S & H COMPUTER SYSTEMS v. SAS Institute, Inc., 568 F. Supp. 416 (M.D. Tenn. 1983)

*417 James V. Doramus, Nashville, Tenn., for plaintiff.

Jay S. Bowen, Nashville, Tenn., J. Thomas Franklin, Boston, Mass., for defendant.

WISEMAN, District Judge.

This action is before the Court on a motion to dismiss and cross motion for summary judgment filed by S & H Computer Systems, Inc. [S & H] and two motions for partial summary judgment filed by SAS Institute, Inc. [SAS]. The present memorandum addresses only the S & H motion to dismiss and the first SAS motion for partial summary judgment.

The standard governing the Court's analysis is clear. Summary judgment is appropriate only where no genuine issue of material fact is present and a party is entitled to summary judgment as a matter of law. In re Atlas Concrete Pipe, 668 F.2d 905, 908 (6th Cir. 1982). In particular, a court "should be reluctant to dispose of a complex case on summary judgment and should do so sparingly where a party might be denied the opportunity to develop fully his case at trial." Id. at 908, n. 5. The moving party has the burden of establishing conclusively that there exists no genuine issue as to a material fact and the evidence together with all inferences to be drawn therefrom must be read in the light most favorable to the party opposing the motion. See Smith v. Hudson, 600 F.2d 60, 63 (6th Cir.1979).

In the late 1960s, James Barr and James Goodnight began work on statistical software which later became known as "the SAS project." Initial funding was provided for research purposes by North Carolina State University. Later, some additional funding was allocated to the SAS project under a pre-existing National Institute of Health [NIH], grant to the University. The NIH grant was terminated in 1973. The Institute of Statistics (an independently chartered organization within the University of North Carolina system) began distributing the 1972 release of SAS pursuant to a license agreement for which it charged a fee for use and support. These agreements *418 prohibited the redistribution of SAS. Prior to 1973, SAS was distributed essentially at no charge (cost of tape only), and without restriction.

In 1973, the Institute of Statistics solicited additional sources of income for the SAS project. The United States Department of Agriculture participated as one of the project sponsors and paid an initial annual fee of $5,000 to help support the project.

In March of 1976, SAS Institute, Inc., was incorporated. SAS executed a separation agreement with North Carolina State University. The University assigned to SAS all the rights in previously copyrighted manuals, along with all other original materials.

During the evolution of the SAS project, several versions of SAS became available. The first production release of SAS occurred in August 1969. After this initial release, SAS grew in scope with additional versions becoming available up through 1972. The development of a new version of SAS began in 1973 and emerged three years later as SAS 76. From 1972 through 1976, the version released to customers was SAS 72. The first release of SAS Institute was known as release 76.2. The next release of SAS was made available on March 22, 1981, as SAS 79.5.

S & H is a Tennessee corporation having its principal place of business in Nashville. It was formed in 1975 and is in the business of developing and marketing computer software. The development of INDAS, the S & H product involved in this litigation, began in the early Spring of 1981. Dr. Charles Federspiel, a professor of biostatistics at Vanderbilt University Medical School, needed a way to analyze the approximately one million Medicare-Medicaid health claim records which are processed in Tennessee every month for health related research. His department had access to a DEC computer, but not an IBM computer. Having heard of the SAS project, Dr. Federspiel called Dr. Goodnight to inquire if SAS would be available for use on the Department of Biostatistics DEC computer. Goodnight stated the SAS project was only available on IBM equipment.

Dr. Federspiel contacted his coworker, Dr. Wayne Ray, along with Philip Sherrod and Harry Sanders (the founders of S & H). These men established the design goals for a statistical software system which became known as INDAS. In connection with their project, S & H obtained a license to use SAS and received a limited portion of the SAS source code. In December 1981, after substantial development of INDAS, S & H contacted SAS inquiring into joint marketing prospects. An exchange of letters and conversations between the two companies ensued, including a letter from Dr. Goodnight explaining the copyright and otherwise proprietary nature of SAS. On June 3, 1982, SAS terminated its license agreement with S & H for alleged violations of the agreement.

The present lawsuit questions: (1) the validity of a copyright in SAS 79.5, (2) alleged similarities between SAS 79.5 and INDAS and (3) violations of the license agreement between SAS and S & H.

S & H asks the Court to dismiss the SAS claim for copyright infringement because SAS 79.5 is not an original work of authorship. S & H alleges that government grants which partially funded the SAS project placed SAS 76.2 in the public domain. Finally, inasmuch as SAS 79.5 is so substantially similar to SAS 76.2 as to not constitute an original work of authorship, S & H argues SAS 79.5 also cannot receive copyright protection.

The S & H argument is straightforward. The Department of Agriculture contract, signed by SAS, that partially funded project S-94 contained the following restrictions:

With respect to the publication of any results of the research conducted under this Agreement ... no copyrights shall subsist in any such publication. Exhibit C to S & H Motion to Dismiss, Paragraph 17(a); see also Paragraph 17(d) (2). The public shall be granted all benefits of any patentable results of all research and *419 investigations conducted and all information, data and findings developed under this Agreement.... Exhibit C to S & H Motion to Dismiss, Paragraph 17(d).

