SAS Institute, Inc. v. S & H Computer Systems, Inc.

605 F. Supp. 816 (1985)

SAS INSTITUTE, INC.
v.
S & H COMPUTER SYSTEMS, INC.

Nos. 82-3669, 82-3670.

United States District Court, M.D. Tennessee, Nashville Division.

March 6, 1985.

John M. Conley, Robinson, Bradshaw & Hinson, P.A., Charlotte, N.C., Karen L. Casser, Gen. Counsel, Cary, N.C., James J. Marcellino, J. Thomas Franklin, Dexter L. Kenfield, Gaston Snow & Ely Bartlett, Boston, Mass., Bass, Berry & Sims, Nashville, Tenn., for SAS Institute, Inc.

James V. Doramus, Doramus, Gideon & Trauger, Nashville, Tenn., Joseph M. Potenza, Schuyler, Banner, Birch, Mckie & Beckett, Washington, D.C., William Ramsey, Neal & Harwell, Nashville, Tenn., for S & H Computer Systems, Inc.

MEMORANDUM

WISEMAN, Chief Judge.

This civil action represents the consolidation of cross actions filed by the parties. SAS Institute, Inc. [Institute] in its action seeks relief against S & H Computer Systems, Inc. [S & H] based on claims of copyright infringement, breach of contract, fraud, trade secret misappropriation, and unfair competition, relating to the preparation by S & H of a computer software system for performing statistical analysis which was allegedly copied and derived from the Institute's copyrighted and widely marketed system, known by its registered trademark "SAS." S & H for its part seeks declaratory judgment with respect to the copyright and breach of contract claims, and seeks recovery against the Institute on claims of breach of a separate contract entered into by the parties in an effort to settle this dispute, and for interference with a contract between S & H and a third party. Jurisdiction of this Court is based on diversity of citizenship, federal question, and 28 U.S.C. § 1338, respecting copyright and unfair competition claims. State law claims are also asserted, based on pendent jurisdiction. The jurisdiction of this Court is not disputed.

The case was tried without a jury on September 12 through 20, 1983, with the Institute proceeding as plaintiff. This memorandum sets forth the Court's findings of fact and conclusions of law. To the extent that findings of fact actually represent conclusions of law, they should be so considered, and vice versa.

Because the Court finds in favor of the Institute on both the copyright and contract claims, and grants the requested relief, the Court declines to address the trade secret, Lanham Act, fraud and pendent state claims.

The Court requested proposed findings of fact and conclusions of law from the parties, and has adopted many of those submitted, verbatim, as the findings of fact and conclusions of law of the Court.

At the request of the Court, the parties agreed to the appointment of an expert to assist the Court in this case and expressly waived the disclosure requirements of Rule 706, Fed.R.Evid. The Court acknowledges with gratitude the invaluable assistance of Dr. Robert I. Winner, Ph.D., the expert appointed under this agreement.

FINDINGS OF FACT

1. Plaintiff SAS Institute, Inc. is a North Carolina corporation which markets a computer programming system for performing statistical analysis under the registered trademark "SAS." SAS presently operates only on IBM computers, and IBM-compatible computers offered by several other companies.

2. The Institute is testing a version of SAS to operate on VAX computers manufactured by Digital Equipment Corporation [DEC]. The Institute decided to develop a VAX version of SAS in response to SAS user desires. The Institute annually distributed "software ballots" to its users, seeking suggestions for enhancements to SAS. After the desire for non-IBM versions of SAS appeared as a strong user desire for two successive years, the Institute in 1981 commenced such a project. The VAX version of SAS has been distributed to over 200 so-called "beta test" sites. Beta testing is the final phase of operational testing prior to actual production release.

3. SAS has been marketed over the years in a series of successively more advanced editions or "releases," each identified by a unique number. This action primarily involves SAS Release 79.5. SAS 79.5, released in March, 1981, was the result of an extensive development process growing out of earlier SAS releases, particularly SAS 76.2, released in July, 1976.

4. The development process which led to SAS 79.5 required approximately five years, and more than 18 man-years of labor. Exhibit P-38. This process included both the addition of new procedures and major enhancements to existing procedures in SAS. One of the Institute's expert witnesses, Dr. David Peterson, testified that approximately 67% of the lines of computer program source code contained in SAS 79.5 are "new," that is, not carried over from SAS 76.2. S & H's only expert, Alan Merten, agreed with the substance of Dr. Peterson's testimony in this regard, his estimate of the number of lines of new source code (70%) differing by at most a few percentage points.

5. The source code for a computer program is the series of instructions to the computer for carrying out the various tasks which are performed by the program, expressed in a programming language which is easily comprehensible to appropriately trained human beings. The source code serves two functions. First, it can be treated as comparable to text material, and in that respect can be printed out, read and studied, and loaded into a computer's memory, in much the same way that documents are loaded into word processing equipment. Second, the source code can be used to cause the computer to execute the program. To accomplish this, the source code is "compiled." This involves an automatic process, performed by the computer under the control of a program called a "compiler," which translates the source code into "object code," which is very difficult to comprehend by human beings. The object code version of the program is then loaded into the computer's memory and causes the computer to carry out the program function.

6. Although S & H argued that lines of new source code might have represented trivial or functionally insignificant alterations, the Institute's president, Dr. James Goodnight, testified that SAS source code was not changed except for significant functional purposes. Both Dr. Goodnight and Dr. Peterson testified that every functional change or enhancement in a computer program must be accomplished by the addition or insertion of new source code. The Institute's evidence that substantial new portions, including totally new procedures and new blocks of source code in existing procedures, were added in SAS 79.5 was not rebutted, and is accepted by the Court.

7. The Court finds that SAS 79.5 contained numerous and substantial additions, enhancements, revisions, and other new material not contained in SAS 76.2, and that this material taken as a whole qualified SAS 79.5 as an "original work of authorship" under the Copyright Act of 1976, as amended.

8. The Copyright Office of the United States Library of Congress issued copyright registration no. TXU 96-620, effective June 23, 1982, which on its face is applicable to SAS Release 79.5 and lists the Institute as the owner of copyright. Exhibit P-2. This copyright registration has not been revoked or otherwise altered.

9. The Institute does not sell copies of SAS. Rather, it markets SAS through individual contracts with each customer, under which the customer is licensed to use SAS in accordance with the terms and conditions of the license agreement. This case primarily involves a license agreement between the Institute and S & H, a Tennessee corporation located in Nashville, Tennessee.

