Jeffery and contained within the Prism

program.

- 192 -

(f) DECEASED PRELOAD

Programming is

a

process which is not strictly

mechanical or even substantially so. It is a process which is in

many respects an "art form" in which an individual programmer is

free to make choices as to the way in which his program will

accomplish certain tasks. While the language utilized dictates a

good many features of the resulting program there is ample room

for the "expression" or "quirks" of the particular programmer.

In this section of the Prism software a somewhat difficult

process is followed. Through a process described as "preloading"

a skeleton record is created when a patient is discharged, signed

out or dies. Initially certain basic information is collected and

the full abstract is completed later when the information is fully

available.

The fifth field on the preload screen of the Prism

When an operator

system asks whether the patient is deceased.

answers yes to that question by entering a "Y" the Prism system

instead of storing that specific answer flips the "Y" answer to a

"N" or no response and stores it in the space reserved for the

"EXIT ALIVE" response in field 14 on Screen 2. This indicates, in

the form of data required by HMRI that the patient did not leave

the hospital alive, a somewhat anomolous method of recording the

fact that they died.

During the abstracting process the Prism

software checks the "exit alive" field and if it locates an "N"

entry it converts it to a blank which is the correct "exit alive"

- 193 -

code specified by HMRI for a deceased patient.

In its cross field edits the Prism system checks this field

for an "N" entry when the coder leaves the abstract and generates

an error message if it encounters one.

In the Q-PRO-4 system the data is collected in the same way

with the answer to the initial question of whether or not the

patient is deceased being stored in the "exit alive" field. The

difference is that the "Y" response is not flipped to an "N" when

it is stored there. As a result the resulting "Y" entry entered

by Mr. Imbert's system indicates that the deceased patient left

the hospital alive. A response which is wrong.

The Q-PRO-4 system also utilizes the same crossfield edit as

the Prism system checking for an "N" entry in the "exit alive"

field despite the fact it can only occur if the coder enters a "N"

response to the first question indicating that the patient is not

deceased. The Q-PRO-4 system then records the "N" response in the

exit alive field indicating the patient died.

This response is

again incorrect. Under this system abstracts would be generated

with a "Y" entry in the exit alive field where HMRI only permitted

three entries "D" meaning discharged, "S" meaning signed out or a

"blank" indicating the patient was deceased.

- 194 -

The plaintiff's submit that the only reasonable inference to

be drawn from this is that Mr. Imbert copied the Prism system

without completely understanding the way in which it worked. In

his cross-examination on this seemingly simple issue Mr. Imbert

maintained that when a coder entered "Y" to the question of

whether a patient was deceased and his program stored the "Y" as

entered in the exit alive field indicting that the patient left

the hospital alive that this was ". . . the way I wanted it to

be."

Mr. Imbert also maintained that this was not a "bug" or an

anomoly in his system despite the fact the users of the system

reported at the Ontario Users group meeting on November 10, 1988

that it was and that after entering the information on the preload

screen they had to change it on the abstracting screen.

Mr. Fisk on behalf of the defendants submits that the absence

of the "flip" in the Q-PRO-4 program makes ". . . it . . . clear

from this fact alone that there is no copying."

I regret to say I do not agree.

If Mr. Imbert wrote this

section of his program without reference to the Prism software and

with the assistance of Wendy King and in full knowledge of the

HMRI requirements it is, in my respectful view, unlikely that his

"logic" would so fail him.

It is more likely that the observed

- 195 -

results arose from a failed attempt to copy than from a unique

approach adopted by Mr. Imbert in a "clean room" without reference

to the Prism software.

In any event a single difference in the logic which is

otherwise identical does not in any event amount to proof that

there was no copying.

On the logic represented by Mr. Imbert's

"solution" to this part of the program there is no explanation

whatsoever for the crossfield edit which checks for an "N" entry

which will not occur.

(g) SERVICE TRANSFERS

This area of the Prism program is perhaps the most

unique and anomolous of those identified by Mr. Jeffery in his

"fingerprint" analysis. The reason for this is to be found in the

reasons which underlie Prism's "solution" to this particular

programming challenge.

On the original manual abstracting FORM 780, space was

provided to record up to three "service transfers."

A service

transfer occurs when a patient is moved from the hospital for a

procedure or test then returned to the hospital. The actual way

in which the manual abstract form is used provides for four data

entry fields for each service transfer, these are (1) the actual

patient service, (2)

the service transfer days (3) the service

- 196 -

transfer doctors, and (4) the service transfer diagnoses.

While

the first two of these have their own fields on the manual form

the last two share fields for regular doctors and diagnosis on

that form. In other words the last three fields provided on the

manual abstract form are used potentially for two different

purposes. Where there are no service transfers they are used to

record regular treating doctors and diagnoses, but, where there is

a service transfer they are used to record the service transfer

doctor and diagnoses.

Mr. Jeffery when he originally wrote the SuperMACS system for

Vancouver General Hospital misunderstood the data to be collected

with respect to service transfer and thought that the only data to

be collected was the transfer services, days and doctors.

He

created a three field format which provided space for these

fields.

When he discovered some six months later that service

transfer diagnoses were required as well there was no space left

in this record layout to store this additional date. His solution

was unique to the situation which faced him, with all the power

and flexibility of the computer at his disposal he adopted the old

manual paper abstract solution of using certain fields for two

purposes as needed.

Prism Savoir-Faire.

This anomolous solution continued through

Although the Prism Service Transfer Screen

- 197 -

allowed a user to enter the required data the service transfer

diagnoses were actually stored by the program in the last three

positions of the regular diagnoses field on the previous screen.

(Screen 3).

Prism's system also attempted to make the maximum

room available for normal diagnosis entries and so when the coder

finished entering a service transfer entry and exited that screen

the program took the data entered and "pushed down" the data

writing it on the lowest of the available fields.

This process

provided the maximum available space for normal diagnoses.

The Q-PRO-4 system duplicates Prism's double use of the

diagnoses fields but like the Prism system and unlike the paper

abstract does not do so with the service transfer doctor field.

It does, however, handle the push down feature differently than

the Prism system pushing down each entry as it is made to the last

available line.

Once again Mr. Imbert's evidence during cross-examination is

instructive, he testified that he originally provided a separate

data field to store service transfer diagnoses but that he later

removed it and then made double use of the same area as Prism did

because that was how the Prism system worked.

Mr. Imbert's evidence on this point is not believable.

It

distorts both coincidence and logic in an attempt to force his

- 198 -

actions into some innocent mold. The differences in the way in

which the Q-PRO-4 push down feature works are cosmetic changes

made to give the appearance of difference without its reality.

ii) THE EXPERT EVIDENCE

(1) PROFESSOR DEXTER

Dr. Albert S. Dexter is an Associate professor in the

Faculty of Commerce and Business Administration at the University

of British Columbia.

Bachelor of Arts degree, with honours from the University of

California Santa Barbara, in 1962, Master of Business

His academic qualifications include a

a

Administration from Harvard University in 1966 and a PhD from

Columbia University in 1970. Aside from numerous academic awards,

teaching positions, and professional positions he had written

extensively and has twenty-five years of experience in the

analysis, design, programming and implementation of computer based

information systems. This includes both academic and professional

positions and a broad experience with many computer languages

including both third and fourth generation languages. He has also

been involved in research and the publishing of papers on the

reverse engineering of computer software. Reverse engineering he

defined as a process of engineering backwards from the source code

to reconstruct the design functions on which the code was

originally based.

- 199 -

Professor Dexter was eminently qualified to undertake the

assignment he was asked to pursue. That particular assignment is

best set out in his words:

I

have

been

requested to provide an independent opinion

concerning the similarities and differences

between

versions

of

software

and

their

supporting material used for the purpose of

medical records abstracting and reporting.

One version of the software is entitled "Prism

version 86.6", (hereafter called Prism 86.6

and is written in the Microsoft QUICKBASIC

language (hereafter called BASIC). The other

software is entitled "HMRI Ontario Savoir-

Faire 2.0", (hereafter called HMRI Version

2.0) and is written in the language QPRO4.

Included in the analysis of the HMRI software

and supporting material is an earlier version

entitled

"Savoir-Faire

DEMO",

(hereafter

called HMRI DEMO) also written in QPRO4.

Collectively the two HMRI versions, DEMO and

2.0, are referred to as HMRI software.

Professor Dexter in carrying out his assignment compared the

Prism and Q-PRO-4 systems at all levels from system flow charts,

menu structure, program structure, sequence and organization and

finally at the source code level.

The results of his analysis are succinct and unequivocal:

. . . The similarities are overwhelming at all

levels of analysis. Overall flow charts, the

data, menu screens, program structure, and

even at the level of the code itself, and I

think the interesting issue with the code

itself is that while a fourth generation

language is different from a third generation

language, and BASIC is classified as third

generation and Q-Pro-4 is classified as fourth

generation, but beyond the capabilities of a

fourth generation language generally, and Q-

- 200 -

Pro-4 in specific, you will find that there

are

certain

things

that

the

automatic

procedures cannot accomplish, and when one has

to deal with that language at that level

procedurally, you will find the same type of

procedural coding in the Q-Pro-4 as was

encountered in the BASIC language program in

Savoir-Faire.

Professor Dexter discussed at some length the procedure which

he followed beginning with a general overview of the Prism and Q-

PRO-4 systems from the "function charts" which gave the general

overview he descended to the ever more detailed.

During cross-

examination he was asked if he relied on the function charts in

making his comparison of the Prism Software to the Q-PRO-4

software his answer makes it readily apparent which level of

analysis formed the majority of the evidence on which he relied:

QI see.

You relied on the function

charts when comparing the Prism

software to the HMRI software?

AOnly in the most general sense to grab

an overview of the development of

the system or the presentation of

the systems. In fact, I relied far

more on the detailed code, screens,

the performance, all of the other

things in the report, to make my

comparison. I did not -- I just --

this happens to be one small piece

of the evidence I examined, and,

actually, a very small portion of

the piece, in my opinion. The vast

majority of the evidence that I

have examined has been at the code

level, at the performance level.

So this was general overview in

getting a -- primarily to get a

briefing myself as to what medical

record abstracting was all about.

- 201 -

In chief he testified similarly indicating that what he meant

in his report when he referred to a more detailed examination of

the programs was:

Well, what's meant by "detailed" here is that

as one begins to go into the sub programs, you

begin to examine the way in which each of the

programs are then, in fact, broken down, and

you begin to show the similar types of

similarities;

mostly

similarities,

some

differences. So, for example, on page 19, you

show under here, and I guess this moves to

Appendix 15, but in general, you will see

Prism shows eight different sub pieces of

functionality; the HMRI, both versions now,

show nine.

So primarily large differences

Number 7, the ICD-9

between functionality.

codes, where the Prism version show procedure

codes and ICD-9s together, whereas there's a

breakout of the ICD-9 codes and the procedures

codes on the HMRI versions. But in general,

the structure is very, very similar as you get

down into the detailed breakdown of the

programming.

Professor Dexter in both his report and his viva voce

evidence dealt extensively with the way in which the Prism program

and the Q-PRO-4 programs handled data at the level of the

individual data field.

In the course of this examination he

conducted three detailed sets of comparisons (1) Items developed

by Prism which were not a part of HMRI's specification, (2) Data

which was captured by Prism either differently or in excess of

HMRI's requirements, and (3) computed items which were not

collected from a Prism abstract screen. His conclusion:

. . . both HMRI versions of the software

follow the Prism 86.6 software much more

- 202 -

closely than these versions follow Form 780

and the Manual System.

From an overall

perspective on the data, there are obvious

similarities which can be traced to the

requirements of Form 780, the Manual System,

and to HMRI.

Notwithstanding, there appear

numerous peculiarities of expression in both

the individual field and record structures of

both HMRI versions which seem to have been

substantially carried over from Prism 86.6.

In these latter instances, there are styles of

data expression by Prism which did not

originate with Form 780, the manual record

abstracting or other HMRI requirements. Both

HMRI

versions

appear

to

use

the

same

expressions as Prism 86.6, although the

closest comparisons are made between the DEMO

and Prism 86.6. The similarities exist, even

to the various sub fields of the files.

In

summary, there appear to be substantial

similarities at the data structure levels

among the various fields, records and files

between Prism 86.6, HMRI Version 2.0, and HMRI

DEMO.

The significance of these conclusions is threefold the

similarities extend to the data structure levels, many of them are

not dictated by or traceable to the Form 780 and the similarities

extend to areas outside of the requirements for the submission of

data to HMRI. In other words, the similarities found by Professor

Dexter are not explainable by the copying of FORM 780 or by HMRI's

data specifications.

During the course of his examination of the Prism code

Professor Dexter located a particular error in the programming

logic that is, in my view, of some significance.

In the

collection of data with respect to operating room deaths it is

- 203 -

necessary to collect the information, as previously described in a

number of categories: (1) in the operating room, (2) less than 48

hours after the operation, (3) between 48 hours and 10 days post

operative, and (4) more than 10 days. In reducing this to program

code and the logic necessary to collect it Prism made an error,

the Prism code and the program does not collect data with respect

to any deaths which occur on the 10th day after the operation.

The categories established in the Prism program are less than 10

days and more than 10 days thus excluding the tenth day.

Professor Dexter identified exactly the same logical error in

the Q-PRO-4 programs he examined.

With respect to the Q-PRO-4 programs duplication of Prism's

unusual and strange way of handling service transfer diagnoses

Professor Dexter in his report described it this way:

. . . The most peculiar portion of this screen

is something that also appears at the deep

code level. The information having to do with

service transfer diagnoses is most peculiar,

and frankly is not explainable by any logical

means that a programmer would follow.

I

understand that originally a Prism programmer

had made a mistake in his design of this

screen and had not provided enough room on the

screen or in the record to include this

transfer information.

As a result the Prism

programmer stored this data in the last three

positions of the diagnoses area or Prism

Screen 3 in fields 59, 51 and 53.

This

mistake would have led to a great deal of

coding and effort to reposition the items and

programs around this error.

It is my

understanding that this error was made early

in Prism's definitions of screens, even before

- 204 -

their entire relationship with HMRI began.

I

can think of no other reasonable explanation

as to why a programmer would have programmed

a system in the convoluted manner I observed.

Astonishingly, the HMRI programs duplicate

this entire convoluted and illogical process!

This evidence and opinion is of some importance, the

structure and coding of this portion of Prism's program was

explainable only in terms of the factual situation in which it was

created. They had misunderstood the data to be collected and had

completed their program based on that understanding.

After the

fact, they sought a solution to the problem with a defined data

structure that had no room to store additional data fields. Their

solution was the one described to make double use of some fields

and to move the data down.

There is no logical explanation for a different programmer

not faced with those peculiar circumstances utilizing this

"solution", it is inefficient, it potentially reduces space for

data storage in another area of the program and it involves a

great deal of code and work to allow the double use of these

fields for no reason other than that faced by Prism originally.