The import of the clauses is clear. The contract expressly states that information developed under this project shall remain in the public domain.

SAS is correct in stating that government funding of a project, in and of itself, does not prohibit a copyright for any developments under the project. See Public Affairs Associates, Inc. v. Rickover, 284 F.2d 262 (D.C.Cir.1960), vacated on other grounds, 369 U.S. 111, 82 S. Ct. 580, 7 L. Ed. 2d 604 (1962), on remand, 268 F. Supp. 444 (D.D.C.1967); Schnapper v. Foley, 471 F. Supp. 426 (D.D.C.1979), aff'd 667 F.2d 102 (D.C.Cir.1981), cert. denied, 455 U.S. 948, 102 S. Ct. 1448, 71 L. Ed. 2d 661 (1982). However, neither of the authorities relied upon by SAS involves a contract that expressly prohibits any copyright.

The Court rejects any argument that SAS 76.2 does not constitute a "publication of any results" under Paragraph 17 of the USDA contract. SAS readily admits that SAS 76 was the culmination of three years of research beginning in 1973. See Affidavit of James H. Goodnight, Ph.D., Memorandum of SAS in Opposition to Motion to Dismiss filed by S & H, Page B-4. The obvious purpose of the clauses is to prevent a private party from gaining an exclusive benefit from publicly funded research. Such purpose is supported by sound public policy. There is nothing in the contract to suggest an unusually narrow interpretation of the phrase. To allow SAS to benefit from such a technical definition of the word "publication" would be in contravention of the clearly worded arms length agreement signed by both parties. Thus, the Court holds that SAS 76.2 is in the public domain pursuant to the agreement entered into with the United States Department of Agriculture.

A genuine factual controversy exists regarding the alleged similarity of SAS 76.2 and SAS 79.5. All or at least some portions of SAS 79.5 may be validly copyrightable. Thus, the Court cannot dismiss the SAS claim for copyright infringement.

Inasmuch as the Court has held that SAS 76.2 is in the public domain, it is unnecessary to address the question of the publication of SAS 76.2 without copyright notices.

S & H has moved to dismiss any claims by SAS regarding violation of trade secrets. There appears little, if any, dispute that the SAS source code could constitute a trade secret as defined in the Restatement of Torts § 758 and as adopted by the courts of Tennessee. See, e.g., Hickory Specialties v. B & L Laboratories, 592 S.W.2d 583 (Tenn.App.1979).^[1] The major arguments on this issue concern the degree of protection required to preserve any alleged trade secret. Both parties present genuine issues of material fact as to whether SAS imposed an obligation of confidentiality upon S & H and whether SAS managed or ever intended to keep its programs secret. The parties disagree over the interpretation of the license agreement. The parties disagree over the intention of SAS ever to protect a trade secret. The parties disagree over the purpose of the SAS source code. The parties disagree over interpretations of statements made by SAS officials. Finally, the parties disagree as to whether there was a "public disclosure" of any alleged trade secrets. In view of these factual disputes, the Court cannot dismiss the trade secret claim at this stage of the proceedings.

SAS alleges that S & H fraudulently entered into the license agreement with SAS for the primary purpose of gaining *420 access to the source code in order to copy and use it in violation of the terms of the agreement. S & H argues the SAS claim for fraud should be dismissed because Tennessee courts do not recognize a cause of action for promissory fraud. In a diversity action, it is the responsibility of the Court to decide the issue as would the Tennessee Supreme Court. See, e.g., Ann Arbor Trust Co. v. North American Company for Life and Health Insurance, 527 F.2d 526, 527 (6th Cir.1975), cert. denied, 425 U.S. 993, 96 S. Ct. 2206, 48 L. Ed. 2d 818 (1976). Upon reviewing the most recent pronouncements of the Tennessee Supreme Court and interpretations made by the appellate courts, this Court is convinced that the Tennessee Supreme Court would recognize a cause of action for promissory fraud under the appropriate fact situation.

In Fowler v. Happy Goodman Family, 575 S.W.2d 496, 499 (Tenn.1978), the Tennessee Supreme Court noted that "although a minority view, the rule established in this state has been that a misrepresentation of intention or a promise without intent to perform is legally insufficient to support a claim for rescission or damages." However, the Fowler court cited dicta from the case of Bolan v. Caballero, 220 Tenn. 318, 417 S.W.2d 538 (1967), in which the court stated that it would be willing to consider adopting the majority rule thereby recognizing an action for promissory fraud, "in a proper case where justice demands ...." Although the Fowler court did not expressly adopt the majority rule, it did apply the majority rule to the facts of the case before it and held those facts legally insufficient to establish an action for promissory fraud.