10. The events leading to the licensing of SAS by S & H were initiated by a small group of faculty and staff at Vanderbilt University in Nashville. Charles Federspiel, a professor of biostatistics, was engaged in early 1981 in research for which he required sophisticated statistical computing capabilities, such as are provided by SAS. After very brief investigation, Federspiel learned that SAS was available only for IBM or IBM-compatible computers. Vanderbilt did not have such computers. Federspiel's department did have a DEC VAX computer, on which SAS could not be executed.

11. Federspiel did not investigate the cost or feasibility of acquisition of IBM or IBM-compatible computers by Vanderbilt, nor did he investigate the feasibility of obtaining the use of SAS from another computer facility. In fact, SAS is widely available, both from commercial computer timesharing companies, and from other university computing centers. Exhibit D-11, p. 4. Federspiel made no efforts to investigate any other way of obtaining the use of SAS.

12. Federspiel discussed his desire to obtain the capabilities of SAS with two of his colleagues, William Vaughn and Wayne Ray, as well as with a part-time employee of the Vanderbilt University Computer Center, Phillip Sherrod. As Federspiel was aware at the time, Sherrod was also employed by S & H, and soon thereafter, Sherrod left the employ of the University to work full-time for S & H.

13. This group, together with Harry Sanders, the president of S & H, discussed the desirability of developing and marketing a statistical analysis package having capabilities comparable to SAS. Sherrod and Ray took some preliminary steps toward the design of a "state of the art" package, which they named PASQUEL. Their intent was that PASQUEL be an original work, not based on or similar to any existing statistical package. However, after determining that marketing such a package would be difficult, as compared with a package modeled after SAS, the PASQUEL effort was abandoned in the summer of 1981.

14. Soon thereafter, the decision was made to develop a conversion of SAS that would operate on VAX computers. The new product would be "portable," that is, it could be transported to and executed on computers other than the IBM equipment for which SAS was designed. Since it was then their understanding that the Institute had no interest in developing non-IBM versions of SAS (the Institute having not yet announced its upcoming VAX version of SAS), it was their hope that once their conversion was complete, the Institute would embrace it, and join in joint marketing, user education, and support efforts.

15. Federspiel, Vaughn and Ray each desired to benefit financially from their participation in the SAS conversion project. However, they were employed full-time by Vanderbilt University, and thus could not become employees of S & H. They therefore organized a Tennessee limited partnership under the name "Portable Statistical Analysis Systems, Ltd.," sometimes referred to as "PSAS, Ltd." Exhibit P-3. Federspiel, Vaughn and Ray were the limited partners in the partnership, and S & H was the general partner.

16. The partnership then executed a Development Agreement with S & H. Exhibit P-4. Under the Agreement, S & H was to develop "a set of computer programs, which provide integrated statistical analysis similar to an existing statistical system known as `SAS' (Statistical Analysis System)," following which the partnership would purchase all rights to the product for a price slightly more than 114% of development costs, to be financed by S & H. The result was that proceeds from the marketing of the portable SAS conversion would go first, to S & H until it had recouped its development costs plus a profit, and thereafter, to the partnership and through it to Federspiel, Vaughn and Ray.

17. Immediately after deciding to abandon the PASQUEL project, S & H decided to obtain a license to use SAS. S & H's primary purpose in obtaining the license was to obtain detailed technical information not otherwise available for use in the preparation of its conversion of SAS to run on VAX computers.

18. Contemporaneous notes prepared by S & H indicate that it required SAS source code for its project, which is referred to in the notes as a "conversion-fully compatible." Exhibit P-19. The notes also show that S & H planned to seek from the Institute all of the source code for SAS, additional technical assistance, and marketing and publication assistance. According to the notes, S & H intended to seek "exclusive rights to [the] VAX version," and "royalty on use of converted code for other machines." These desires were never communicated to the Institute.

19. Even before licensing SAS, S & H had made preliminary contact with DEC, the manufacturer of the VAX, with respect to its planned statistical package. S & H first contacted DEC employee Stephen Nagy in the spring of 1981.

20. Mr. Nagy testified that S & H contacted him again in September or October, 1981, after S & H had licensed SAS. At that time, S & H described its product as a "conversion" of SAS, and that that it would release the product to the market in the spring of 1982. Nagy expressed his concern as to the legality of S & H's product, and S & H responded that it had received legal advice that its conduct was proper. S & H has never disclosed the source or content of any such legal advice.

21. In January, 1982, Nagy told S & H that DEC would not consider any arrangement with respect to the S & H product until its legal status was clarified. There has been no subsequent contacts between S & H and Nagy or DEC.

22. S & H presented testimony at trial, including but not limited to testimony by Mr. Sherrod and Mr. Ray, that its intent at the time of acquiring the license was solely to execute SAS in accordance with the license, running statistical analyses on test data, so that the results produced by SAS could be compared with the results produced by the S & H product. However, undisputed evidence discussed below demonstrates that immediately upon receiving SAS, S & H used it in ways inconsistent with this trial testimony. In view of these facts, as well as the demeanor of the witnesses on the stand, the Court disbelieves the S & H testimony with respect to its purpose in licensing SAS. The Court concludes that S & H's purpose in licensing SAS was to obtain SAS source code for use in the preparation of its product, in the manner discussed below.

23. With a view to obtaining a SAS license, S & H contacted the Institute. Its purpose in acquiring the license was not disclosed. This fact was material, and had S & H disclosed its true purpose, the Institute would not have licensed SAS to S & H.

24. In its first communication with the Institute, S & H specifically inquired as to whether SAS materials distributed to licensees included the source code for SAS.

25. In August, 1981, the parties executed a license agreement under which S & H was authorized to use SAS, but only on a specifically identified IBM computer located at Tennessee State University in Nashville. Exhibit P-1. S & H paid a $4,500 license fee. The license agreement prohibited S & H from further distribution of SAS, from allowing timesharing use of SAS, and from making copies of SAS except for backup purposes. The license agreement permitted S & H to modify SAS, but only for its own use, and to employ SAS as part of an updated work, but no redistribution of such an updated work was allowed. The license agreement granted to S & H only those rights enumerated in the agreement.

26. At the time S & H executed the license agreement, it intended to use SAS in ways contrary to the terms of that agreement (as it subsequently did). That intention was not disclosed to the Institute, and had it been disclosed, the Institute would not have licensed SAS to S & H.