The explanation offered by Mr. Imbert is neither logical nor

sensible. It borders on the preposterous. The irresistible force

of this evidence, taken on its own, is that this section of the

Prism code was copied by Mr. Imbert in his creation of the Q-PRO-4

- 205 -

versions examined by Professor Dexter.

definitely does not stand alone.

This evidence most

Professor Dexter in his examination of the software went

through a process whereby the entry of data into the systems he

tested and compared the operation of the programs and the way in

which they processed data observing both similarities and

differences:

. . . The net result of that is they're

primarily very similar, a few differences, and

I have outlined those differences in the

report.

But

the

main

difference between this versus the appearance

just of the screens with no data on it is the

fact that all of the performance of the

processing underneath the screen is supported

by code which essentially means that the same

things are going on at the same time, the same

logical sequences, and so forth, the same

types of edits are going on as was for all

three systems.

The Q-PRO-4 programs at the surface level have made changes,

notably by use of the function keys found on IBM or IBM compatible

keyboards. Professor Dexter viewed the differences he found to be

mostly "cosmetic" and concluded that ". . . the behaviour was so

close that you had to work very hard to find differences. . ."

Professor Dexter examined in detail five programs within the

Prism and Q-PRO-4 systems and having considered the expected

differences between a third and fourth generation language he

- 206 -

concluded that in terms of the structure, the sequence of

execution and their organization the Prism and the Q-Pro-4

programs he examined were "very, very similar."

This similarity included not simply the broader structure of

the programs but extended to include the individual line by line

code comparison where ordinarily the personality and quirks of the

individual programmer would prevail.

The concluding portion of Professor Dexter's evidence places

this in perspective:

QIn terms of the line-by-line command

structure, if you're getting to

where you're going in terms of

either having examined the system's

function and finding that they

behave

the

same

the

way,

is

it

necessary

that

line-by-line

structure of the two systems would

be the same or --

ANo, it could differ. Again, these are

idiosyncrasies that appear at the

programmer level and they could

differ. What you see is that they

don't differ, and that's a bit

surprising, but they could if one

was, in fact, not using a model of

one to write the other; and it's my

belief the reason they're so close

is the fact that one was being used

as a model to write the other one.

We can talk about the differences,

but those are differences that I

would expect just looking at a

third generation versus a fourth

generation language.

QAssume the Q-Pro-4 author, the person

that was writing the system in Q-

- 207 -

Pro-4, sets out to copy generally

the screen structure but otherwise

is writing the software in what I

understand in your world is called

a clean room --

AYes.

Q-- so all he's got is the screens,

nothing else, is the result that

you have found in your analysis --

would you expect it to be similar

as you see here or would you expect

--

AIf this was a clean room environment,

this result is astonishing.

They

are so close that one would have

had to be a wizard to come up with

something that close.

differ and --

They do not

THE COURT:

Just a moment.

I take it your

evidence, if I read into it what I think is

there, Professor Dexter, is that this person

would not only have to be a wizard to do it

but a particularly characteristic wizard to

match --

AWith the same personality, sir.

The critical underpinning of Professor Dexter's evidence lay

in his detailed examination and comparison of the underlying code.

While cross-examination touched on peripheral matters such as the

source of individual documents which Professor Dexter reviewed it

in no way succeeded in shaking either the central features of his

evidence or the opinions he expressed. It is his examination of

the programs and the code itself on which Professor Dexter's

opinions rest.

- 208 -

The defendant's submit that the version of Prism Savoir-Faire

examined by Professor Dexter was different than those in the

possession of HMRI at the time of the rewrite and yet in all the

seemingly endless days of evidence in this trial they neither

demonstrated nor identified any substantive differences.

Professor Dexter's evidence and his opinions, in my

respectful view emerged from cross-examination not only intact but

virtually unscathed.

(2) MR. VAN TOEN

Gordon Van Toen is a management consultant with Holocene

Management Consulting Limited, a company providing ". . .

information technology related management consulting services to

the public and private sectors." His academic credentials consist

of a Bachelor of Science degree with majors in Mathematics and

Physics from Sir George Williams (now Concordia) University in

1962. He has extensive work experience in the area of computers

and information management including lengthy periods of employment

with the Ministry of Health and other Provincial government

departments in Ontario.

Mr. Van Toen's report despite it's structure lacks the depth

of analysis found within Professor Dexter's report. It frequently

contains the assertion of opinions without either supporting

- 209 -

analysis or identification of underlying facts on which the

opinions are based.

In other instances underlying assumptions were inconsistent

with the evidence led at trial.

I have already touched on a

number of areas of Mr. Van Toen's report within these reasons

where they bear on other specific issues.

Of even more concern, however, was Mr. Van Toen's demeanour

and attitude during his evidence.

His reluctance to accept

propositions put to him during his cross-examination reached

disturbing proportions. A series of examples are illustrative.

During his cross-examination Mr. Van Toen refused to accept

propositions put to him that the Prism code was reasonably well

commented or written:

QWill you agree with me at least, sir,

that the Prism code is reasonably

well commented?

ANo, because I see certain things missing

from it.

QWill you agree with me that it is

reasonably well written?

AIt seems to be sort of broken out into -

- there's no great use of things

like Go To which is -- no, looking

through it, it's not current BASIC

programming.

constructs

It's -- some of the

that are older

constructs, there are new ways of

doing things in the language which

aren't used.

- 210 -

QSo does that mean that you think it is

or is not reasonably well written?

AIt doesn't use constructs which would

make it a better written program.

For the level -- I would say that -

- no, that some of the modules are

real great size, the mixture of

generic code and specific code

applications, specific code.

QNow, I'm not a programmer and so I don't

have great ability to deal with the

editing,

the

editorial

remarks

you've made, so can you just tell

me in a yes or no format, if that's

possible, and if it isn't, say so,

is the Prism code reasonably well

written?

AIf the totality of the code itself plus

the commenting is what we're taking

about, the answer is no.

QI was breaking --

AI don't think you can. I think you have

to look at it as a package, okay.

The reason is that if somebody

comes to make

a

change to

a

program, how easy is it for them to

make that change.

QOkay. So it is neither reasonably well

commented

written?

nor

reasonably

well

A

Right.

The thrust of the cross-examination became apparent when

counsel produced to Mr. Van Toen a preliminary report he had

prepared for HMRI in September of 1987. Within the body of that

report which was entitled "Comparison Between the Basic and Q-PRO-

- 211 -

4 Versions of Savoir-Faire" Mr. Van Toen expresses the view that

". . . the [Prism] programs themselves are reasonably well

commented. . ." and went on to say that:

Although, given the number of pages of code,

it was not practical to read every line, it

appears that the code is reasonably well

written.

In another area Mr. Van Toen was asked whether or not two

programmers working in clean rooms would have produced report

formats and data entry screens as similar as those in the Prism

and Q-PRO-4 versions of software. The sequence is:

QPrism, Q-Pro-4 Savoir-Faire are similar

from a user's perspective at least

in part because, as you understand

your

instructions,

a

deliberate

attempt was made to make them look

similar?

AThat's what I've been told, yes.

QOtherwise, now there being a clean room,

as I understand that jargon, so

we're not trying to make one look

like another, do you agree that it

is unlikely that the menus of

another system would be as similar

as they are in the Prism and Q-Pro-

4 -- of between Prism and Q-Pro-4?

AWell, if a professional programmer was

asked, given the HMRI input form

and asked to make a system that

replicated the process that's -- or

the entry process on that form,

then you would expect a high degree

of similarities in form.

QDo you agree that the report formats and

the data entry screens would not

likely be as alike as are Prism and

Q-Pro-4

for

this

third

party

- 212 -

program system?

AI'd really have to -- no, I don't think

I agree with that. I would have to

know what the constraints were on

the other -- on the other party.

In his preliminary report Mr. Van Toen seemed to have been

quite prepared to express a contrary opinion:

Thus any experienced systems developer, given

the HMRI manual and the task of programming a

data entry system for HMRI abstract dat, would

have produced a product similar to both the

Prism and HMRI systems.

However, it is not

likely that the menus, report formats, data

entry screens and display screens would be as

much alike as the two existing systems. This

is because a deliberate attempt was made to

make the HMRI version look like the Prism

version in order provide a level of comfort to

users making the switch.

(emphasis added)

There were other areas of seeming disagreement between the

opinions expressed and conclusions reached by Mr. Van Toen in his

preliminary report and those given at trial.

For the reasons I have expressed I prefer the evidence of

Professor Dexter wherever they differ from those expressed by Mr.

Van Toen. Quite apart from the other factors Professor Dexter's

superior qualifications and, more importantly, the depth of their

respective analysis leads to that result.

iii) CONCLUSIONS

- 213 -

The evidence in this case overwhelmingly supports the

conclusion that Stephen Imbert in producing Q-PRO-4 Savoir-Faire

resorted to Prism's source code and copied Prism Savoir-Faire.

This conclusion does not flow from individual pieces of

evidence viewed in isolation but from the entire body of evidence

and the individual pieces of evidence viewed within it and in

relation to the rest of the evidence.

The defendants admit that they set out to and did in fact

generally copy the screen formats of the Prism systems. Given the

features of a fourth generation language certain consequences flow

from that decision and the choice of a language such as Q-PRO-4

those consequences are best described by Mr. Imbert in the

evidence put to him from his discovery:

QLet me ask one question before we leave

that area though.

I

have an

imperfect understanding of these

things, as is probably obvious, but

let me give you a proposition and

you tell me if I'm right or wrong.

In very broad terms, writing a

program such as you wrote for HMRI

in Q-Pro-4, does that language

dictate or was that it convenient

with that language to start with

the screens, then to define the

fields, then to write the edits and

go in that order?

A

Yes.

QSo that in effect the screens the user

will be working with are going to

dictate the program structure; is

that fair?

- 214 -

AIn Q -- yes. The way Q-Pro-4 is set up,

yes.

Were you asked those questions and did you

give those answers?

A

Yes.

QAnd were those answers true?

AYes.

In their decision to copy the screens, a decision which was

the first step in the process and the decision to utilize Q-PRO-4

the programmer, Mr. Imbert, decided on a very practical level to

copy the program structure of Prism Savoir-Faire.

In addition, the defendants admit that Mr. Imbert looked at

the Prism source code to obtain the "public domain formula" for

the OHIP check digit.

They also admitted, during the course of

the trial that Mr. Imbert wrote a specific Q-PRO-4 program

(ICD9IN.PIC) to copy Prism's ICD-9 codes from their program into a

Q-PRO-4 format which was then used by Mr. Imbert in the

preparation and distribution of his Alberta Demonstration version

of the Q-PRO-4 software.

The fact that these codes, taken

directly from the Prism software, were changed is not surprising,

Alberta was a former CPHA province and it was always intended to

provide to them the capability of utilizing the ICD-9 CM codes.

The fact that in March 1987 the Board of Directors had learned of

the "rewrite" and that shortly thereafter Mr. Colman was fired and

that on May 8, 1987 this lawsuit was commenced must be considered

- 215 -

to have focussed considerable attention on Mr. Imbert and his

program.

The existence of changes after those dates must be

considered in the light of those events.

The fact remains that in producing the demo version for

Alberta Mr. Imbert directly accessed Prism's program and copied

those codes. Whether or not the codes originated with the World

Health Organization and whether of not they were subsequently

removed from the commercial version of the software, is beside the

point, the fact of this copying was not disclosed until trial at a

time after Mr. Jeffery had discovered the existence of the program

which the defendants had unwittingly disclosed.

Until that time

they denied any copying except the single public domain formula.

The defendant's retention of the University Hospital

computer, its use by Mr. Imbert in his development of the Q-PRO-4

program, the presence on the hard disk of both Prism source and

object code during that process, and the efforts of Mr. Imbert

and/or Ms. King to conceal those facts by erasing and wiping the

hard disk before it was returned to Prism also form a part of the

factual backdrop and demonstrate the lengths to which Mr. Imbert

was prepared to go to conceal his activities and to, in the end,

replace the Prism software with his own.

Similarly the scheduling and attendance at the October

- 216 -

training session was utilized not to advance HMRI's capability of

maintaining the Prism software, a project they had already

abandoned, or never undertaken, but to assist Mr. Imbert in

understanding and copying the Prism software.

Mr. Jeffery's

description of how his curriculum was summarily dismissed and how

he was directed to explain specific areas and features of his

software coupled with the fact that some portions of the printed

source code were annotated in areas subsequently reproduced with

amazing similarity in Mr. Imbert's program lead to the inference,

that I choose to accept, that Mr. Jeffery's explanations were

being used to assist in the copying of his program.

The scheduling of the course at a time when Mr. Imbert was

well into his project and the efforts made consciously and

deliberately to keep the fact of that project from Mr. Jeffery

taken together with the absence of any formal notice that there

were major problems in need of immediate repair all tell heavily

against the version of events advanced by the defendants. These

are the actions of a thief prowling at night not those of an

aggrieved

contractual

partner

seeking

solutions

within

a

contractual setting.

Mr. Jeffery's "fingerprints" when viewed within the framework

of events are in themselves significant. One or two coincidences

can be expected in any similar situation, let alone one where a

- 217 -

piece of software is being written to serve essentially the same

purpose.

extreme.

These "coincidences" however, stretch credulity to the

Five of these "fingerprints" are, in my view, of particular

significance.

The unknown birthdate, the way in which operating

room deaths are handled, the ICD-9 prefixes, the deceased patient

preload and finally the way in which the systems handle service

transfer diagnoses all have in common one or more unique

characteristics which are peculiar in the sense of having

programming or logical errors within their coding or peculiar and

distinctive solutions to the programming.

In the case of the

unknown birthdate it is the choice of testing against a similar

range of years which, in error, permits the entry of a future date

of birth. In the case of the handling of operating room deaths it

is the use of a "mask" to convert a single alpha character into

four separate mutually exclusive fields coupled with an error in

the logic which precludes the inclusion of a death which occurs on

the tenth day after an operation.

In the case of the ICD-9

prefixes it is the use of the "P" prefix for procedure (even

though Q-PRO-4 records these in a separate file) and the use of a

"VV" code for contributing factors but only a single "E" prefix

for external causes.

With the Deceased Preload section it is

Prism's flip of the initial "Y" response to a "N" on the Exit

Alive response which necessitates the crossfield edit test to see

- 218 -

if an "N" exists in the second screen.

Finally in the case of the service transfers diagnosis Prism

adopted the strange and anomolous double use of another data field

because of an original programming error.

The presence of any one of these unique features, within the

work of a different programmer might be explainable by mere

coincidence although in my respectful view the application of

logic by the programmer would avoid the majority of them.