27. Shortly after executing the license agreement, S & H received the SAS distribution package. This included a magnetic tape containing the complete object code for SAS, and the source code for those portions of SAS which perform the various statistical analyses, as well as for portions which constitute the "linkage" between the supervisor and statistical procedures. SAS is divided into three parts consisting of statistical analyses, linkage and supervisor. The statistical analyses and supervisor form a very high fraction of the total and they are of approximately equal size. The SAS distribution package did not include the supervisor portion of the source code. Immediately upon receiving the tape, S & H made a backup copy of the tape, as authorized by the license agreement.

28. S & H then transported one of the tapes to Vanderbilt University, where its contents were loaded onto the VAX computer located in Federspiel's department. Because SAS could not be executed on the VAX, only those portions of the tape containing source code for SAS were loaded. However, VAX computers store text by means of a different scheme of numeric coding (acronym "ASCII") than that used on IBM computers (acronym "EBCDIC"). Hence, in order to load the SAS source code onto the VAX, S & H found it necessary to translate from the IBM coding scheme to the VAX coding scheme. The result of this process was the creation on a disk storage device attached to the VAX computer of an exact copy of the SAS source code, using the VAX coding scheme. This copy was not intended, and could not be used, for backup purposes on the IBM computer, and thus was not authorized by the license agreement.

29. Since SAS could not be executed on the VAX computer, the loading of SAS source code onto the VAX could not have been for the purpose of executing test runs of SAS, as testified by S & H. This conduct, admitted by S & H, supports the Court's conclusion that S & H licensed SAS for the purpose of examining SAS source code in connection with the preparation of its conversion of SAS to run on VAX computers.

30. Once having loaded SAS source code onto the VAX computer, that source code was available for examination and study to all S & H employees and consultants working on its conversion project. Printouts of the SAS source code were made available on demand to S & H's programmers, and could be made by those programmers themselves on an unlimited basis. In addition, through video screen terminals located at S & H's office and connected to the Vanderbilt VAX, S & H's programmers were able to call up any desired portion of the SAS source code on the screens. Each S & H programmer had access to that portion (the non-supervisor part) of the SAS source code stored on the VAX.

31. In addition to storing the SAS source code, the VAX was capable of editing the source code, through the use of a "text editor." This capability, available on a unlimited basis to all S & H programmers, allowed them to manipulate the SAS source code much as a word processor is able to do. It is possible to select particular lines or groups of lines for editing, to rearrange or move single lines or blocks of lines of code, and to make any desired change in single lines or blocks of lines of code. These text editing capabilities were also used by S & H for the development of its SAS conversion.

32. The use by S & H of the Vanderbilt VAX was billed to S & H under the account name "SAS." This name was provided to Vanderbilt for billing purposes by S & H.

33. The Court finds that the SAS source code was extensively and systematically used by S & H in the preparation of S & H's product. The specific grounds for this finding are detailed below.

34. One of the Institute's experts, Dr. Peterson, testified that it was his expert opinion that early source code for the S & H product was substantially similar to the source code for SAS. The Court finds his testimony credible, and accepts it.

35. Dr. Peterson's opinion was supported in part by some 44 specific instances identified by him in which the S & H product showed evidence of direct copying from SAS. These examples are summarized in Exhibit P-27.

36. S & H's expert, Dr. Merten, did not disagree with Dr. Peterson's factual observations, with the possible exception of three out of the 44 examples found by Dr. Peterson (which he viewed as "misunderstandings" of the S & H product). Although Dr. Merten drew the conclusion that the more than 40 examples were not evidence of copying, this conclusion is entitled to little or no weight. Dr. Merten testified that it was based on his evaluation of the credibility of S & H programmers with whom he discussed the matter, and his decision to believe their statements. This Court by contrast has found testimony presented by S & H programmers to be not credible. In addition, one of the S & H programmers on whose credibility Dr. Merten relied was not produced by S & H to give testimony at trial. This programmer, Frank Kyle, was the author of the S & H "TTEST" procedure, in which Dr. Peterson found clear evidence of copying. Exhibits P-26, P-27, Ex. 42-44. The Court draws the inference that his testimony would not have been favorable to S & H, further weakening S & H's case, and further undercutting Dr. Merten's conclusion. Furthermore, at least 18 of the examples provided by Dr. Peterson were described by Dr. Merten himself as "similar to" SAS source code or documentation. Exhibit P-29.

37. The Institute's other expert, Prof. William Kennedy, testified to the substantial similarity between a key portion of the S & H source code, the "GLM" procedure, and the comparable SAS source code. Dr. Merten did not offer an opinion in this regard. The only contrary evidence introduced by S & H was testimony by Wayne Ray, the author of the S & H GLM procedure. This testimony was not credible.

38. S & H argued at trial that the 44 examples of copying presented by Dr. Peterson were trivial, in the context of the claimed 186,000 total lines of source code in S & H's product. The Court cannot accept this argument.

39. As part of early efforts to resolve this dispute, S & H permitted two employees of the Institute, including its president and one of the principle authors of SAS, Dr. Goodnight, to examine printouts of portions of the source code for the S & H product on July 12, 1982. Dr. Goodnight testified at trial that during that examination, he observed numerous instances of literal, near literal and organizational copying from SAS. Subsequently, the Institute informed S & H of these findings. Exhibit D-13. The Court finds Dr. Goodnight's testimony to be credible, and accepts it.

40. Within the next few days, S & H destroyed those printouts.

41. It was established by the Institute, and not rebutted, that the S & H source code examined on July 12, 1982, could not be reproduced as it existed on that date. Dr. Peterson, forced to work with surviving fragments of S & H code, confirmed some but not all of Dr. Goodnight's testimony. The Court concludes that were complete printouts of the July 12, 1982, version of the S & H product available, they would support Dr. Goodnight's testimony.

42. The Court's conclusions are still further strengthened by conduct of S & H before the confidentiality order was entered. Dr. Peterson testified to numerous instances in which examples of copying which were present in the S & H product at the time of the July, 1982, inspection were altered between that time and the commencement of the reports to Dr. Winner. The effect in each instance was to make the S & H source code appear on its face to be less similar to SAS. In many of these instances, no functional reason for the change could be suggested by Dr. Peterson, Dr. Merten, or S & H.

43. Certain such examples are particularly striking. In December of 1981, the S & H product was programmed so as to print at the beginning of its output the label, "Statistical Analysis System," using the same title and in the same style as displayed by SAS itself. Subsequently, this caption was changed after S & H altered the name of its product. Exhibit P-26, Ex. 40.

44. Another striking example is the use in early versions of S & H's product of the name "SAS" in at least 145 separate lines of source code. Dr. Peterson demonstrated, and the Court accepts as true, that at some point the VAX text editor was used to systematically change each such occurrence, either by eliminating it entirely or by changing the term "SAS" to a different term. Exhibit P-26, Ex. 27. Such changes are easy to accomplish, and require only seconds through the use of a text editor. The results of the changes left in many instances ungrammatical or incoherent statements in the S & H source code. In one instance, the result was to alter a reference to SAS, but to leave intact a reference to SAS documentation. The Court thus concludes that these changes do not represent any effort to improve the S & H product, but rather represent an effort to mask and disguise evidence of copying.

45. Further evidence of the use of and copying from SAS source code is found in a document prepared by S & H and distributed to its programmers, purporting to set forth coding standards to be used by them in preparing the S & H product. Exhibit P-5. This document, prepared by S & H on a VAX computer, is headed "SAS—STANDARDS—TXT; 11-SAS," and its capitalized title is "Coding Standards for the Portable SAS Implementation." The document instructed the S & H programmers to select "driver routine" names "as close as possible" to the equivalent SAS names. Those names are internal to the source code, and are not visible to a user. S & H programmers were required to describe "any differences between the implementation of [each] procedure and the corresponding SAS implementation." Such information could be learned only by close examination of the SAS source code. S & H programmers were specifically instructed not to use a particular routine from SAS; "the VARY subroutine, documented in the SAS programmer's guide, should not be used...." This document further demonstrates that the modular structure of INDAS (the S & H product) was not developed independently, but was specifically intended to duplicate that of SAS.

46. Still more support for the Court's conclusion is found in the lack of design documentation for the S & H product. As noted above, the S & H personnel originally contemplated an original statistical package to be known as PASQUEL, and actually produced preliminary design documentation for that system. That project was abandoned. No comparable design documentation was ever prepared for the subsequent S & H product. The evidence at trial demonstrated, and the Court finds, that S & H prepared no design documentation at any time for its product. On the contrary, the Court finds that S & H adopted SAS, wholesale and in detail, as the design documentation for its product.

47. The argument put forward at trial by S & H, that a document known as "DOCOUT" (Exhibit D-45) was the design document for its product, is not credible and is not accepted by the Court. This document is in fact an extraction from the S & H product of lines of source code considered as "comments" describing the functioning of the accompanying code. The DOCOUT document could not have been produced until after those comments were prepared. These comments are organized into a structure of programs and sub-programs similar to that of SAS. Therefore, the Court finds a large part of the design of INDAS to have been appropriated from the SAS source code prior to and during the preparation of DOCOUT.

48. The Court particularly notes that the DOCOUT document deals largely if not entirely with "interface routines." These are the portions of the S & H product which deal with the inter-relationship between the "supervisor" portion and the "procedure" portions of the product. Dr. Merten testified that of 70 interface routines contained in SAS, 69 were included in the S & H product. The Court accordingly specifically finds that the interface routines of the S & H product were based on and derived from SAS.

49. Although the distribution tape for SAS provided to S & H did not contain source code for the supervisor portion of SAS, the Court cannot find that S & H produced its supervisor through independent design efforts. There was, as noted, no design document ever produced by S & H for its supervisor. In addition, the S & H supervisor contains the interface routines, as described above, which were derived from SAS copyrighted interface routines, to which S & H had access.

50. Additional telling evidence of copying with respect to the S & H product relates to the existence of "undocumented options." These are elements of the SAS user language, the set of commands and instructions employed by a SAS user in order to cause the SAS program to perform a particular set of statistical analyses. Certain options and other features of the SAS user language are not described or documented in any SAS publication, and thus their very existence can be learned only by close examination of the SAS source code.

51. An example will help illustrate the significance of this point. During the development of SAS 79.5, there was concern about the amount of time required for the computer to perform certain calculations. In order to evaluate this, Dr. Goodnight provided in the SAS source code for a user to include the option, "OBSTIME" in his commands to SAS to perform a particular statistical analysis. After testing, the option was deleted, except for one overlooked reference. The existence of this option was never disclosed outside the Institute, and it never appeared in any SAS user manual or other documentation. Only a vestigial and non-functional remnant of the employment of "OBSTIME" appears in the released SAS 79.5 code, and Dr. Goodnight had entirely forgotten its existence prior to this litigation.

52. When Dr. Peterson examined the S & H source code, he discovered in it a totally functionless appearance of "OBSTIME," exactly comparable to the vestige remaining in SAS. Exhibits P-26, P-27, Ex. 8. The only conceivable explanation is that S & H copied this non-functional feature from the SAS 79.5 source code. A second such example was provided by Dr. Peterson as his Example 38.

53. S & H admitted at trial that it specifically sought out, through the use of computer programs specially prepared by S & H for this purpose, each such undocumented option or other feature contained in the SAS source code. Each undocumented option or features located by S & H was then carefully and precisely reproduced in the S & H source code.

54. The argument put forward by S & H at trial that its actions were necessary in order to obtain "upward compatibility" with SAS. Specifically, Dr. Merten testified that the creation of compatible software is a common practice in the computer software industry. By this, Dr. Merten meant the original creation of a computer program which will perform the same overall task as an existing program. However, Dr. Merten did not suggest that this practice extends to the preparation of such software through the misuse of proprietary and copyrighted materials, as was done by S & H in this instance.

55. Still additional evidence of the use and copying of SAS materials by S & H is found in the organizational patterns used by S & H. At the origin of its project, all S & H programmers had access to the entirety of SAS source code and other materials. Mr. Sherrod and Dr. Ray later changed this practice, so that individual programmers would be assigned a particular, narrow task, and provided only that information relevant to the task.

56. For example, Beth Barnwell testified that she was provided a specific statement of the task to be performed by a program module, and was given information as to the ways in which that module of code would interface with the remainder of the S & H product. She was not provided information as to the source of that material.

57. It was Mr. Sherrod and Dr. Ray who provided her with that information, information which in fact was derived and copied from SAS. Sherrod and Ray in fact wrote the source code for the routines which would link Barnwell's modules to the remainder of the S & H product. Their protestations to the contrary have already been found by the Court to be not credible and are not believed. By placing themselves at the hub of the S & H wheel, and carefully controlling the information provided to the programmers at the spokes, Sherrod and Ray hid from the very people involved the extent to which their work was derived from SAS. Thus, the testimony of other S & H programmers, while perhaps true to the best of their knowledge, is in the end unpersuasive.