The

presence of all of these unique features albeit with some minor

variations within the work of a single programmer who had access

to the source code of Prism's program must lead to but one

conclusion, that he was both accessing that source code and

copying it.

When the presence of all of these unique features is combined

with all of the evidence in this case including the conduct of the

parties and the credibility findings I have made the conclusion

and the evidence supporting it becomes irresistible.

There is another aspect of the evidence concerning the

operating room death routine and the service transfer diagnosis

routine which is important.

In, my respectful view, no

experienced and capable programmer writing independently is likely

- 219 -

to duplicate the errors in logic and design inherent in these two

routines. The errors inherent in the Q-PRO-4 replication of the

Prism program are not explainable in the context of the design and

programming process in which the programmer's attention is

focussed on the design and logic of the programming solution, but

by a programmer whose attention is focussed on the rewriting of

another programmer's logic into a language of his choice.

To all of this is added the evidence and opinions of

Professor Dexter, which I accept.

(C) LEGAL IMPLICATIONS

Earlier in these reasons I have dealt with the formation

of the contract, the licence agreement and the relationship

between Prism and HMRI. In beginning the discussion of the legal

results and conclusions I will deal initially with the contract

issues and then those arising in copyright.

i)

CONTRACT

1) INTRODUCTION

The defendants begin their written submissions on the issue

of contract with the following statement:

589.

the

There is no part of

where plaintiff's and

present

case

defendants differ so much as in their

interpretation of [the] contract.

- 220 -

The plaintiff's respond to this assertion in their reply by

saying that:

. . . there is no part of the case which

demonstrates more clearly the one-sided,

miserly, and overly technical approach to the

relationship between Prism and HMRI than the

defendants interpretation of the contract by

the defendants.

stripped of

In its starkest terms, and

all the verbiage, the

interpretation urged by the defendants amounts

to virtually all of the benefits flowing to

HMRI with Prism being entitled to nothing more

than the bare minimum of royalty payments

stipulated in the agreement.

The touch of animosity which emerges from this response may

be real but behind it lies a serious point. In their defence of

this action the defendants have adopted an overall approach which

has consistently sought to isolate portions of evidence or

submissions from the surrounding circumstances or other evidence

and then to dismiss that particular piece of evidence or

submission from consideration as being irrelevant.

I have commented already within these reasons from time to

time on such positions when dealing with particular issues. The

pattern continues in this area of the submissions.

In their introduction to the contractual issue the defendants

begin with a discussion of principles which they submit should be

applied in the present case.

It begins with a discussion of

collateral oral contracts and the restrictions on their use in the

- 221 -

There is no allegation by the

interpretation of contract.

plaintiff's in the present case of any oral collateral agreement.

This portion of the defendant's argument is irrelevant to the

issues in this action.

The defendant's then embark on a

discussion of the contra preferentem rule in which they urge the

use of this rule against the plaintiffs with respect to the

interpretation of the contractual documents.

This is not a case in which one side dictated the form of the

contractual agreement between the parties but rather, on the

evidence, one in which both sides participated in the negotiation

and finalization of its terms.

The fact is that much attention

was focussed during this trial by the defence on the fact that

HMRI after its review of the proposal refused to allow within it

any positive duties with respect to Prism's desire and intention

to develop enhancements and add-on products.

portion of the Proposal was redrafted.

As a result this

The defendant's submissions are based in large part on the

fact that Prism had the contractual documents reviewed by their

lawyer who in fact drafted at least some portions of them. What

the defendants seek to isolate and ignore in these submissions is

that there is not the faintest suggestion that these documents

were prepared by Prism and presented to HMRI for their signature

in that form. HMRI not only participated in the drafting of the

- 222 -

final form of agreements they too sought advice by having them

reviewed.

In their case they were reviewed by officials within

the Ministry of Health of Ontario whose comments and suggestions

were returned to HMRI as part of the negotiation process.

The contra preferentem principle is applied in circumstances

where an ambiguity in the document leads to two or more possible

interpretations and the other party did not participate in the

drafting of the contract.

The defendant's in advancing their

argument rely upon a passage from the decision of Le Dain, J. in

Hillis Oil & Sales Ltd. v. Wynn's Canada [1986] 1 S.C.R. 57 at p.

69.

The passage which immediately precedes that quoted by the

defendants reads as follows:

. . . It is true that this rule has been most

often

invoked

with

reference

insurance

in such

to

the

construction

particularly

of

clauses

contracts,

contracts

purporting to limit or exclude the insurer's

liability.

Statements of the rule and its

application in such cases may be found in the

decisions of this Court in Consolidated-

Bathurst, supra, and McClelland and Stewart,

supra. The rule is, however, one of general

application whenever, as in the case at bar,

there is ambiguity in the meaning of a

contract which one of the parties as the

author of the document offers to the other,

with no opportunity to modify its wording.

(emphasis added)

To suggest that these circumstances exist in the present case

requires one to ignore the evidence as a whole and instead focus

- 223 -

only on the activities of Prism.

This is a suggestion and a

process which I decline to adopt. The whole of the evidence must

be looked too and, the contra preferentem principle has no

application when this is done.

2)

THE RELEVENT CONTRACTUAL PROVISIONS

In an earlier portion of these reasons I have already found

that the commercial purposes of this contract were to develop and

distribute a low cost PC based medical abstracting system which

HMRI would place in hospitals across Canada.

In p. 597 of their written submissions the defendants submit

that the relevant clauses of the agreement are 1, 2, 5, 6, 8, 9,

10, 14 and 15. Notably absent from this list is Article 11 which

provides:

11.

All copies of the

Licensed Programs made by HMRI including

translations, compilations and partial copies

with modified or up-dated work are the

property of PRISM.

HMRI will reproduce and

include copyright notice on any such copies in

accordance with such copyright instructions as

may be provided to HMRI by PRISM.

The significance of Article 11 lies in its application to the

defendant's alternate submission that the Q-PRO-4 version of

Savoir-Faire constituted a modification of the Licensed Programs

within the meaning of Article 8. This alternative submission will

be addressed first in this discussion in light of the findings of

- 224 -

copying which have already been made.

The license granted to HMRI by Article 1 of the agreement is

". . . to use the Licensed Programs in its business operations in

Canada and to sub-license the Licensed Programs to HMRI's Canadian

Clients under the terms and conditions of this Agreement."

The

terms of the license are further defined by the terms of Articles

5 and 6 of the agreement.

Article 5 both defines and restricts the use and purpose for

which HMRI may use either the Licensed Programs or the source

code:

5.

HMRI

shall

be

entitled to use the Licensed Programs granted

hereunder in connection with its business

operations in Canada but it is expressly

understood and agreed that HMRI shall not

sell, lease, sub-lease, market or otherwise

make available in any manner the source code

for such Licensed Programs to any person or

entity without the prior written consent of

PRISM and in the event HMRI does so sell,

lease, sub-lease, or otherwise make available

in any manner the source code for such

Licensed Programs, PRISM may terminate this

Agreement forthwith.

Article 6 extends the license and defines it to be linked, so

far as the executable programs are concerned to uses on its "data

processing equipment in the course of their business operations in

Canada." The last sentence purports to and, in fact, limits the

use of compiled object code of the licensed programs to that

- 225 -

defined purpose or use:

6.

HMRI

is

hereby

authorized to use the Licensed Programs in

compiled, object code form on its data

processing equipment in the course of their

business operations in Canada. HMRI is hereby

also authorized to sub-license the Licensed

Programs in compiled, object code form to

their Canadian clients for the client's use on

their data processing equipment in the course

of their business operation in Canada.

Save

and except for these purposes, HMRI shall not

have the right to use, print, copy or

otherwise make available in any manner the

compiled

object

code

for

such

Licensed

Programs or other such material as are

supplied hereunder.

The heart of the agreement between the parties is found

within these provisions which define the scope of the license and

the purpose for which the licence is extended.

In my view they

contain no ambiguity. The object of the license is to allow both

HMRI and its client hospitals to whom it is sublicensed to utilize

the Prism software on their respective data processing equipment

to process that data which they collect through the use of the

program to fulfill their data submission requirements for HMRI and

to manipulate and use that data themselves for their own purposes.

3)

THE MODIFICATION CLAUSE

In Article 8 HMRI is granted an additional right which is

both separate from the license contained within the agreement and

dependent upon it:

- 226 -

HMRI may modify any

8.

Licensed Programs and/or merge such Licensed

Programs into any other program material to

form an updated work for the purposes as

detailed in this Agreement, provided that in

the

event

the

Licensed

Programs

are

discontinued as a result of default by HMRI

hereunder,

associated

the

materials

Licensed

will

Programs

be completely

and

removed from the updated work and dealt with

under this Agreement as if permission to

modify and/or merge had never been granted.

It is fundamental to the right to modify the licensed

programs granted by this provision that the modifications must be

done ". . . for the purposes as detailed in this Agreement. . ."

This provision is not and cannot be construed as providing HMRI

with any form of unlimited right to modify or change the licensed

programs. Such a construction would be not only contrary to the

wording of Article

8

itself but would also ignore the

acknowledgment of Prism's ownership of the programs contained in

both the initial recital and Article 11.

The wording of Article 11 is crucial in the present action

because it contains an unequivocal acknowledgment that all copies

made by HMRI including "modified or up-dated works" ". . . are the

property of Prism." The Article also requires HMRI to reproduce

Prism's copyright notices on such "modified" works.

On the facts of this case and in terms of the defendants

alternative position, if the Q-PRO-4 version is found to be a

- 227 -

modification authorized by Article 8 it is, and remains, the

property of Prism under the terms of Article 11.

The first question to be addressed is that of whether or not

the Q-PRO-4 version of Savoir-Faire constitutes an "updated work"

within the meaning of Article 8.

The defendants submit that Article 8 gives to HMRI ". . . a

right to modify, without putting any limitations on that right."

They then invite the court to find that the license agreement

prevails where there is any conflict between the proposal and the

license agreement. I have earlier in these reasons indicated my

conclusion that the unusual wording of Article 15 has the effect

of preserving the proposal and elevates it to form a part of the

contract document itself.

It is not therefore a matter of a

conflict arising between the two and one prevailing; Prism and

HMRI each signed both documents and in the wording of Article 15

of the License Agreement chose wording which specifically

preserves the proposal from the operation of the general

exclusionary clause. As such, the agreement of the parties must

be found in the two documents. In construing the agreement of the

parties the use of language in one which explains the use of

language in the other without contradicting its terms is to be

given its full effect.

- 228 -

Even if, as the defendants suggest, there is some ambiguity

in the obligation of HMRI to make every reasonable effort to

market "the Licensed Programs and/or related material. . ." as set

out in Article 14 of the License Agreement. The meaning is clear

when viewed in the light of paragraph 6 of the Proposal which

reads in part:

This

proposal

is

also

based

upon

the

understanding that HMRI will at all times

apply their best efforts toward the marketing

and distribution of this software, with a

concentration of their initial efforts being

directed toward their clients in Ontario and

in Quebec.

(emphasis added)

The defendant's go on to suggest that the purpose of the

agreement is to be found in the phrase that allows HMRI to

sublicense to their Canadian clients and that the purpose should,

in effect, be restricted to anything which serves this limited

purpose. They conclude this position in this paragraph of their

written submissions:

613.

There is nothing in

T6 CONTRACT which says that any modified or

updated work by HMRI has to be in BASIC.

Further, the only mention of BASIC in all of

T6 CONTRACT is at T6 PROPOSAL 2, where it says

the software will be written in BASIC at the

request of HMRI.

Note that, as discussed

above in DFARG 599, the license cannot be

revoked for any alleged breach of T6 PROPOSAL,

but only for

a

breach of the license

In any event,

agreement, i.e. 6 AGREEMENT.

the term in T6 PROPOSAL 2 is clearly for

HMRI's benefit, as seen from the words "at the

request of HMRI".

- 229 -

This reasoning with a skillful blend of exclusion and myopia

has now reached the point where the object and purpose of a

contract reached between two parties has been reduced to the

object and purpose of one.

Prism's purpose in entering the

contract and writing the software so as to achieve its penetration

of the market and the placing of its software in hospitals across

Canada to provide it with a market for the enhancement and add-on

products it wished to sell has now been completely subsumed by a

purpose which the defendants boldly assert is to allow HMRI to

achieve its goal of ensuring it receives electronic abstracts and

to completely change both the language in which the software is

written and the way in which data is stored by the software in its

history file; so as to eliminate Prism's access to the market it

bargained for.

In p. 613 of their argument the defendants assert firstly

that as the License Agreement itself does not mention BASIC and

that the only mention of BASIC is in the Proposal which says that

"At the request of HMRI the software will be written in BASIC. .

." that this is a term ". . . clearly for HMRI's benefit."

I reject this submission if it is intended to suggest that

this was in any way a term which, at its option, HMRI could waive

or change. Once again this view of the contract focuses entirely

- 230 -

on the interest of one party to the contract, and ignores Prism's

rights of ownership.

In addition to the market penetration it would achieve

through HMRI's placing of its software in hospitals across Canada,

the other advantage gained by Prism in completing the contract was

its knowledge of the software, its structure and the data storage.

It was these very features and their intimate knowledge of them

which gave Prism both the capacity and the advantage it felt was

necessary to exploit the market for enhancements and add-on

products.

To suggest that HMRI had the capacity to unilaterally and

secretly change both the language and the data structure of the

software is to permit them to take all of the fruits of the

contract available to them and then to harvest those left for

Prism.

This does violence not just to the language of the

contract but reason itself.

The defendants then go on to assert that within the

modification provision contained in Article 8 "There is no

obligation in this clause to keep the program in the BASIC

language."

Once more this court has been asked to ignore the

whole of the agreement between the parties, and all of the

surrounding circumstances and to place a single clause of the

- 231 -

agreement under a microscope and to make a finding as to HMRI's

rights not based on what is contained within the clause but on the

basis of what isn't there.

invitation.

Once again I must decline this

Underlying the entire agreement of the parties and central to

it as the primary purpose of the contract is the writing of

Savoir-Faire in the BASIC language by Prism.

The purpose for

which this program is to be written is so that HMRI can market and

distribute the resulting program to its client hospitals.

Within their contractual framework HMRI was given the right

to sub-license the software to its client hospitals and after

October 1, 1986 Prism's obligation to provide software maintenance

for the Licensed Programs ended.

This is the factual and contractual setting in which Article

8 is to be construed. Under the rights granted in this provision

HMRI had the right to modify "any Licensed Programs" and/or to

"merge" them "into any other program material to form an updated

work for the purposes as detailed in the Agreement. . ."

The Oxford English Dictionary, second edition defines

"Modify" as meaning "To make partial changes in; to alter without

radical transformation."

To suggest that the Q-PRO-4 rewrite

- 232 -

constitutes a modification of the existing Prism software within

this definition is absurd.