58. The Court further finds that, to the extent that S & H copied from the SAS source code and other copyrighted SAS materials, such copying was of expression, and not merely of ideas.

59. The testimony of Dr. Goodnight, one of the principal authors of SAS, as well as both the Institute's and S & H's experts, was that throughout the preparation of a complicated computer program such as SAS, the author is faced with a virtually endless series of decisions as to how to carry out the assigned task. Beginning with a broad and general statement of the overall purpose of the program, the author must decide how to break the assigned task into smaller tasks, each of which must in turn be broken down into successively smaller and more detailed tasks. At the lowest levels the detailed tasks are then programmed in source code. At every level, the process is characterized by choice, often made arbitrarily, and only occasionally dictated by necessity. Even in the case of simple statistical calculations, there is room for variation, such as the order in which arithmetic operations are performed. S & H itself presented evidence that there are at least two substantially different ways of programming the "MEANS" procedure, one of the simplest statistical analyses. As the sophistication of the calculation increases, so does the opportunity for variation of expression.

60. S & H presented no evidence that the functional abilities, ideas, methods, and processes of SAS could be expressed in only very limited ways. On the contrary, the Court finds that to the extent that similarities between SAS and the S & H product have existed, they represent unnecessary, intentional duplication of expression.

61. The Institute's testimony, including that of Dr. Goodnight, Dr. Peterson and Prof. Kennedy, established that the S & H program follows the organizational structure of SAS, down to a detailed level. This testimony was not rebutted, except for conclusory denials of copying by Sherrod and Ray, which the Court has found not credible.

62. An illustration is Dr. Peterson's Example 29, concerning the "MEANS" procedure. Dr. Peterson testified that S & H employed a very close paraphrase of the SAS structure and organization. However, Dr. Peterson demonstrated that S & H included a particular logical test (an "IF" statement in source code) which, due to an apparently unrelated change made by S & H, is totally functionless. The inclusion of this statement by S & H can be explained only as a result of slavish copying of structural detail.

63. S & H argued at trial that its approach was essentially to adopt the SAS organizational scheme, but to write the actual program code for each of the lowest level tasks independently of the actual SAS code used to accomplish that task. Even were the Court to accept the S & H testimony as credible, which it does not, the method described is strikingly similar to that described in Meredith Corporation v. Harper & Row Publishers, Inc., 378 F. Supp. 686 (S.D.N.Y.), aff'd., 500 F.2d 1221 (2d Cir.1974), opinion after trial, 413 F. Supp. 385 (S.D.N.Y.1975). That court found a process by which a textbook was outlined by certain employees of the defendant, the outlines were distributed to other personnel, and they wrote "original" text based on the outlines, resulted in duplication of expression, and not merely of ideas.

64. This result is also supported by Synercom Technology, Inc. v. University Computing Co., 462 F. Supp. 1003 (N.D. Texas 1978), the principle authority upon which S & H relies. Synercom noted that the programming in source code of a detailed English-language statement of a problem solution would result in duplication of expression and copyright infringement; here, S & H has taken the simpler route of copying directly from SAS source code.

65. On the evidence presented at trial, the Court finds that the actual method employed by S & H was not the method described in its testimony. Rather, it is clear that S & H's programmers actually referred and relied, systematically and pervasively, on the SAS source code and other proprietary and copyrighted materials. This is the only rational inference to be drawn from the nature and scope of the similarities demonstrated at trial, from the admitted conduct of S & H both prior to and during this litigation, from the deliberate non-disclosure by S & H of its intent in licensing SAS, from S & H's expressed desire that its product be adopted by the Institute as the VAX version of SAS, from S & H's use of the name "SAS" in connection with its product, from S & H's original intention to use "PSAS" as the trademark and marketing name of its product (see Exhibit P-9), from S & H's own description of its product as a "conversion" of and as "modeled after" SAS (Exhibits P-11, P-14), and from S & H's stubborn insistence at trial on an incredible story.

66. The advantage derived by S & H from its use and misuse of SAS materials was not limited to its copying and derivation from those materials. By expropriating a completed and proven design, S & H was saved the substantial time and effort to developing its own original design. Because the structure of SAS was proven, S & H was able to avoid the risk and expense of false starts, dead ends, re-designs, and other fruitless efforts. S & H's conduct is analogous to stealing the blueprints for a skyscraper, thus avoiding the labor and expense of engineering studies, structural analysis, and innumerable design decisions. The extent of this advantage achieved by S & H is difficult to estimate with precision, but was clearly substantial.

CONCLUSIONS OF LAW

I. Validity of Copyright in SAS 79.5.

The copyright registration certificate for SAS Release 79.5, Exhibit P-2, is prima facie proof of the validity of the copyright. 17 U.S.C. § 410(c). The burden of overcoming this presumption of validity is on S & H, Williams Electronics, Inc. v. Artic International, Inc., 685 F.2d 870 (3d Cir.1982). The only reason S & H has suggested why the copyright might be invalid is the relationship between SAS 79.5 and earlier releases of SAS, particularly Release 76.2. The Court has found, as a matter of fact, that SAS 79.5 represents overwhelmingly a new and original work of authorship, above and beyond the pre-existing work contained in earlier release of SAS. Accordingly, the copyright in SAS 79.5 is valid and fully enforceable. 17 U.S.C. § 103(a) (subject matter of copyright included derivative works), § 101 (definition of derivative work). This is so regardless of whether SAS 76.2 is or is not in the public domain. In connection with earlier summary judgment motions, the Court held that SAS release 76.2 was in the public domain.

S & H has argued, although this point was never raised in the pleadings, that enforcement of the Institute's copyright should be denied on the grounds of unclean hands or fraud on the Copyright Office. However, S & H has not demonstrated any withholding by the Institute of information which might have resulted in denial of registration, nor has S & H shown that it or anyone else was in any way misled by the Institute's registration application. Accordingly, the purposes of the statutory registration requirement have been met, and the S & H argument must be rejected. Midway Mfg. Co. v. Artic International, Inc., 547 F. Supp. 999, 1009-10 (N.D.Ill.1982), aff'd, 704 F.2d 1009 (7th Cir. 1983).