The Q-PRO-4 rewrite was neither a

partial change nor an alteration without radical transformation.

The same dictionary defines "Merge" as "To sink (a lesser

estate, title, etc.) in a greater one.

Hence gen., to cause

(something) to lose its own character or identity in something

else." This is closer but still far off the mark. There was no

merger on the facts of this case, there was no greater "something"

into which the plaintiffs' software was merged instead it was

taken and transformed into a qualitatively different product

designed differently in certain key areas but intended to serve

precisely the same function and to eliminate the plaintiff's

product.

On the basis of these definitions alone I would be inclined

to find that the Q-PRO-4 rewrite could not be brought within the

provisions of Article 8, however, when the rights provided by that

section are examined within its proper context of "the purposes of

the agreement" there is no doubt. The Q-PRO-4 version of Savoir-

Faire cannot come within the scope of Article 8.

The right to

modify or merge is governed by the overall purposes of the

contract and are intended, in my view, to permit HMRI to maintain

and support the software after October 1, 1986.

I find that it

does not constitute an "updated work" emerging as a result of an

- 233 -

authorized modification or merger process.

What the defendants

set out to do was to engineer the complete replacement of the

plaintiff's software and in retrospect to justify that process.

The absence of any compliance with the requirement that Prism's

copyright be displayed on any updated work puts this submission in

its proper perspective as an attempt to force a contractual

interpretation which is wholly out of harmony with the evidence

and is on its face manifestly wrong.

4)

THE OBLIGATION TO PROTECT THE SOURCE CODE

By Article 10 of the License Agreement HMRI agreed to take

action ". . . by instruction, agreement or otherwise with any

persons permitted access to the Licensed Programs and/or

associated material so as to enable HMRI to satisfy [their]

obligations under this Agreement."

Rather than doing so they provided full access to all of

Prism's source and object code to Mr. Imbert while he set about

copying the software and creating a version of the software which

would effectively replace the Prism product and deprive them of

the market for which they had bargained. This entire process was

carried out, during its initial phases, in deliberate secrecy. If

this was not bad enough, they allowed Mr. Imbert to utilize a

computer system they had received from Prism to do so.

The

evidence is far from complete as to who at HMRI may have known

- 234 -

about the use of the University Hospital computer but it is clear,

and I find, that Wendy King did.

The use of the source and object code by Mr. Imbert to copy

and otherwise assist him in the writing of his Q-PRO-4 version of

the software constituted a breach by HMRI of its obligations under

Article 10 of the Licence Agreement.

HMRI did not seek or obtain a written consent from Prism when

it "made available" to Mr. Imbert the Prism Source Code.

The

irony of this situation is that HMRI by their deliberate deception

enlisted Mr. Jeffery to disclose much of the source code and to

offer his assistance to Mr. Imbert in his efforts to copy the

Prism software by misleading him with the suggestion that Imbert

was undertaking the maintenance of the Prism software.

This

deception was decided upon and carried out with the active

participation of a senior staff member, John Stenabaugh, whose

background in computer matters must have made him conscious of

what was going on.

The active deception coupled with the provision of the source

code for a purpose not only outside the scope of the agreement but

antithetical to its legitimate purpose constituted a breach of

HMRI's obligations under Article 5. Such a breach entitled Prism

to terminate the License Agreement forthwith.

This particular

- 235 -

provision is a measure of the seriousness with which the agreement

treats the wrongful disclosure of Prism's source code.

For greater certainty, I construe the provisions of Article 5

and HMRI's rights to use either the Licensed Programs or the

source code with which they were provided to be restricted by the

overall purpose of the agreement.

That purpose does not, and

cannot include the use of the program materials and code to copy

the programs and replace them. This result would ignore both the

acknowledgments of Prism's ownership of the software contained

within the agreement and overall purpose of the agreement between

the parties. Such acts are best viewed in the way in which Mr.

Stenabaugh, Mr. Imbert and Ms. King viewed them as acts which must

be kept secret, in particular, from Prism.

characterize the legitimacy of their actions.

This does much to

I find as well that Mr. Imbert during the course of his

copying of the Prism software had access to and made use of the

compiled object code of the Prism software. This is a use again

outside the scope of what is contemplated by Article 6 and the

License Agreement as a whole. Particularly in light of the last

sentence of that provision.

I find that this conduct constitutes a breach of HMRI's

obligations under Article 6.

- 236 -

5)

HMRI'S FAILURE TO PROVIDE SOURCE CODE

The overall fact which emerges from the evidence concerning

the Q-PRO-4 project is that in the contract documents HMRI

acknowledged Prism's ownership of the programs comprising Prism

Savoir-Faire.

They also acknowledge in their alternative

submission that the Q-PRO-4 software was a modification of the

licensed software under Article 8 and that the Q-PRO-4 software

belonged to Prism but was licensed to HMRI.

It is an interesting concept that a licensee can alter

radically the licensed object and withhold from the owner any

notice of the changes or their purpose.

On September 28, 1986 in anticipation of HMRI undertaking

maintenance of the software Mr. Trueit writes to address the issue

of the exchange of source code:

For the record, we would like to remind you -

and we would ask that you advise others within

HMRI - that the License Agreement restricts

the manner in which HMRI may make the code

accessible

to

non-HMRI

personnel.

The

Agreement also provides that the modifications

are the property of Prism and we would ask

that a copy of all source (including source in

progress) and the object of Savoir-Faire be

sent to us at least once per month beginning

November 1, 1986. Details for this can be can

be(sic) worked out over the next month.

Like so many other letters written by Prism this request went

- 237 -

unanswered. Yet the defendants who did not reply to this request

in their written submissions boldly assert that "Prism took no

steps to 'work out details', and it would seem that both parties

forgot about the request."

This assertion is breathtaking in its simplicity, and does

much to characterize the manner with which the defendants have

approached this action and, more importantly, their treatment of

their contractual partner during the currency of their dealings.

Prism, the owner of the software that HMRI is using has

written to HMRI to underline its contractual rights to any

modification of the program code and to suggest a regular exchange

of code.

HMRI who must be taken to have knowledge of their

employee Mr. Imbert's activities ignore this perfectly reasonable

request and then criticize Prism for not taking steps to work out

the details.

It is not a question of whether Prism "needed" the source

code as the defendants suggest in p. 634 or of whether the

contract specifically required HMRI to provide it as they argue in

p. 635, or of whether a collateral oral agreement to require it

should be found as they address in p. 636. Prism's right to the

source code flows from the ownership rights acknowledged by HMRI

within the License agreement. In my respectful view, Prism needed

- 238 -

no other grounds to assert their right of ownership but if any

were needed those outlined in Mr. Trueit's correspondence of

September 28, 1986 are both sufficient and compelling.

In paragraph 631 and 632 the defendants advance two other

propositions which must be addressed:

631.

It is also submitted

that HMRI had no obligation to prepare a

modified or updated work only when or if

Prism's programs were in some way defective.

The Modifications Clause contains no such

limitation, nor does it require Prism's

approval to make modifications.

632.

Although T6 CONTRACT

does not require HMRI to give notice to Prism

of its intention to make modifications, HMRI

did in fact give Prism adequate notice so that

Prism could make its own arrangements for

marketing in Alberta and New Brunswick, as

well as other provinces, if it wished.

HMRI's obligations were to exercise the rights afforded to it

under the contract within the contractual framework in which they

were granted and recognizing the contractual purpose for which

they were given.

The defendants failed to do this and in their

assertion of their interpretation of the "Modification Clause"

seek to assert ownership rights themselves. This was neither the

purpose nor the intent of the agreement.

6)

THE MARKETING PROVISION

In p. 632 the defendants again seek to limit the notice

- 239 -

requirement to a narrow issue which serves HMRI's needs.

The

point of the lack of notice was not to allow Prism to make

separate marketing arrangements for Alberta and New Brunswick but

to allow Prism to respond to a course of action embarked upon by

HMRI which amounted to a fundamental breach of the contractual

arrangement. This breach was not aimed at taking advantage of the

markets in Alberta and New Brunswick but at replacing Prism's

software at every site at which it had been installed and

recanting on its contractual obligations to ". . . make every

reasonable effort to competitively market the Licensed Programs

and for related material in every Province and Territory in Canada

as the respective versions of the Licensed Programs become

available to HMRI." (emphasis added).

Whether the decisions were made in anticipation of taking

advantage of the money they knew was likely to be made available

by the Ministry of Health of Ontario to allow hospitals to

purchase Savoir-Faire, to seize the advantage of its takeover of

the CPHA provinces or because of an unfortunate combination of

managerial weakness and, in Mr. Imbert's case, a forceful

personality, or

a

combination of the three is of little

consequence. HMRI embarked upon a course of conduct which was

aimed not at correcting real difficulties but at taking for

themselves the advantages the contract gave to Prism.

- 240 -

This conduct constitutes a breach of their obligations under

Article 14 of the License Agreement.

7)

REPUDIATION OR BREACH

HMRI gave Prism no written notice of its present

allegation that it considered problems with the software as being

serious enough to warrant a "rewrite" of the software. In fact,

on the whole of the evidence both in their conduct and the

information they provided they led Prism to believe that things

were going well, with the exception of continuing problems at the

multi-user sites.

Whether or not the position HMRI advances in this action

amounts to a repudiation by them at the time is a difficult

question.

The critical point, however, is that the "problems"

The relationship between the

were not communicated to Prism.

parties was such as to require notice and its absence leads to the

strong inference that the present complaints are afterthoughts,

more illusory than real. On a practical level there must be taken

to arise from the requirements of the project and the joint nature

of their efforts some obligation to communicate any problems or

dissatisfaction.

In Unified Systems Ltd. v. Clearwater Lobsters Ltd. (1986) 58

Nfld. & P.E.I.R. 138 the court had before it facts which involved

- 241 -

an agreement for the design, licensing and installation of a

computer system.

When the defendant purchased and installed a

different system prior to the completion of the contract the court

observed at p. 146:

. . . Implementation of a computer system

requires effort by the customer and seller.

There was no written notice from Clearwater to

Unified of any dissatisfaction of Clearwater

with the work of Unified prior to the

termination

letter

of

Clearwater

dated

February 18, 1985.

These observations are particularly appropriate in the

circumstances of this case.

In M.L. Baxter Equipment Ltd. et al v. Geac Canada Ltd. et al

(1982) 133 D.L.R. (3d) 372 the Ontario High Court in considering a

case involving a computer equipment lease found that:

. . . Clearly, the plaintiff could not,

without reasonable notice, come along one day

and say to the defendant that he was in

fundamental breach of his obligations and

terminate the contract: see Charles Rickards

Ltd. v. Oppenheim, [1950] 1 K.B. 616.

plaintiff must give reasonable notice of his

intention to do so. What is reasonable

The

depends on the circumstances of the case.

The reasons at p. 380 go on to find that:

The termination of

an agreement is a very serious matter. It is

the cessation of a legal relationship between

two parties which most often leads to costly

legal actions, such as this one, in order to

settle the differences between them.

A

- 242 -

notice, therefore, which is the commencement

of the termination procedure must be one which

brings home to the party to be charged the

purpose thereof so that it may be said that

there was no doubt that the defendant realized

the serious consequences of his inaction.

If these are the requirements upon termination, in my view,

they apply with equal force, to a decision to embark on a course

of action as fundamental as a replacement of the plaintiff's

software in the present case. A positive obligation to notify the

plaintiff of their action arose on the circumstances of this

action whether or not the defendants actually thought the

plaintiff's software was seriously flawed or that the rewrite was

authorized by Article 8.

This the defendants failed to do in either a timely or

forthright manner.

In considering Prism's actions in the months following the

training course in October 1986 and leading up to the commencement

of this action on May 8, 1987, it is important to recognize that

Prism had available to it only limited information as to the

defendants activities.

The true nature of those activities was

deliberately withheld by those involved at HMRI. The staged way

in which HMRI chose to inform Prism of their intentions had two

effects, it allowed Mr. Imbert additional time to prepare his copy

for marketing, and it "stonewalled" Prism and delayed any informed

- 243 -

response on their part. The suggestion that Prism in some manner

acquiesced in HMRI's activities would require a knowledge and an

appreciation which not only did not exist but was precluded by

HMRI's deliberate actions.

Prism in the face of a deteriorating relationship and limited

knowledge sought to maintain the relationship and work out the

problems. When their attempts to do so were rebuffed and as they,

over time, became aware of HMRI's activities and intentions they

sought legal relief.

HMRI did not, in my view repudiate the contract, instead they

fundamentally breached it while trying for their own benefit to

hide that fact from Prism to gain time to secure their position.

Prism was entitled to and by the commencement of this action

accepted those breaches and exercised their right under the

contract to terminate both the agreement and the licence it

contains.

ii) COPYRIGHT INFRINGEMENT

1) INTRODUCTION

The Copyright Act, R.S., C C-30 defines "computer

program as:

. . . a set of instructions or statements,

expressed, fixed, embodied or stored in any

manner, that is to be used directly or

indirectly in a computer in order to bring

about a specific result;

- 244 -

A computer program in its source code form has been found to meet

the necessary criteria of representing an ". . . expression of

thought in an original form." IBM v. Ordinateurs Spirals, (1984)

2 CIPR 56 at p. 62.

The protection afforded by the Copyright Act is to be found

in s. 3(1):

3.(1)

For the purposes of

the work or any

this Act, "copyright" means the sole right to

produce

or

reproduce

substantial part thereof in any material form

whatever, to perform, or in the case of a

lecture

to

deliver,

the

work

or

any

substantial part thereof in public or, if the

work is unpublished, to publish the work or

any substantial part thereof, and includes the

sole right

(a)

to

produce,

reproduce, perform or publish any

translation of the work, and to

authorize any such acts.

The object of s. 3 was examined by Reed, J. at the trial

level in Apple Computer Inc. v. Macintosh Computers Ltd. (1986) 28

D.L.R. (4th) 178; affirmed (1987) 44 D.L.R. (4th) 74; affirmed

[1990] 2 S.C.R. 209. In her decision Reed, J. at p. 213 wrote:

. . . The legislation historically, in my view

had two purposes: to encourage disclosure of

works for the "advancement of learning", and

to protect and reward the intellectual effort

of the author (for a limited period of time)

in the work.

- 245 -

Another aspect of this decision is important in the present

action. Madam Justice Reed in the course of her decision had to

determine whether a computer program which had been recorded in a

silicon ROM chip was to be afforded copyright protection, at p.

198 she wrote:

In

my

view

the

conversion of a work into a code, or the

conversion of a work originally written in one

code

into

another

code

constitutes

a

translation for the purposes of the Act. In

addition, as noted above, a programmer, in

creating a program, is not thinking in terms

of the specifications of the ROM chip, either

in writing the assembly code version or when

the hexadecimal notation is prepared.

What

media is finally chosen for embodiment of the

program is irrelevant to the programmer.

Accordingly, I find it difficult to accept

counsel for the defendants' argument that the

hexadecimal form of the program is a different

literary work and not a translation of the

original.

Two of the three judges of the Federal Court of Appeal held

that Reed, J. in determining that this constituted a translation

within the meaning of s. 3(1) had erred, however all three held

that the process described constituted a "reproduction" within the

meaning of the Act.

The unanimous decision of the Supreme Court of Canada was

given by Cory, J. at p. 215 he writes:

Like Reed J. and for

the same reasons that were expressed by her, I

am of the opinion that the programs embedded

- 246 -

in the silicon chip are a reproduction of the

programs in assembly language and as such are

protected by copyright under s. 3(1) of the

Copyright Act. In addition, I agree with her

that these programs constitute a form of

expression

functionally unique and cannot be regarded as

a merger of idea and expression. Since the

that

is

conceptually

and

programs contained in the silicon chip are

therefore protected under the Act, it is

unnecessary for me to determine whether the

silicon chips can be regarded as a translation

under s. 3(1)(a) or a contrivance under s.

3(1)(d).

In the end whether it is a reproduction alone or both a

reproduction and a translation the copying of the programs was

found to contravene the protection provided by s. 3(1)

In his text Computer Law

:

Acquiring and Protecting

Information Technology Carswell:Toronto) 1989 Barry Sookman writes

at p. 3-186:

. . . The right to reproduce the work

encompasses

a

wide variety of acts of

reproduction by any process for the indefinite

multiplication of copies, whether mechanically

or otherwise.

involves two elements:

actual use of, the copyright work.

The concept of reproduction

resemblance to, and

Two

elements are required to make out a cause of

action for infringement by reproduction:

first, the reproduction must be either of the

entire work or of a substantial part (or, as

it is sometimes said, there must be sufficient

objective similarity between the infringing

work

and

the

copyrighted

work,

or

a

substantial part thereof), and second, there

must be some causal connection between the

copyrighted work and the infringing work. In

other words, the copyrighted work must be the

source from which the copy is derived.

In

- 247 -

Apple Computer Inc. v. Mackintosh Computers

Ltd. Hugessen J. stated that the sole right of

reproduction

includes,

by

necessary

implication, the sole right to produce the

means of reproduction of the work, and the

sole right to produce anything used or

intended to be used to reproduce the work.

The use of the phrase "substantial part" is derived in part

from the language of s. 3(1) which refers to the ". . . sole right

to produce or reproduce the work or any substantial part thereof

in any material form whatever. . ." but there is a second aspect

to the phrase as well. It is a rare case in which proof of actual

copying is available, in those cases where it is not, an

evidentiary test can be applied through the comparison of the

original and the impugned work.

Where this comparison

demonstrates the existence of a substantial part of the original

work or "sufficient objective similarity" between the two the

evidentiary burden has been met and the inference of copying may

be drawn.

To be actionable the copying must be of the expression, in

effect, the author's original or literary work not the ideas,

concepts or underlying facts.

Infringing is also defined by the Act:

"infringing", when applied to a copy of work

in which copyrights subsists, means any copy,

including any colourable imitation, made, or

imported in contravention of this Act;

- 248 -

It is also the specific subject of s. 27(1) which provides:

27.(1)

Copyright in a work

shall be deemed to be infringed by any person

who, without the consent of the owner of the

copyright, does anything that, by this Act,

only the owner of the copyright has the right

to do.

The term "expression" is sometimes equated with the term

originality. Copyright protection is not afforded to collections

of facts but rather to the original component contributed by a

particular author.

This distinction is illustrated in a portion

of the reasoning of the Supreme Court of the United States in

Feist Publications Inc. v. Rural Telephone Service Co. Inc. (1991)

18 MED L.R. 1889. In this decision in which the court refused to

extend copyright protection to a telephone directory they wrote at

p. 1893:

This protection is

subject to an important limitation. The mere

fact that a work is copyrighted does not mean

that every element of the work may be

protected.

Originality remains the sine qua

non of copyright' accordingly, copyright

protection may extend only to those components

of a work that are original to the author.

Where

compilation author adds no written expression

but rather lets the facts speak for

themselves, the expressive element is more

elusive. The only conceivable expression is

the manner in which the compiler has selected

the

and arranged the facts.

Thus, if the

selection and arrangement are original, these

elements of the work are eligible for

- 249 -

copyright protection.

No

matter

how

original the format, however, the facts

themselves do not become original through

association.

A good example of this distinction arises in the present

case, in the form of the ICD-9 codes which were published by the

World Health Organization Mr. Trueit examined the list of more

than 10,000 codes and truncated more than half of the literal

descriptions in order to fit them within the 59 character field

provided for them.

act of truncation while leaving the remaining portion sufficiently

distinctive to be identifiable. Of the more than 5,000 codes

The element of originality emerges from the

truncated by Prism, HMRI in their review of Prism's work changed

approximately 200 before Prism incorporated them within their

software.

In my view, the resulting list of truncated ICD-9 codes

utilized by Prism by the creative process described were afforded

the necessary "originality" to justify the extension of copyright

protection.

This does not, in any way give Prism the right to

control the use of the ICD-9 codes generally, but it does afford

them protection for the truncated codes that they produced through

the process described for use in their software.

The defence submits that:

- 250 -

Once one had the

470.

idea that one would reduce the ICD9 codes to

59 characters and use particular prefixes,

very little scope for originality remains, as

one must truncate in a way to convey the

information of the ICD9 in a medically correct

form. The idea has thus been merged with its

expression.

Two difficulties arise with respect to this submission,

firstly the "idea" or decision to use 59 characters instead of 20

or 100 taken together with the choice of an identifying prefix,

may together with the truncations themselves constitute an

original expression. The opinion expressed by counsel that "very

little scope for originality remains, as one must truncate in a

way to convey the information . . . in a medically correct form."

is made without any evidence having been led to support it.

In order to succeed in a copyright infringement action the

plaintiff must demonstrate that the defendant has made an unlawful

use of the plaintiff's copyright work. It is not necessary to

demonstrate actual copying direct from the plaintiff's work if the

surrounding circumstances are such as to meet the evidentiary test

and permit the inference previously discussed to be drawn. In the

end what must be demonstrated is that either directly or

indirectly through his actions the defendant has taken the fruits

of the plaintiff's labour.

2)

THE APPLICABILITY IN CANADA OF AMERICAN LAW

- 251 -

In the United States both the size of the computer

industry and the extent of litigation have generated

a

considerable number of decisions involving the application of

copyright law in computer cases. These decisions are not binding

on courts in Canada but they are entitled to both consideration

and respect where they touch on areas which have not been

considered or on issues that are similar in both American and

Canadian legislation.

In his text on Computer Law, Sookman writes:

In Compo Co. Ltd. v. Blue Crest Music Inc.

Estey J. commented on the applicability of

American decisions that interpret American

copyright

laws

in

interpreting

problems

He pointed

arising under the Canadian Act.

out that the American copyright laws, in both

their present and earlier forms, have many

similarities to the Canadian legislation as

well as to pre-existing Imperial copyright

statutes. However, American decisions must be

scrutinized very carefully because of some

fundamental

differences

from

Canadian

legislation in copyright concepts that have

been adopted in the legislation of that

country.

definitions

Additionally, the wording in the

in the American Act is

significantly different from that in the

definitions under the Canadian Statute.

. . .

In

spite

of

the

differences between the Acts, Canadian courts,

including the Supreme Court of Canada, have

found

assistance

for the

in

examining

of

American

decision

purpose

resolving

difficult legal issues arising under the

Copyright

copyrightability of computer programs American

authorities have been cited and,

Act.

In

cases

- 252 -

notwithstanding the differences between the

wording of the Acts, have been given qualified

approval.

3)

THE PROTECTION OF THE EXPRESSION OF IDEAS

In Johnson Controls v. Phoenix Central Systems Inc.

(1989) 886 F. Zd. 1173 the United States Court of Appeals for the

Ninth Circuit had occasion to consider what constitutes a computer

program and in what circumstances copyright protection is extended

at p. 1175:

. . . A computer program is made up of several

different components, including the source and

object code, the structure, sequence and/or

organization

interface, and the function, or purpose, of

the program. Whether a particular component

of

the

program,

the

user

of a program is protected by a copyright

depends on whether it qualifies as an

"expression" of an idea, rather than the idea

itself.

It is settled law that once a computer program reaches the

point of source code or object code it represents a form of

expression which is subject to the protection afforded by

copyright.

Although what is protected by copyright is the expression and

not the idea, and this term is sometimes expressed as originality

the protection afforded does not depend on original ideas. This

seeming conundrum was addressed by Peterson, J. in University of

London Press Limited v. University Tutorial Press, Limited, [1916]

- 253 -

2 Ch 601 at 608:

Assuming that they

are "literary work," the question then is

whether they are original. The word

"original" does not in this connection mean

that the work must be the expression of

original or inventive thought. Copyright Acts

are not concerned with the originality of

ideas, but with the expression of thought,

and, in the case of "literary work," with the

expression of thought in print or writing.

The originality which is required relates to

the expression of the thought.

But the Act

does not require that the expression must be

in an original or novel form, but that the

work must not be copied from another work -

that it should originate from the author.

In addressing the question of originality in this context it

is not appropriate to dissect a work removing portions which may

not be considered original before examining what remains, this is

particularly so when the work is a form of compilation (such as

the ICD-9 codes) in the present case. It is important, however,

to remember one significant distinction, when you turn from the

consideration of the concept of originality of expression to

determine whether two works are substantially similar, the court

must exclude from its consideration the portions of the work which

are not properly the subject matter of copyright. The reason for

this narrower approach on this issue is the requirement that the

plaintiff demonstrate that his expression has been copied. The

ultimate question for consideration is whether the expression of

the author's idea has been taken.

- 254 -

The approach to be taken was well set out in the decision of

McLachlan, J., then of this court, in Slumber Magic Adjustable Bed

Co. Ltd. v. Sleep-King Adjustable Bed Co. Ltd. (1984) 3 C.P.R.

(3d) 81 at p. 84:

. . . The defendants suggest that there is no

copyright in the brochure because it used

ideas and elements which are also found in the

brochures

however,

copyright.

of

does

other

not

competitors.

defeat claim

That,

for

a

It is well established that

compilations of material produced by others

may be protected by copyright, provided that

the arrangement of the elements taken from

other

sources

is

the

product

of

the

It

plaintiff's thought, selection and work.

is not the several components that are the

subject of the copyright, but the over-all

arrangement of them which the plaintiff

through his industry has produced. The basis

of copyright is the originality of the work in

question.

So long as work, taste and

discretion have entered into the composition,

that originality is established. In the case

of a compilation, the originality requisite to

copyright is a matter of degree depending on

the amount of skill, judgment or labour that

has been involved in making the compilation:

Ladbroke (Football), Ltd. v. William Hill

(Football), Ltd., [1964] 1 All E.R. 465

(H.L.).

Where copyright is claimed in a

compilation it is not the correct approach to

dissect the work in fragments and, if the

fragments are not entitled to copyright, then

deduce that the whole compilation is not so

entitled; rather, the court should canvas the

degree of industry, skill or judgment which

has gone into the over-all arrangement.

Perhaps the most difficult problem in an infringement action

lies in the determination of when in the transition from idea to

expression in a computer program that line has been crossed and

- 255 -

the idea has been transformed into protectable expression.

4)

THE COPYING OF PRISM SOURCE CODE

The defendants have sought in a variety of ways to

explain portions of the evidence offered by the plaintiff's to

prove Mr. Imbert's copying. As I said earlier the test of

substantial similarity has a second component beyond the wording

of s. 3(1) of the Copyright Act. The law as it has developed in

the United States has applied a similar approach. In Broderband

Software Inc. v. Unison World Inc. (1986) 64 F. SUPP 1127 at p.

1135 it is expressed this way:

. . . Because direct evidence of copying is

rarely available, copying may be established

by proof of access and substantial similarity.

Roth Greeting Cards v. United Card Co., 429

F. 2d 1106, 1110(9th Cir. 1970).

This is the evidentiary test to which I earlier referred.

A

similar approach was taken by the United States Court of Appeals

for the Third Circuit in Whelan Associates Inc. v. Jaslow Dental

Laboratory (1986) 797 F. (2d) 1222 at p. 1231:

. . . As it is rarely possible to prove

copying through direct evidence, Roth Greeting

Cards v. United Card Co., 429 F. 2d 1106,

1110(9th Cir. 1970), copying may be proved

inferentially by showing that the defendant

had

access

to

the

allegedly

infringed

copyrighted work and that the allegedly

infringing work is substantially similar to

the copyrighted work.

The Whelan decision at p. 1236 proposed a method for

- 256 -

distinguishing between idea and expression in a computer program

... the line between idea and expression may

be drawn with reference to the end sought to

be achieved by the work in question. In other

words,

the

purpose

or

function

of

a

utilitarian work would be the work's idea, and

everything that is not necessary to that

purpose or function would be part of the

expression of the idea. ... Where there are

various

means

of

achieving

the

desired

purpose, then the particular means chosen is

not necessary to the purpose; and hence, there

is expression, not idea.

The Whelan decision has been controversial to say the least

and no small portion of the immense debate which has followed the

delivery of this decision has centered on this particular passage.

At the center of the debate lies the suggestion that the decision

fails to recognize the utilitarian nature of computer programs,

the respective roles of copyright and patent laws, and the

inappropriate application of the idea/expression distinction to a

consideration of a program's structure, sequence and organization.

Recent decisions in the United States, most notably Computer

Associates International, Inc. v. Altai, Inc. (1992) 23 U.S.P.Q.

(BNA) 1241 and Apple Computer, Inc. v. Microsoft Corp. (1992) 799

F. SUPP 1006 appear to signal a significant retreat from the

principles advanced in the Whelan decision.

In the present case the defendant Imbert was not only

afforded unlimited access to the Prism source code, but I have

- 257 -

found as a fact that he had both the source and object code

installed on the computer on which he developed the Q-PRO-4

"version" of the software.

It is clearly established on the

evidence, indeed it is not disputed, that Mr. Imbert had access,

but in so far as it is necessary I reject his evidence that this

access was limited in any way, and find that he closely consulted

and copied the code available to him.

The next question is whether or not the allegedly infringing

work is substantially similar to the Prism software.

In the

determination of what constitutes "substantial" it is not

necessary to prove a particular percentage or quantity of code,

indeed, in terms of numbers the amount copied may be small, that

affords no defence. In Breen v. Hancock House Publishers Ltd. et

al (1985) 6 C.I.P.R. 129 Mr. Justice Joyal of our Federal Court

found that:

. . . Although the proportion of cribbing from

the plaintiff's work to the total of the

author's work was quantitatively small, the

qualitative aspect indicated to me that it was

more

than

"fair

dealing"

and

that

it

constituted an appropriation by the author of

the skill and time and talent of the

plaintiff.

entitled to an injunction.