II. The Institute's Contract Claims.

The Court has previously ruled in connection with the Institute's first motion for partial summary judgment that S & H violated the specific terms of the license agreement. These violations included the use of SAS on an unauthorized CPU, the creation of unauthorized copies of SAS, and the use of SAS subsequent to the proper termination of the license agreement by the Institute on June 3, 1982. Exhibit D-29. The Court's memorandum opinion issued on July 25, 1983, indicated that these rulings were dependent upon proof of the validity of the copyright in SAS 79.5. Copyright validity having been found, the Court's findings with respect to breach of contract and termination of license agreement are hereby adopted and incorporated, and are made unconditional. The Court notes the Institute's argument that it was not necessary for the Court to make those rulings conditional, as the license agreement is valid and enforceable even in the absence of a valid copyright. However, since the Court has found copyright validity, it need not reach that question.

In the summary judgment ruling, the Court did not reach the Institute's claim that S & H also violated an implied covenant in the license agreement. There can be no doubt that the license agreement between the Institute and S & H included implied duties of good faith and fair dealing. "Every contract imposes upon each party a duty of good faith and fair dealing in its performance and its enforcement." Restatement (Second) of Contracts, § 205 (1979). The precise scope of the covenant must be considered with respect to the facts of each case. "Good faith performance or enforcement of a contract emphasizes faithfulness to an agreed common purpose and consistency with the justified expectations of the other party.... Subterfuges and evasions violate the obligation of good faith in performance even though the actor believes his conduct to be justified.... [T]ypes of bad faith [include] evasion of the spirit of the bargain [and] willful rendering of imperfect performance...." Id., Comments a, d.

The content of this implied covenant has been considered previously by the courts in the context of copyright licenses. Professor Nimmer states the rule that a licensee violates the covenant "by creating a new work based upon the same idea, theme, or title, even if a stranger could create a new work with such idea, theme or title without infringing the grantor's copyright." 3 Nimmer on Copyright § 10.11[B] (1983). S & H correctly points out that this is true where a licensee has a duty to promote the work, such as a publisher. However, S & H fails to persuade the Court that the covenant is limited to those situations.

On the contrary, S & H ignores cases such as Uproar Co. v. National Broadcasting Co., 81 F.2d 373 (1st Cir.1935), cert. denied, 298 U.S. 670, 56 S. Ct. 835, 80 L. Ed. 1393 (1936), cited by the Institute. In Uproar, a comedian was engaged to write and perform radio broadcasts; he subsequently published and sold the scripts. The District Court held this to violate the implied covenant, and the Court of Appeals affirmed. "The principle is well established: `That in every contract there is an implied covenant that neither party shall do anything which will have the effect of destroying or injuring the right of the other party to receive the fruits of the contract, which means that in every contract there exists an implied covenant of good faith and fair dealing.'" 81 F.2d at 376-77, quoting Kirke La Shelle Co. v. Paul Armstrong Co., 263 N.Y. 79, 188 N.E. 163, 164 (1933), also relied on by the Institute.

In County of Ventura v. Blackburn, 362 F.2d 515 (9th Cir.1966), the same result was reached in the context of a license to reproduce and sell copyrighted maps. The District Court held "That the contract implicitly requires the [licensee] to exercise good faith in the performance thereof and to do nothing whatsoever that would injure or destroy the value of the copyright to the [licensor] and that the [licensor], reciprocally has the same duty to the [licensee] with respect to the subject matter of said contract." 362 F.2d at 918. The Court of Appeals remanded the case with respect to damages, and affirmed "in all other respects," id. at 522. Cf. Manners v. Morosco, 252 U.S. 317, 327, 40 S. Ct. 335, 336, 64 L. Ed. 590 (1920) (enjoining licensee of play from producing motion picture version, conditioned on licensor not doing so while license in effect); Harper Bros. v. Klaw, 232 F. 609, 613 (S.D.N.Y.1916) ("negative covenant" implied; the "law is analogous to that which implies, from a covenant to make a certain use of property, a covenant negative against doing anything else with it").

In short, although it may be true, as S & H argues in its trial brief, that it "has never agreed, impliedly or expressly, not to produce a piece of statistical software designed to operate on VAX Computers," it did agree as a matter of law not to use proprietary SAS materials in the process, as it did. S & H's conduct cannot be said to comply with its legal duty of good faith and fair dealing.

III. The Institute's Claim of Copyright Infringement.

The Institute's copyright claim has two components. First, the Institute claims that in the course of preparing its product, S & H made an unknown number of unauthorized, and thus infringing, exact copies of some or all of the SAS source code, as well as engaging in other unauthorized, infringing conduct. Second, the Institute claims that the S & H product itself is either a "copy" of, or a "derivative work" based upon, SAS, in either case making it an infringing work. The claim is that the S & H product, while perhaps containing some original elements, was in substantial part produced by copying or derivation from SAS.

The first claim was decided in the Institute's favor in connection with the Institute's first motion for partial summary judgment. The Court there found, subject only to proof of validity of the copyright in SAS 79.5, that S & H's use of SAS on an unauthorized computer, the making of innumerable copies for unauthorized purposes, and any use whatsoever of SAS following the lawful termination of the license agreement by the Institute constitute copyright infringement. Gilliam v. American Broadcasting Companies, Inc., 538 F.2d 14 (2d Cir.1976); National Bank of Commerce v. Shaklee Corporation, 503 F.Supp. 533 (W.D.Texas 1980). The Court has found above that the Institute's copyright in SAS 79.5 is indeed valid. Accordingly, the findings of the Court on the summary judgment motion are hereby adopted and incorporated, and are made unconditional.

The Institute's second claim was also discussed, although not decided, in connection with the summary judgment motions. As the Court there noted, the critical issue is whether S & H appropriated from SAS only ideas and concepts, or whether it also appropriated expression. This decision, difficult and ad hoc in any case, see Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir.1960), is particularly difficult in the case of complex computer software. As stated by the National Commission on New Technological Uses of Copyrighted Works [CONTU] in its final report, "one is always free to make the machine do the same thing as it would if it had the copyrighted work placed in it, but only by one's own creative effort rather than by piracy." The Court observed in connection with the Institute's summary judgment motion that if it succeeded in proving its allegations, then the Institute would be entitled to recover on this claim. The Court must now decide whether the Institute has sufficiently proven those allegations.

In this case under the Copyright Act, the Court starts with the statute itself. See generally Apple Computer, Inc. v. Franklin Computer Corporation, 714 F.2d 1240 (3d Cir.1983). Section 501(a) of the Act (references are to Title 17, U.S.C.) defines copyright infringement: "Any one who violates any of the exclusive rights of the copyright owner as provided by Sections 106 through 118 ... is an infringer of the copyright." Under section 106, "the owner of copyright under this title has the exclusive rights to do and to authorize any of the following: (1) to reproduce the copyrighted work in copies or phonorecords; (2) to prepare derivative works based upon the copyrighted work...." Since it is undisputed that the use of SAS in the preparation of the S & H product was not authorized by the Institute, the only question is whether that product constitutes a "copy" or a "derivative work" based upon SAS.