As a result, the plaintiff was

In the United States a similar approach emerged in SAS Institute

Inc. v. S. & H. Computer Systems (1985) 605 F. SUPP 816 where a

District Court found that forty-four examples of copying in

186,000 lines of source code was not trivial.

- 258 -

The approach to this issue is to assess the evidence on a

qualitative rather than a quantitative interpretation of the word

substantial. The area of judgment involves an assessment of the

work as a whole and a determination of the importance of the

portions of the work found to be similar.

The defendants submit that there is no way to translate

mechanically from Basic to Q-PRO-4 and go on to suggest that as a

result the evidence demonstrates ". . . that there was no literal

copying."

In support of this proposition they refer to the

following passage in the evidence of the Plaintiff's expert

Professor Dexter taken from a portion of his cross-examination:

QOn a line-by-line basis, there is no

easy way to copy BASIC into Q-Pro-

4, is there?

AIt would be analogous to copying English

into French, and it is not

a

literal

copying,

it

is

translation.

QBecause there is no -- there is no sort

of automatic program for converting

BASIC to Q-Pro-4?

ANo, there is not a precompiler, to the

best of my knowledge.

It is important however in assessing evidence to keep it in

context. In the series of questions and answers that immediately

follow these questions the context and opinion with respect to the

- 259 -

comparisons emerges:

QAll right. And would it be correct to

say that in your view, based on the

analysis you've done, what HMRI did

was to modify the Prism programs to

prepare a system in Q-Pro-4?

AWhen you say "modification,"

there --

a direct

there's very little modification.

There

was

virtually

translation when one looked at the

Prism system versus the first demo

system.

That looks like a almost

direct translation. There was some

modification by the time you got to

the version 2.0.

For example, the

seven screens have now matriculated

to nine screens.

That's what I

mean by "modification."

QAnd it's a modification in itself to

translate because you have to make

some changes --

AWell, it's a question of translation.

The translation is the difference

in the language but not difference

in the substance of the screens,

for example.

That doesn't -- I

mean, that seems to be, I would

say,

directly copied; the

translation

language.

clearly

a

different

QA translation is to do the same thing

but in a different language; is

that right?

AFunctionality, there is very little

difference even though one's in one

language and one is in the other

one.

same thing, the same function, they

are virtual analogues of each

other.

So in that case, to do the

QWell, in the same function, they both

collect

data

so

for

the

HMRI

mainframe,

does

manual

system. We have to get to a lower

- 260 -

level of function before we --

AYes, but the expression is really --

it's not the function that's, to

me, at issue here. To me it's the

expression

which

seems

to

be

directly copied.

That's the issue

to me that we are attempting to

address from this aspect of the

investigation.

QSo are you telling me then that the --

you

translation

there was

said

that

there

was

and

of

a

then

the

initially

modification

translation; is that right?

AThere was some modifications.

QSo what we are ending up with then after

the modifications, we have the

version 2.0, which you have shown?

A

Yes.

QSo, therefore, version 2.0 is a -- in

your view, modified -- HMRI

a

version 2.0 is, in your view, a

modified

programs?

version

of

the

Prism

AOnly marginally modified.

this very clear.

Let me make

To me, the

similarities of 2.0, which is the

most

different

system

that

I

observed, are so strikingly similar

that the concept of modification is

being

stretched.

They

are

marginally different. So there has

clearly been some changes.

Q

A

Some modification?

Very minor.

I have already indicated that I accept Professor Dexter's

opinions and prefer them over any conflicting opinion expressed by

- 261 -

Mr. Van Toen.

I have reached this conclusion not because of

personalities or manner, although that was a minor factor in the

consideration, but because on review Professor Dexter's depth of

analysis and technique of comparison is both compelling and

persuasive.

The absence of "literal copying" does not prevent the

allegedly infringing work from being either a "translation" or a

"reproduction" within the meaning of the Copyright Act.

In this case, if they stood alone, I would find Professor

Dexter's opinions persuasive on the issue of substantial

similarity. They, however, do not stand alone they are supported

by a body of evidence, which in my view demonstrates in this case

unequivocally that Mr. Imbert copied the plaintiff's computer

programs, and did so in a substantial manner.

I refer at this point to the "fingerprint" evidence offered

by Mr. Jeffery and the additional programming errors identified by

Professor Dexter as occurring in both systems.

I have earlier

dealt with these specific pieces of evidence in detail, I will not

repeat those conclusions here beyond saying that the reproduction

of all of these unique features in Mr. Imbert's work when taken

together with the other similarities are conclusive.

The minor

The

variations are, in my view, primarily cosmetic in nature.

- 262 -

errors evident in some aspects of Mr. Imbert's copying of the

Prism software far from demonstrating that he did not do so simply

demonstrate that, on occasion, he did so badly. Bad copying is in

the end still copying, I know of no requirement that it be

completely accurate before it is actionable.

Finally, there is in support of these conclusions the conduct

of the defendants surrounding the rewrite and the evidence

presented in this case.

In my respectful view, the court in

determining these issues is entitled to look to the conduct of the

parties at the time of the critical events and to weigh as part of

the process that conduct in determining the validity of "innocent"

explanations offered by the defendants.

The conduct of the

defendants at the time was both secretive and deliberately

deceptive, this was not the conduct of individuals seeking to

protect a legitimate trade or business secret but that of

individuals knowingly concealing unlawful activity.

Of perhaps even more significance is the evidence which I

have accepted which demonstrates that the defendants in the

persons of Mr. Imbert and Ms. King took active and positive steps

to conceal their activities and eliminate evidence (in particular

the wiping of the hard disk of the University Hospital computer).

When this evidence is combined with the way in which the

disclosure of Mr. Imbert's direct copying of Prism's ICD-9 codes

- 263 -

by means of the Q-PRO-4 program ICD9IN.PIC occurred at trial I

have little confidence that all of the evidence of Mr. Imbert's

activities has been disclosed during this trial.

I conclude that actual copying has been demonstrated in this

case, that access to the source code has been demonstrated and

that the programs are substantially similar on the positive

evidence available. Even if on the evidence before me I had not

reached the conclusion that the copying in this case is

substantial within the meaning of the tests I have described, the

conduct of Mr. Imbert and the other defendants and the evidence

that they concealed information likely to disclose their

activities justifies the drawing of an inference adverse to the

defendants on this point.

This copying, on the evidence which I have accepted, goes

beyond

a

reproduction

explainable

by

reference

to

the

specifications for data collection laid down by HMRI or required

by the structure of its mainframe programs, it extends into areas

where Prism's software collected data which was not a part of

HMRI's requirements, areas where data was collected either

differently or in excess of HMRI's requirements, and into areas of

program design and data storage which are both anomolous and

unique. In particular I refer to the five fingerprints I have

identified and the unique aspects of the handling by Prism of the

- 264 -

service transfer diagnosis.

Those unique aspects of Prism's solution to the idea of a

medical record abstracting system that met HMRI's data

requirements represent a form of expression entitled to copyright

protection.

In the circumstances of this case I find that the Prism

software represents a "solution" to the collection of medical

abstracts which is conceptionally and functionally unique and

therefore has the necessary originality to constitute a form of

expression protected by the Act.

In addition we are not dealing in the present case with the

conversion of the plaintiff's program into a machine language, but

into another "human" programming language, Q-PRO-4.

I find that

in this process the defendant Imbert created a "translation" of

the plaintiff's software within the meaning of s. 3(1)(a).

The defendants have suggested that even if some copying

occured in an early version of Q-PRO-4 software it was

subsequently removed and as a result the commercial version had

been "purged".

Even if I was inclined to accept such a suggestion as being

- 265 -

correct in law, the copying I have found extends well beyond any

particular portions which may have been removed from the

production or commercial version of the Q-PRO-4 software, and is

only demonstrated by the particular examples identified, not

limited to them. Professor Dexter's evidence and opinions go well

beyond those specific examples.

The decision relied upon by the defendants in advancing this

argument Blackie & Sons Ltd. v. The Lothian Book Publishing Co.,

[1921] 21 C.L.R. 396 appears never to have been considered or

applied by a court in Canada. In addition it is not a case

involving computer software. Even if it is good law with respect

to computer programs, in the circumstances of this case where the

defendant Imbert used Prism's work to enable him to produce a

replacement copy, subsequent minor changes will not, in my

respectful view, change the character of the initial taking and

the advantage derived from it.

Perhaps Sookman gives the best response to this particular

submission when he says at p. 3-279

Making modifications

or changes to the allegedly infringing work

will not negate antecedent copying, nor will

it reduce the extent of the reproduction to

anything

less

than

substantial,

if

the

modifications have been made to a work which

was copied from the copyright work and the

allegedly

infringing

work

substantially

reproduces the copyright work. The fact that

the defendant has added some additional

information or made some changes of his own

- 266 -

will not be a defence to an allegation of

infringement, and it has been well recognized

"that the most successful form of copying is

to employ enough points of similarity to

confuse the public with enough points of

difference to confuse the courts.

I find and conclude that both the demonstration version and

the subsequent production version of Q-PRO-4 Savoir-Faire are

reproductions and translations of Prism's software within the

meaning of s. 3(1)(a) of the Copyright Act.

5)

THE COPYING OF THE USER INTERFACE

The plaintiff's have urged on me the position that the

user interface or the "look and feel" of the Prism software can be

the subject of copyright protection either on its own or,

alternatively, as

a

part of the "structure, sequence and

organization" of a computer program.

I recognize the importance of such a finding in the area of

copyright law. As a result of the findings and conclusions I have

already reached a finding on this point is unnecessary to the

issues in this action.

In light of recent developments in the

United States which appear to signal a significant retreat from

the principles enunciated in Whelan (supra) it is best to leave

the analysis and decision of these points to a case in which they

are required for its determination.

- 267 -

This same consideration also makes it unnecessary for me to

determine whether or not Deanna Hickey's assistance or that of

Vancouver General Hospital in the design of SuperMACS had any

impact on the protection afforded to Prism's screen displays.

Given the findings and conclusions I have already expressed

coupled with the acknowledgment in writing by both the present

defendant HMRI and Vancouver General Hospital that Prism was the

owner of the programs and code, I need go no further in this

action.

6)

CONCLUSION

The defendants acknowledge in p. 587 of their written

submissions that "If HMRI is held not to have acted within its

license, and is held to have infringed copyright, then it is

acknowledged that Mr. Imbert and HMRI are jointly liable for any

acts of Mr. Imbert which infringe the copyright."

As these reasons make apparent I have found that HMRI did not

act within its licence and that Mr. Imbert by his actions

infringed Prism's copyright in the programs. Mr. Imbert and HMRI

are jointly liable to Prism.

XVI RELIEF

I have found the defendant HMRI in breach of the provisions

of Articles 5, 6, 10 and 14 of the licence agreement. In addition

- 268 -

I have found that both HMRI and the defendant Imbert by their

actions have infringed Prism's copyright in Prism Savoir-Faire and

as a consequence are jointly liable for damages flowing from that

infringement.

Mr. Imbert is not a party to the contract and

cannot be liable as a result of the breaches I have found. The

action as it proceeded before me did not proceed as against the

defendants John Stenabaugh and Lynne Abbott.

There is no evidence or basis in law to justify an award to

the remaining plaintiffs.

Prism seeks orders for prerogative relief, compensatory and

punitive damages, interest, and costs on a special scale.

The thrust of the prerogative relief sought is to compel the

removal of the infringing software from the market and to compel

its return to the plaintiff. Prism is entitled to this relief but

while some portions of this particular relief are easy to identify

others are more problematic. So far as possible the orders given

should be tailored to attempt to place Prism back in the position

they would have been in but for the breaches of contract and

copyright.

In achieving this it is necessary to recognize that

the infringing software has been marketed widely by HMRI and has

been installed in a large number of hospitals across Canada. Any

orders made must provide for an orderly transition with a minimum

- 269 -

of disruption to innocent third parties.

(1)There will be a declaration that Q-PRO-4 Savoir-Faire is

an infringing work which violates Prism's copyright.

(2)There will be an injunction issued prohibiting HMRI and

Imbert from further copying of either Prism's work or of

the infringing software.

(3)The terms of the prohibition last described will permit,

for a transitional period, HMRI to maintain the software

at its existing sites.

In the event the parties are

unable to agree on the necessary transitional period

either party may speak to the terms of such an

exception.

(4)There will be an order directing HMRI to deliver up to

Prism any and all infringing copies within their

possession, save and except for materials reasonably

necessary to enable them to maintain the existing

software for the transitional period.

HMRI is to

deliver up these materials forthwith and to provide

Prism with a list of any materials they retain for the

purpose mentioned together with a description sufficient

to specifically identify all such material in their

- 270 -

possession.

(5)There will be a further order directing the defendant

Imbert to deliver up to Prism any and all infringing

copies in his possession.

This is to include any and

all copies, paper or electronic, in his possession or

control.

(6)There will be an order directing that the infringing works

in hospitals, under license from HMRI be withdrawn over

a transitional period. Once again in the event counsel

are unable to agree there is leave to speak to the

details and time periods of the transition.

(7)HMRI will provide Prism forthwith with all necessary

information, particulars, and data to enable Prism to

convert

Q-PRO-4

Savoir-Faire

historical,

current

ancillary data files and any other necessary files to a

compatible SuperMACS format and will pay all reasonable

costs incurred by Prism to do so.

(8)HMRI will provide to Prism and/or a particular hospital,

at no cost to them, historical data from their mainframe

computer for that hospital if it elects to install the

Prism software and wishes access to its historical data.

- 271 -

The period provided will include that period for which

data was submitted from that site by the Savoir-Fare

program either Prism or Q-PRO-4.

(9)The defendant HMRI will pay all reasonable costs

associated with the changing of any existing Savoir-

Faire site to Prism SuperMACS. These costs will include

the costs of conversion, installation and training. It

may well be that these costs will as a result of

SuperMACS current incarnation include training in the

use of a report writer.

The bundling of the report

writer with the software occurred as a result of Prism's

attempt to survive in the market created by HMRI's

wrongful acts and in no small measure by the fact that

they were forced to compete in the market with a

"version" of their own software.

If that fact

necessitates some modest cost for additional training it

is one which flows from the defendants wrongful acts.

(10)There will be an order that HMRI immediately cease

marketing the infringing software.

(11)Upon the withdrawal of the infringing software from the

market all remaining Savoir-Faire materials including

user manuals, system diagrams, and source and object

- 272 -

code whether in paper or electronic form within the

control of HMRI is to be delivered to Prism forthwith.

(12)Where, to the knowledge of HMRI, copies, in paper or

electronic form, of the infringing material have been

delivered to individuals or organizations and are

outside HMRI's direct control they will forthwith

disclose to Prism the name, address and location of such

individual or organization together with particulars of

the materials in their possession.

In concluding this section of these reasons I wish to make

clear that the "reasonable cost" of converting both data and sites

will include compensation for the time of Prism's principals or

employees to carry out these activities. The fixing of a specific

fee for each site, as suggested by the plaintiff does not in my

view properly reflect the costs of the conversion process which

may in fact be minimal in some cases and substantial in others.

Finally I know of no principle which would permit me to or

warrant the imposition, as the plaintiff suggests of a form of

license fee based on existing sites electing to install a

different system. Such an order would seek, in effect, to impose

as a penalty what should properly be assessed as damages which

flow from one or more of the defendants breaches. It is proper,

- 273 -

however, to recognize that during any transitional period, the

continuing presence of an infringing work will be and remain a

legitimate concern of Prism's, I will therefore impose an

additional term during the transitional period.

(13)During the transitional period any and all communications

from HMRI, or on its behalf to a Savoir-Faire site

involving either maintenance of the infringing software

or its replacement will be communicated forthwith to

Prism.

XVII DAMAGES

In this action the plaintiff seeks to recover damage in a

number of categories including specific claims for lump sum fees,

losses of hardware sales, royalties (or per abstract fees), claims

for loss of future benefits or opportunities in the form of lost

sales of enhancements and add-on products and punitive damages.

On a factual basis many of the claims advanced do not fit

clearly within specific, traditional heads of damage.

The assessment is complicated further by Prism's financial

health at the time of these events. At the time of the initial

contact from HMRI although their software was installed in a

handful of major hospitals in British Columbia, the revenues

- 274 -

generated were not sufficient to support the principals who had to

avail themselves of other opportunities, including contract work

in Saudi Arabia.

Despite these difficulties Prism was not

inactive and it was pursuing its relationship with Vancouver

General Hospital and government agencies as means of promoting

both itself and its products.

The argument advanced by the plaintiff on this aspect of the

case is predicated on their assertion that the contract with HMRI

was Prism's big chance, the backing and market "clout" offered by

HMRI would allow Prism to not only break into the market on a more

significant scale than they had yet done but to do so nationally.

The difficulty is that Prism was not a healthy company which lost

a fixed and demonstrable number of sales to a specific competitor,

rather it was a company which saw its "big opportunity" in

securing a future market for enhancements and add-on products by

bargaining away and providing its abstracting system to HMRI who

would insure through its marketing position the installation of

its system ". . . in every Province and Territory in Canada."

Having written the software and produced the first of its

add-on products, a report writer which it delivered to HMRI for

review and testing, it was in a position to begin to capitalize on

the opportunity and the market.

It was at this critical point,

when the potential and the accuracy of their hopes and

- 275 -

expectations were to be tested that HMRI and Imbert through the

production of their "copy" of Prism's software turned HMRI's

marketing position and "clout" against their contractual partner

and the contractual obligations found in Article 14.

The defendants in their submissions, focus their attention on

Prism's modest success before the contract with HMRI, their

failure to produce additional add-on products during a time when

the production of Prism Savoir-Faire strained their limited

resources to the breaking point, and their failure to produce

these "future products" after the breakdown of their relationship

with HMRI left them fighting for their lives and against both

their own software and the very marketing "clout" that HMRI had

contracted to give them.

The defendants submit that there was no "crippling blow"

dealt to Prism and that it came out of the contract with HMRI with

software which was now available for marketing on a microcomputer

and with the edit requirements for eight provinces not just one.

While these submissions do touch on some aspects of the

situation existing between the parties after the breakdown of

their relationship they do not address the issue of what Prism

lost as a result of HMRI's wrongful acts and what damages flow

from them. It is no answer to a claim for breach of contract to

- 276 -

say I may have taken these advantages from you but I left you

those others.

There are two clear basis for awarding damage in the present

case, the first is on the basis of copyright infringement and the

second is breach of contract.

It is necessary to consider each

separately before determining the appropriate damage award.

Prior to turning to that consideration it is useful to set

out some general principles which affect the assessment of

damages. The burden of proof is on the plaintiff to show what its

losses are. In the circumstances of this case it is to a degree

speculative to embark upon the process of saying what would have

happened if the defendant HMRI had not breached its contract and

if Mr. Imbert had not copied the plaintiff's software, but the

burden of proof imposed on the plaintiff should not be so tightly

circumscribed so that the law loses sight of the fact that the

difficulty of proof arises in large part from the defendant's

wrongful acts.

In the assessment of damages the defendants are

entitled to be shielded from mere speculation, but in considering

what may have happened after September of 1986 in the now

impossible future, the defendant's cannot be allowed to hide

behind difficulties of proof arising from their own wrongful acts.

A.

COPYRIGHT

- 277 -

Damages for the infringement of copyright are generally

determined as those which the owner of the copyright work may have

suffered due to the infringement.

Damages are specifically

described in the Act and are to be found in s. 35(1) which

provides:

35.(1)

Where

any

person

infringes the copyright in any work that is

protected under this Act, the person is liable

to pay such damages to the owner of the right

infringed as he may have suffered due to the

infringement, and in addition thereto such

part of the profits that the infringer has

made from the infringement as the court may

decide to be just and proper.

The damages described are cumulative not alternative heads,

but, perhaps the most important feature underlying remedies

available in copyright is that it is a right of property and the

owner is, primarily entitled to have his property rights

protected. This is so even if the owner is unable to show actual

damage at which point he may only be entitled, in addition to any

protective orders, to an award of nominal damages.

The fact remains that the amount of damages in copyright

infringement is not necessarily derived by the actual damage

caused by the infringement. At p. 3-314 and 3-315 Sookman puts it

in these words:

. . . Damages for breach of copyright are at

large, and may be dealt with broadly and as a

matter of common sense without professing to

be minutely accurate.

Where the sum to be

awarded is under consideration, the plaintiff,

- 278 -

rather than the defendant, ought to have the

benefit of any doubt there may be as to the

amount of the award.

The wording and concepts expressed by the author in this

passage appear to originate in the language of Cozens-Hardy, M.R.

in Meters Ltd. v. Metropolitan Gas Meters Ltd. (1911) 28 R. P.C.

157 at p. 161:

. . . Therefore, in a case such as the

present, where licences are not granted to

anyone who asks for them for a fixed sum, it

is a matter which is to be dealt with in the

rough - doing the best one can, not attempting

or professing to be minutely accurate - having

regard to all the circumstances of the case,

and saying what upon the whole is the fair

thing to be done.

. . .

We must arrive at a

conclusion as best we can, not tying ourselves

down by any hard and fast rule, not requiring

the Plaintiffs to establish before us that any

definite number of retailers would have come

to the Plaintiffs if the Defendants had not

supplied infringing instruments, but dealing

with the matter broadly, and as best we can as

men of common sense.

In addition it is clear and well established that exemplary

damages are, in appropriate circumstances, available even in the

absence of monetary loss or derived profits. Once again Sookman

expresses the view, at p. 3-317 that "The amount of exemplary

damages should be commensurate with the gravity of the action

committed and should reflect the indignation of the court at the

conduct of the plaintiff." At the risk of being overly technical

- 279 -

I suggest the learned author meant "defendant" and suffered from a

momentary programming error.

Returning for a moment to the underlying nature of copyright

protection Robert Goff, L.J. wrote in Paterson & Co. v. Merfarken

Ltd. [1986] 3 All E.R. 525 at p. 538:

. . . The law of copyright is a self-contained

branch of the law, concerned with the

protection of a particular proprietary right.

The provisions regulating the protection of

that right, including, for example, those

which

provide

for

strict

liability

or

liability based on fault, are framed with

regard to a particular interest, viz the

proprietary right in question.

It would be

undesirable as a matter of policy to extend

the statutory remedies available for the

protection of that interest to the recovery of

damages in respect of the invasion of a

different interest.

In their submissions on damages the defendants assert that

"Profits which plaintiff's might have made had they obtained a

contract

which

the

infringer

instead

obtained

are

not

recoverable."

The defendant's cite three authorities for this

proposition British Insulated Wire Company v. Dublin United

Tramway Company (1900) 47 R.P.C. 14 at p. 21; Meikle v. Maufe

(1941) 3 All E.R. 144 at p. 154 and Paterson & co. v. Merfarken

Ltd. [1986] 3 All E.R. 522 at p. 538.

With the greatest of

respect, none of these decisions assert the general proposition

which the defendants advance.

In the first of these decisions the court found on the facts

- 280 -

that the plaintiff did not and may not have, in any event,

obtained the contract in question.

As a result they found that

the profit margin the plaintiff may have made from a contract they

neither entered nor performed was not the proper measure of

damages.

The heart of this decision is found in the passage

immediately following that cited by the defendants where the court

makes it clear that the defendant could have maintained its

agreement with the American supplier and simply not brought the

infringing cable into the country until after the date on which

the plaintiff's patent expired.

In Meikle a similar situation existed, the plaintiff in that

case argued that the measure of damages in a case where there was

an infringement of copyright in the plans for a building was the

fee he would have obtained if he had been hired as the architect.

The court acknowledged that the fact the copyright was held by

the architect was one of the factors to be considered but that the

proper measure of damages was not the fee for performing a

contract the architect never had, nor performed.

The Paterson decision is even further removed. This decision

in my respectful view, stands for the narrow proposition that a

person (in this case a printer) who infringes the plaintiff's

copyright is not liable to him for the use made of the infringing

material by

a

third party without his authority.

These

- 281 -

circumstances do not in my view support the general proposition

advanced nor do they have application in the present case.

The critical difference in the present case is that the

contract did exist and the infringing parties were one of the

parties to the contract and an employee of that company.

In

addition the defendants in this case used their contractual

relationship to enable them to both infringe the copyright and

capture the potential market provided to Prism.

The general measure of damages to be awarded in cases of

copyright infringement flow from the fact that it is a property

right, the specific wording of s. 35 and the general principles

that apply to the assessment of damages. What is recoverable by

the copyright holder are damages to compensate him initially for

all losses he suffered which flowed from the infringement. If the

evidence establishes the necessary causal link to the infringement

then a loss flows from it. If that loss is both foreseeable and

not too remote it is recoverable.

On the second level contained within the wording of s. 35(1)

an infringer may be, in appropriate circumstances, forced to

surrender all or a portion of the profits which they have received

from the infringing product.

- 282 -

What is different in the present case is that the

infringement took place within a contractual setting and that the

acts of infringement in this case also constituted breaches of the

contract between the parties.

Were it not for the difference

between the liability of Imbert for copyright infringement and

that of HMRI for both copyright infringement and breach of

contract it would perhaps be possible to approach the damage issue

on a more simplistic basis. It is important to recognize that in

this case the infringing acts are not simply the individual sales

of the infringing program itself but also the acts of Imbert in

copying Prism's program itself.

A portion of the reasons for judgment of Fletcher Moulton,

L.J. in the Meters Ltd. v. Metropolitan Gas Meters Ltd. decision

is both helpful and instructive in terms of the general principles

to be followed in these cases and the dangers inherent in

deviating from them. The passage beginning at p. 163 is lengthy

but it is worthwhile to set it out fully:

. . . The Plaintiffs have, I think, done their

duty to the Court in putting forward all the

elements which enable the Court to judge of

the magnitude and nature of the infringement.

I take it that this is all they can be

required to do.

The defendants seek to

diminish

the

damages

by

to

a

variety

that

of

the

affidavits

particular

intended

purchasers

these

show

for

whom

they

were

manufactured

infringements

customers who would not have purchased from

the Plaintiffs if they had not purchased from

them. I am not for a moment going to say that

evidence of that kind may not be relevant, but

the argument based upon it was, that where a

- 283 -

plaintiff proves the sale of infringing

instruments by the defendants he does not

establish any right to damages unless he shows

how many of those particular instruments would

have been purchased from him if the defendant

had not sold them; and the Counsel for the

Defendants were bold enough to say that in

this case of infringement on a large scale

there ought to be only nominal damages. The

estimation of damages in cases of breach of

contract,

or

cases

of

tort,

is

often

necessarily a most difficult matter. In some

cases of contract, as is the case of the sale

of articles capable of being purchased in the

open

market,

there

is,

under

ordinary

circumstances, a method so obviously excellent

of measuring the damages - measuring the

injury done to the plaintiff by the breach of

contract - that that method of estimating them

has almost become a rule of the law. As soon

as you leave that type of case the assessment

of damages depends upon circumstances so

various that in most instances there is no

rule of law other than the very idea and

principle of damages - that is an equivalent

for the wrong done to the plaintiff.

It is

very dangerous to allow secondary rules to

come in and be treated as rules of law unless

you are certain that, whatever be the

particular circumstances of the case, the

nature of the breach, or the nature of the

wrong is such that these secondary rules will

always do justice between the parties.

The

Defendants have set up here - the burden of

proof is on them - that there is this

secondary rule of law, that where a defendant

has sold infringing articles the plaintiff can

only recover damages in respect of those which

he can show would have been bought from him,

if the defendant had not infringed.

In my

opinion there is no such secondary rule. I am

quite aware that a good practical method of

arriving at a fair estimate of the wrong done

may, in some cases, be by forming a conception

of how many sales of a particular article

would have been made by the plaintiff, and

then giving him the full manufacturing profit

for that proportion, but there is no rule of

law which requires the Court to do that in all

circumstances, and there are innumerable cases

in which such a rule of law would be quite

- 284 -

inapplicable.

Once again returning to the general principle it is aptly

summarized by Buckley, L. J. in the same decision at p. 165:

. . . The matter to be ascertained is the

pecuniary equivalent of the injury which the

Plaintiffs have sustained by the wrongful acts

done by the Defendants. In ascertaining that,

we are not to include every injury, however

remote; we have to find the pecuniary

equivalent of the injury resulting as the

natural consequences of the acts done by the

Defendants.

In order to answer that inquiry

the proper mode is, I think, to ascertain, as

far upon the evidence as it is possible to

judge, the extent to which the trade of the

Plaintiffs has been interferred with by the

acts of the Defendants.

The facts of this case involve the defendant HMRI placing the

infringing work written by the defendant Imbert at 163 hospital

sites in Canada, as at the date of trial. At approximately 80 of

those sites they placed copies of a product called Q-Query which

was a commercial report writer which served the same or similar

purposes as the report writer developed by Prism and provided to

HMRI for testing.

There is no suggestion that Q-Query was an

infringing work or that HMRI was restrained by a contractual term

from offering to its hospital sites some other report writer.