The term "copies" is defined in section 101 of the Act as "material objects, other than phonorecords, in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device." This definition, obviously, is focused on the "fixation" requirement, rather than on the degree of similarity which must be shown to establish that a challenged work is a "copy." The Court therefore must fall back upon judicial construction of this term, and concludes that it must determine whether or not the S & H product is "substantially similar" to SAS. Substantial similarity, of course, does not require literal identity; "a play may be pirated without using the dialogue." Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49, 55 (2d Cir.), cert. denied, 298 U.S. 669, 56 S. Ct. 835, 80 L. Ed. 1392 (1936).

It is not disputed that there are many similarities between the S & H product and SAS. Even S & H's expert agreed that at least 18 of Dr. Peterson's examples show clear similarity. S & H argues, however, that such similarities as have been demonstrated are similarities of idea rather than of expression, and that in any event, Dr. Peterson's examples of copying are so few and far between as to be as a matter of law trivial and insufficient to establish substantial similarity. Whether the similarities shown between the S & H product and SAS are similarities of idea or expression is a question of fact. Goodson-Todman Enterprises, Ltd. v. Kellogg Company, 513 F.2d 913 (9th Cir.1975). The Court has found as a matter of fact that the expression, and not merely the ideas, of SAS was duplicated, and thus this branch of the test is satisfied.

The question of the substantiality of the similarity is also a question of fact; the piracy of even a quantitatively small fragment ("a rose by any other name would smell as sweet") may be qualitatively substantial. Roy Export Company Establishment v. Columbia Broadcasting System, Inc., 503 F. Supp. 1137, 1145 (S.D.N.Y. 1980), aff'd, 672 F.2d 1095 (2d Cir.), cert. denied, 459 U.S. 826, 103 S. Ct. 60, 74 L. Ed. 2d 63 (1982). It certainly cannot be said that 44 specific examples of copying as a matter of law are insubstantial. Further-more, the Court has found as a matter of fact that evidence of copying was much more extensive, before S & H embarked upon its program of destroying and masking evidence and disguising its conduct.

In addition, the copying proven at trial does not affect only the specific lines of code cited by Dr. Peterson in his testimony. Rather, to the extent that it represents copying of the organization and structural details of SAS, such copying pervades the entire S & H product. In this respect, the Court finds persuasive the reasoning of Meredith Corporation v. Harper & Row Publishers, Inc., supra:

Passing the issue of whether an admitted substantial infringer is entitled to Court review of every corner of the infringing book for the purpose of excising the plagiarized portions from those which are not, the simple answer to the contention raised is that the record demonstrates and I find ... an extensive taking of the structure and topical sequence of the Mussen book in addition to the eleven percent of the Meredith book admittedly plagiarized.... Thus I conclude that while the Meredith text contains some independent ideas of the author, some independent research, some additional topics and some different structure, the topic selection and arrangement of the Meredith book are in substantial part the result of copying of the Mussen book not attributable to independent effort by Meredith or the necessary result of limited possibilities for organizing and presenting the material to be covered.... The copying thus affects the entire text and not just the eleven percent admittedly infringed. 413 F.Supp. at 386, 387 (emphasis original, citations omitted).

The facts in Meredith are also strikingly similar to the facts in this case, including the selection of a leading work in the field to serve as a model (Sherrod and Ray described their product in letters to third parties as "modeled after" SAS); the duplication by leaders of the project of the organization of the model (proven here by the testimony of Dr. Peterson and Prof. Kennedy, but hidden by Sherrod and Ray even from the programmers working under them); the actual writing by subsidiary personnel under the control of the leaders of the project; the provision of portions of the model to the actual authors (S & H programmers all had access to SAS source code, and were provided printouts on request); the performing of some original research after the duplication of the basic outline; the substitution of references in order to disguise copying; and the completion of the task in but a fraction of the time required by the original authors. I therefore conclude, as did the Meredith Court, that the works are substantially similar. The S & H product infringes the copyright in SAS 79.5.

This result follows even more clearly on the second point of the Institute's claim. A "derivative work" is defined in the Copyright Act as "a work based upon one or more pre-existing works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be re-cast, transformed or adapted." 17 U.S.C. § 101. The definition includes "editorial revisions, annotations, elaborations, or other modifications, which, as a whole, represent an original work of authorship...." Thus, it is irrelevant that the S & H product may contain, as S & H argues, significant original portions contributed by S & H. The Court has found as a matter of fact that the product was substantially and pervasively "based upon" SAS.

The statute deals with human conduct, the nature of the activity which resulted in the defendant's product. There can be no doubt that, as even S & H's expert agreed, S & H "targeted" SAS for duplication, and that it then engaged in improper conduct to achieve what S & H itself called its "conversion" of SAS. The S & H product is therefore well within the statutory definition of a derivative work, and not having been authorized by the Institute, constitutes an infringement of the Institute's copyright. Midway Mfg. Co. v. Artic International, Inc., 704 F.2d 1009 (7th Cir. 1983) ("speed-up kit" for video game, which causes characters to move in different and more complex ways, is derivative work).

S & H's argument that it can cure its infringement by simple excision is flatly inconsistent with the statute. The argument essentially is that S & H can misappropriate every internal and external feature of a building, can then cure any impropriety by changing the tint of the windows and the color of the siding. The argument must be rejected. Tennessee Fabricating Company v. Moultrie Manufacturing Company, 421 F.2d 279 (5th Cir.), cert. denied, 398 U.S. 928, 90 S. Ct. 1819, 26 L. Ed. 2d 91 (1970); Davis v. E.I. duPont DeNemours & Co., 240 F. Supp. 612 (S.D.N. Y.1965).

IV. S & H's Claims.

With respect to S & H's claims of breach of contract and interference with contractual relations, the Court finds in favor of the Institute. S & H presented little or no evidence in support of these claims. The Institute's conduct concerning Tennessee State University was entirely reasonable, and privileged. Restatement (Second) of torts § 772 (1979). With regard to the July, 1982, source code examination, S & H failed to demonstrate any breach of the agreement by the Institute, or any injury resulting therefrom. Judgment will enter in favor of the Institute on these claims.

V. Attorneys' Fees.

The Court notes that regardless of the willfulness of S & H's unlawful conduct, it is unable to make an award of attorneys' fees to the Institute under the Copyright Act. 17 U.S.C. § 412.