It is not correct to say, however, that HMRI is not in the

end responsible to account for profits earned through the supply

of the Q-Query report writer. The Q-Query product was provided by

- 285 -

HMRI to add functionality and features to the infringing product

and to manipulate the data collected by the infringing work. In

this sense the product formed a part of HMRI's marketing strategy

for the infringing product and represented both a profit they

earned from the infringement and a portion of the appeal of the

infringing work itself.

In addition, in so far as the actions of HMRI in marketing

the infringing work forced Prism to reduce its price to compete

that reduction is recoverable as flowing directly from the

infringement.

I can see no meaningful distinction between

reducing the price of the copyright work and adding features which

you would otherwise have charged separately for to achieve the

same result.

In this case the plaintiff bundled with its

SuperMACS software both its Report Writer and a package of

specialized reports. These two items are "extras" which are the

equivalent of a price reduction forced on the plaintiff by HMRI's

infringement.

"reduction".

They are entitled to compensation for that

I reject the defendant's submission that the plaintiff is

restricted by any principle of the assessment of damages to the

"paid up" lump sum license fee and the per abstract royalties.

Such an interpretation of copyright protection has the effect, in

the present case, of limiting the damages recoverable for

- 286 -

copyright infringement to the benefits directly receivable by the

plaintiff under contract while permitting the defendant to breach

the contract with impunity.

Those factors which forced the plaintiffs to effectively

lower their price also restricted their market for their add-on

products, and, in my view, are causally linked and flow directly

from the infringing activity of the defendant HMRI and Imbert.

These items, as such are recoverable in copyright against both

HMRI and Imbert.

These categories are succinctly summarized by

McLachlin, J.A., then of our Court of Appeal in Houweling

Nurseries Ltd. v. Fisons Western Corp. (1988) 49 D.L.R. (4th) 205

at p. 210:

In my view, the law

may be summarized as follows. The basic rule

is that damages for lost profits, like all

damages for breach of contract, must be proven

on a balance of probabilities.

Where it is

shown with some degree of certainty that a

specific contract was lost as a result of the

breach, with a consequent loss of profit, that

sum should be awarded.

However, damages may

also be awarded for loss of more conjectural

profits, where the evidence demonstrates the

possibility that contracts have been lost

because of the breach, and also establishes

that it is probable that some of these

possible contracts would have materialized,

had the breach not occurred. In such a case,

the court should make a moderate award,

recognizing that some of the contracts may not

have materialized had there been no breach.

The matter may be

Even though the plaintiff

put another way.

may not be able to prove with certainty that

it would have obtained specific contracts but

for the breach, it may be able to establish

- 287 -

that the defendant's breach of contract

deprived it of the opportunity to obtain such

business.

The plaintiff is entitled to

compensation for the loss of that opportunity.

But it would be wrong to assess the damages

for that lost opportunity as though it were a

certainty.

The unique and distinguishing feature in the present action

is that the acts of infringement and the licensing of the

infringing work at 163 sites including those at which the

plaintiff's work was already installed are inextricably and

inherently tied to the defendant HMRI's breach of its contractual

obligation to market the plaintiff's software, in my view, with

the exceptions I will set out, the damages flowing from the

copyright breach in this case, mirror those arising in contract.

Two other features of HMRI's marketing are significant. The

evidence establishes that the employees of HMRI were severely

overloaded in their responsibilities and workloads.

This was a

major factor in the suspension of installations and the way in

which those installations were scheduled. A side effect of this

was that an undetermined number of installations were delayed for

periods of time not easily ascertainable on the evidence. As the

contract required HMRI to pay Prism fees for each abstract

submitted by means of their software for a five year period

commencing on April 1, 1986 and ending on March 31, 1991 this had

the effect of reducing the "royalty" payments due to Prism by

- 288 -

pushing some of the abstracts out of the five year period, eg. a

three month delay in the installation of site "A" caused by HMRI

and due to factors unrelated to problems with the software itself

would mean three months less of abstracts submitted within the

royalty period.

This is not an insignificant factor when one considers that

much of HMRI's activities from October of 1986 until late 1987

were aimed at preparing the Q-PRO-4 product and that this involved

the person, Wendy King, responsible for those installations. The

installations actually began in Alberta in April 1987, but

although HMRI continued to install Prism Savoir-Faire at some

sites while awaiting the availability of the particular Q-PRO-4

software it is a fair inference on the evidence that the number of

installations and their promptness suffered significantly because

of the development of the infringing software.

It is equally

probable that these factors significantly reduced the number of

abstracts submitted on Prism's software during the royalty period.

The other feature of the defendant HMRI's marketing efforts

which is of significance in assessing the importance of that

effort, if the contractual commitment had been kept, to Prism is

the fact that HMRI used its abstract fee as a tool to encourage

the conversion of hospitals from paper to electronic submissions

by pricing the cost of submitting paper abstracts well in excess

- 289 -

of that for electronic submissions.

B.

BREACH OF CONTRACT

The leading case in this area is Hadley v. Baxendale

1854) 9 Ex. 341, 156 E.R. 145. In addressing the issue of damages

the court set up tests based on foreseeability and found the

contract breaker liable for such damages as:

. . .may fairly and reasonably be considered

either arising naturally, i.e., according to

the usual course of things from such breach of

contract itself, or such as may reasonably be

supposed to have been in the contemplation of

both parties, at the time they made the

contract, as the probable result of the breach

of it.

Now, if the special circumstances

under which the contract was actually made

were communicated by the plaintiffs to the

defendant, and thus known to both parties, the

damages resulting from the breach of such a

contract,

which

they

would

reasonably

contemplate, would be the amount of injury

which would ordinarily follow from such breach

of contract under these special circumstances

so known and communicated.

In my respectful view, in contract the change in the

programming language and the change in the way the Q-PRO-4 program

stored its data were changes which impacted directly on Prism's

ability to compete and market its products both existing and

intended.

contemplation of the parties and were in fact the subject of

documentation within the proposal itself. Whether or not those

aspects of the matter represented a contractual obligation of

These prospects were certainly reasonably within the

- 290 -

HMRI's is not relevant in my view. HMRI knew of Prism's interest

and intention, and by their wrongful acts they deprived them of

the opportunity to obtain that business. The present case, in my

view, falls squarely within the principles enunciated in the

passage from Houweling Nurseries Ltd. v. Fisons Western Corp. to

which I earlier referred.

C.

SPECIFIC DAMAGE AWARDS

It is not possible in this case to measure Prism's

losses with precision, at times it is difficult to even categorize

some items within traditional heads of damage. What must be done

is to approach the matter by attempting so far as possible to

place Prism in the position they would have been in but for the

defendants breaches of copyright and contract.

In attempting to

of the

do so it is necessary, because of the very uncertainty

process to build into these awards a recognition of the many

contingencies.

i)

LUMP SUM FEES

On page 18 of the proposal it was agreed that:

. . . Prism will provide each of the remaining

Provincial and Territorial versions on an as

required basis.

The conversion fee for each

additional version will be dependant upon the

complexity relative to the current product,

however, the cost in most cases will be the

minimum charge of $10,000 (which will increase

at the rate of 10% per year after the end of

1986 fiscal year).

Specifications will be

based on the latest published HMRI manual as

appropriate for each version.

- 291 -

In addition on page 3 the Proposal includes a provision that:

The delivery schedule for other Provincial and

Territorial versions has not been established

and is subject to negotiation.

Prism seeks to recover the lump sum fees for each of the

additional provinces and territories. This claim is a matter of

construction of the proposal.

It is to be noted that on the

evidence HMRI notified Prism of its intention to prepare software

for Alberta and New Brunswick in correspondence dated November 5,

1986. The wording of that notice is important:

What we are now saying, is that since we have

hired staff to maintain your software, we wish

our own personnel to develop the Alberta and

New Brunswick versions from which you will

derive royalty payments.

Despite the fact that Imbert is over a month into the rewrite

Prism is being told he was hired to maintain [Prism's] software.

The key point, however, is HMRI advising Prism that they wish

their own personnel to develop the provincial "versions" and that

royalties will be paid.

The wording of the proposal is important, firstly it uses the

phrase "Prism will . . . provide . . . as required . . ." and

secondly that there will be "a minimum charge of $10,000 . . ."

The actual fee for conversion, however could vary based on

- 292 -

complexity.

I find that the proper construction of the agreement is that

HMRI received in the contract a license for all of Canada, but

that they were to pay a "minimum charge" for "each additional

version." That charge is distinguished within the proposal from a

"conversion fee" which would depend on complexity.

In my view

what was intended was that HMRI would pay a minimum charge of

$10,000 for each additional version. The suggestion that this is

to be construed as meaning that the payment is only necessary if

Prism is "required" by HMRI to produce it is simply not in accord

with the language used.

A new version is to be produced "as

required" HMRI required it in the case of both Alberta and New

Brunswick but chose to fill their need with the infringing Q-PRO-4

software.

The evidence before me does not establish that new versions

were required for the Yukon or Northwest Territories and the

evidence, such as it is, appears to indicate that they use the

Alberta -CM version. In my view the minimum charge is payable for

a new version of the software to, in effect, license the "engine"

of the plaintiff's software, it is not payable for each separate

territory even if a different version is not required.

The

evidence does not establish that a new version has been prepared

for Manitoba.

- 293 -

The plaintiff Prism is entitled to recover from HMRI, in

contract, the sum of $10,000 each for the Alberta and New

Brunswick versions of Savoir-Faire together with an additional sum

of $2,000 representing a 20% increase for the New Brunswick

version which was required in 1988.

Under this category Prism

will recover the sum of $22,000 together with interest at the

Registrar's rates.

ii) HARDWARE SALES

For a period of time in 1986 and 1987 HMRI agreed to and

did recommend Prism to its client hospitals as a preferred

hardware vendor for hospitals west of the Ontario/Manitoba border.

This agreement did not form a part of the main contract between

the parties and was not reduced to writing.

The defendants say

that there was no consideration given for this oral agreement and

that it was both unenforceable and incapable of supporting a claim

for damages.

In the alternative the defendants say that this

agreement was terminable at will, and at most given the length of

time (approximately one year) it had been in place, only a 30 day

notice was required.

Prism says that the termination of this status resulted in

the loss of hardware sales and arose from the breakdown of the

relationship of the parties which flowed in turn from the decision

- 294 -

of HMRI to rewrite Prism's software.

The essence of the

plaintiff's position is that quite apart from any binding

contractual obligation HMRI knew Prism was supplying hardware to

Savoir-Faire sites and that when they chose for reasons associated

with their decision to produce infringing software to drop Prism

as a recommended hardware supplier Prism would foreseeably lose

hardware sales and the opportunity to make them.

It is, necessary at this point to turn briefly to a

consideration of the expert evidence led with respect to damages

generally. There were three experts called by the parties and one

of the principals of the plaintiff, John Jeffery was permitted to

testify in the area of the marketing and sale of microcomputer

systems for Health Records applications.

The first of these witnesses was Robert D. McKay an expert in

the area of business valuation and Income Loss.

Mr. McKay's

engagement was a limited one in which he was asked ". . . to

review Prism's claim for income loss and summarize and comment on

the method by which it was calculated."

The limited nature of

this engagement left the court with an expert whose opinion deals

with the methodology adopted by Prism.

The remaining two experts Mr. Ronald Johnson and Earl

Bardswich were qualified as having expertise in the area of

- 295 -

hospital related computer systems and software and, in the case of

Mr. Bardswich, in the area of the market for hospital computer

systems.

Mr. Johnson was a witness whose major experience lay in the

use of computer systems and software at hospitals in the United

States. Mr. Johnson did provide evidence with respect to Prism's

proposed pricing structure for its add-on modules by comparing

them to the similar products available in the United States.

Mr. Bardswich, was by contrast a Canadian expert with

experience in the Canadian market.

Once again it is necessary to deal with a bias allegation

advanced by the plaintiffs in relation to an expert witness called

by the defendants.

Mr. Bardswich was employed by the Ontario

Ministry of Health in 1987 and 1988 and during that time he was

directly involved in the administration of the $2 million grant

provided by that Ministry to HMRI.

When he was originally

approached by counsel for the plaintiffs, Mr. Chaster spent some

time reviewing at least some aspects of the plaintiff's claim for

damages.

This approach occurred at a time before Mr. Bardswich

was retained by HMRI. Mr. Bardswich declined to become involved

with the litigation indicating he had a relationship with HMRI and

that he might be seen to be biased.

- 296 -

I am troubled by Mr. Bardswich's appearance and his apparent

reluctance to admit the concern about the bias he originally

expressed. Mr. Bardswich was not an impartial expert, but one who

appeared to have an agenda which he pursued in his evidence. I am

also concerned about his role in the administration of the $2

million grant at a crucial time in the relationship between these

litigants.

It is also relevant to a consideration of Mr.

Bardswich's partiality that, as noted earlier HMRI opposed the

plaintiff's application for the release of material from the

Ministry of Health concerning the grant he was administering for

the Ministry, and as a result documents which may have disclosed

something of Mr. Bardswich's activities at a crucial time were not

produced.

In summary I am not impressed with the evidence of Mr.

Bardswich.

More importantly, the evidence of these experts does

little to provide me with a framework of assumptions which could

provide a method for the mathematical calculation of damages in

the present case.

I am satisfied that but for the breaches by HMRI Prism would

have continued to have been a supplier of computer hardware for

hospital sites in Western Canada installing the Savoir-Faire

system, and that HMRI's removal of Prism's status as a recommended

- 297 -

or preferred hardware vendor both arose from the infringement and

impacted on Prism's ability and opportunity to obtain that

business. The way in which HMRI chose to make that announcement

to its client hospitals prior to any notice to Prism served to

maximize its impact and to minimize Prism's ability to respond.

In its analysis of the evidence Prism estimated that:

(1)Of the 59 installed Savoir-Faire sites

west of Ontario they would have

sold hardware to 50% or 27 after

deducting the two sites they had

already supplied with hardware.

(2)Of the 198 sites which purchased

systems

other

than

Savoir-Faire

Prism suggests that but for the

difficulties and problems arising

with the Q-PRO-4 version, delays in

installations and a half hearted

marketing effort by HMRI 99 of

these sites would have selected

Savoir-Faire and 49 of these would

have purchased hardware from Prism.

(3)AS of March 31, 1990 Prism estimated

- 298 -

143

that

of

the

sites

still

reporting to HMRI on paper 72 were

located in the West and would have

converted to Savoir-Faire and one-

half of these would have purchased

hardware from Prism.

(4)As of April 1, 1986 there were 115

sites reporting electronically to

HMRI which did not convert to Q-

PRO-4 Savoir-Faire, Prism estimated

that

38

of

these

would

have

converted to Savoir-Faire, that of

these 19 were in the west and that

of those they would have made

hardware sales to 10.

(5)Finally Prism while not attempting to

quantify

pointed to the existence of some

450 hospitals which did not

this

final