VI. Relief.

Having decided that the Institute is entitled to injunctive relief, the Court must determine its appropriate extent. S & H argues that the Court's power is limited by law to ordering excision of only the specific lines of source code referred to in Dr. Peterson's examples of copying, while the Institute argues that, since the Court is not in a position to separate the "poison" of copying from the "soup" of the S & H product, the entire product should be enjoined.

The Court starts with the premise that the appropriate scope of injunctive relief is within the sound discretion of the Court. The Copyright Act specifically authorizes final injunctive relief "on such terms as [the Court] may deem reasonable to prevent or restrain infringement of a copyright." 17 U.S.C. § 502(a). The Court cannot accept S & H's arguments that its discretion under this broadly-stated provision is narrowly constrained.

First, S & H argues that it has "cured" its original infringement by removing from its product the particular examples identified by Dr. Peterson. It cites cases involving motion pictures and television programs for the proposition that only the final, marketed version of the product is relevant. However, in each of those cases, the plaintiff claimed only that the final movie or TV program was "substantially similar" to the plaintiff's prior work. None involved a claim that the production was a "derivative work" prepared in violation of section 106(2) of the Copyright Act; both the early versions of the S & H product and S & H's efforts to destroy or alter that evidence are central to that claim.

Furthermore, evidence at trial showed that the S & H product has no "final version." Unlike a movie or television program, a computer program lives a life of continuous change. It is constantly being edited, revised, and altered, even after marketing, and those alterations can as easily introduce infringing elements as remove them. That is especially important here, where S & H has continued to alter its program during the litigation.

The Court also notes Dr. Peterson's testimony that his examples are by no means exhaustive, but were limited by the limited time and resources available in view of the sheer magnitude of the task. Considering the pervasive nature of the copying by S & H and S & H's efforts to destroy, alter and disguise the evidence, the Court considers it highly doubtful that all results of copying have been, or will ever be, excised.

The Court further observes that Dr. Merten offered no opinion on the similarity between the current S & H source code and SAS. On the contrary, he testified that he has never examined S & H source code prepared since the reports to Dr. Winner began.

Second, S & H argues that the "supervisor" portion of its product is separable from the statistical procedures, and that therefore the Court cannot as a matter of law enjoin marketing of that portion. Although the Court agrees in the abstract with Professor Nimmer's suggestion that a "defendant may be able to avoid a permanent injunction if the infringing portion of defendant's work can be removed without destroying the usefulness of the remainder of the work," 3 Nimmer on Copyright § 14.06[B] at 15-54, the argument here founders on the facts.

Much of the testimony concerning copying and derivation concerned the interface routines through which the supervisor and statistical portions of the S & H product communicate. The Court has found that these routines were significantly derived from proprietary SAS materials.

S & H makes much of the fact that it was not provided with the source code for the SAS supervisor, which it therefore could not have copied. However, this does not mean, as S & H argues, that its supervisor is free of copying. In fact, the evidence showed that the boundary between supervisor and nonsupervisor portions (a boundary which is any event is not crystal-clear) is different in SAS and the S & H product. Dr. Peterson testified that code for handling input and output was placed by the Institute outside its supervisor, but was placed by S & H inside its supervisor. Dr. Peterson's Example 34 related to the "PROBHYPR Function," for which S & H did have SAS source code; according to Dr. Merten, S & H placed its comparable (and copied) code in its supervisor. Exhibit D-2. The evidence concerning the interface routines demonstrates that the S & H supervisor contains information copied from non-supervisor portions of SAS.

This Court is not a computer programmer, and is simply not able to determine, out of 186,000 lines of integrated product code, which of them reflect misappropriation. Nor is the Court willing to rely on adjudged plagiarists, whose courtroom testimony the Court found not worthy of belief, to make that determination. It is enough, as the Fifth Circuit has said, that "the starting point for [S & H's] redesign was the [Institute's] work." Tennessee Fabricating Company v. Moultrie Manufacturing Company, 421 F.2d 279 (5th Cir.), cert. denied, 398 U.S. 928, 90 S. Ct. 1819, 26 L. Ed. 2d 91 (1970); Orth-O-Vision v. Home Box Office, 474 F. Supp. 672 (S.D. N.Y.1979) (injunction against entire product where division is technologically unfeasible); Davis v. E.I. duPont DeNemours & Co., 240 F. Supp. 612 (S.D.N.Y. 1965); cf. Burroughs v. Metro-Goldwyn-Mayer, Inc., 683 F.2d 610, 620 n. 9 (2d Cir.1982) (court "should not assume the role of film editor").

The Court's injunction of the supervisor portion of INDAS need not rest upon principles of copyright law.

The Court's conclusion that a broad injunction should issue is reinforced by consideration of fundamental principles of contract law. The Court is mindful that contractual remedies have broad purposes, including the protection of the parties' expectations and the prevention of unjust enrichment. See Restatement (Second) of Contracts, § 344. Any equitable relief granted by this Court must be tailored to accomplish these purposes so far as practicable under the facts of the case.

Here, a broad injunction can be supported on two distinct contractual grounds. First, it is the only way in which the Institute can be afforded the benefit of its bargain. Central to the contract between the parties is that the use of SAS materials by S & H was strictly limited. Obviously, the Court cannot turn back the clock and prevent S & H from making impermissible use of those materials. It can, however, put the parties in the same position they would be in had S & H complied with the license agreement. Had S & H done so, it would not have developed its present product, or at the very least, not in its present form and not in the time actually required. It would not, therefore, be in a position to market that product, and hence the Court should not allow it to do so.

Second, a broad injunction is necessary in order to ensure that S & H not be unjustly enriched at the expense of the Institute. Restitutionary relief is appropriate as a remedy for breach of contract, as well as upon rescission. Restatement (Second) of Contracts, § 373. S & H should be required to return not only the materials it was provided, but also all benefits and gains "resulting from the use by [S & H] of whatever [it] received" under the contract. Id., § 376, Comment a and Illustration 5. The Court has found as a matter of fact that proprietary SAS materials were used and relied upon to a substantial degree by S & H; accordingly, the prevention of unjust enrichment can be achieved only by enjoining the marketing and further use or development of the resulting S & H product.

VII. CONCLUSION.

For the reasons set forth herein, judgment will enter in favor of the Institute in both actions. An injunction will issue, enjoining any marketing of INDAS, or any other product copied or derived from INDAS or SAS.

An appropriate order will enter.