In analyzing the Fowler court's opinion, a Tennessee appellate court recently noted that "the Tennessee Supreme Court will adopt the majority rule if the party alleging promissory fraud produces evidence supporting the allegations." Brungard v. Caprice Records, Inc., 608 S.W.2d 585, 590 (Tenn.App.1980). The Brungard court upheld a finding of fraud on the alternative grounds of "misrepresentation of a material existing fact" and "misrepresentation of intention or promissory fraud." Id.

The foregoing authorities persuade the Court that the Tennessee Supreme Court would now recognize an action for promissory fraud.^[2] A genuine issue of material fact exists as to what, if any, misrepresentations were actually made. Thus, the Court cannot dismiss the SAS claim for fraud.

SAS presents four grounds upon which it bases its claim for breach of contract regarding the SAS-S & H license agreement.

1. S & H used the licensed program on at least one non-designated CPU, the Vanderbilt VAX, and probably on a second, S & H's own VAX, in direct violation of paragraph 3 of the agreement. Such use was admittedly not for the permitted backup purpose.

2. S & H created unauthorized copies of SAS in violation of paragraph 6 of the agreement on a variety of occasions, including (a) when it transferred SAS to the Vanderbilt VAX; (b) when its employees or consultants printed out portions of the SAS source code; (c) when printouts of the code were printed and distributed to S & H programmers on request; and (d) when SAS source code was apparently transferred onto a disk on S & H's own VAX.

3. S & H continued to use the licensed program beyond the lawful termination of the agreement on June 3, 1982.

4. S & H used the licensed program for the purpose of competing with SAS in violation of a covenant of good faith implied in every license agreement.

S & H does not deny the allegations of making additional copies and transferring the SAS program onto non-designated computers. The fundamental S & H argument is that it did not improperly "use" the SAS *421 program in contravention of the license agreement. Paragraph 3 of the agreement expressly states that S & H is "to use the licensed IPP in any machine-readable form on a single central processing unit (CPU) designated by type/serial number." See Memorandum of SAS Institute, Inc. in Support of its Motion for Summary Judgment, Page A-2. S & H argues that the term "use" as understood in the computer industry, means to execute a computer program or to run statistics. Thus, because S & H only viewed the SAS programs on the non-designated computers, rather than actually executing the program, they did not violate the license agreement. Conversely, SAS argues that extrinsic evidence should not be imported to lend an element of confusion to an otherwise plain and unambiguous term of a contract. The Court finds the SAS logic to be persuasive.

The fundamental purpose of this or any license agreement is to prevent the unauthorized exploitation of the product involved. Access to a particular product is governed by the express terms of the license agreement. The SAS-S & H license agreement and the terms contained therein must be examined in light of these policies. Initially, the Court finds nothing in the agreement itself to suggest any intent to deviate from the everyday meaning of the word "use." As SAS indicates, the common understanding of the term "use" is "the act or practice of employing something." See Memorandum of SAS in Support of its Motion for Partial Summary Judgment, Page 51; see also The American Heritage Dictionary 1410 (New College Edition 1976).

From the admitted facts supplied thus far in the case, there can be no question that S & H gained access to the SAS program for some purpose related to INDAS. Consequently, the SAS program was "used" as a means to achieve some end. Courts have long held that words of a contract are to be given "their plain, ordinary, and popular meaning, in the absence of relevant evidence indicating that the words were used with a different meaning." See Corbin on Contracts § 542 at 106-07 (1960).^[3] After examining the underlying policies of the license agreement, the plain meaning of the term and a lack of evidence that these parties intended some special meaning of the term, the Court is convinced the term "use" should be given its normal meaning. Thus, partial summary judgment for SAS is granted regarding claims for breach of contract involving the use of the SAS program on a non-designated CPU and making unauthorized copies of the SAS program.

SAS also seeks partial summary judgment on the ground that S & H further breached the agreement by making any use whatsoever of the licensed program beyond the June 3, 1982, termination of the agreement. S & H argues only that there was no initial breach of the agreement, thereby making any termination of the agreement by SAS improper. However, inasmuch as the Court has determined that S & H did breach the license agreement, and S & H raises no claims that the termination was procedurally defective, the Court must hold that the agreement was properly terminated. Therefore, because S & H has admitted using the SAS program after the June 3, 1982, termination date, there has been a further breach of the license agreement and summary judgment must be granted.

The Court's finding of a breach of contract based upon the express contract claims of SAS makes any further finding of a breach of contract based upon an implied covenant unnecessary. It should be further noted that the Court is fully cognizant of the fact that all of the SAS breach of contract claims rest upon the ultimate issue that SAS has a valid copyright in SAS 79.5. If no valid copyright exists, an issue which remains open for trial in this case, there would be no basis for the license agreement or, at least, no valid claim for damages. By its ruling, the Court holds only that S & H *422 has taken several actions in violation of the license agreement as written.

SAS seeks partial summary judgment on the issue alleged in Count One of its complaint arguing that S & H violated the SAS copyright in SAS 79.5, S & H having admitted that in the process of developing its product INDAS it repeatedly infringed a valid copyright in SAS 79.5 by making copies of the SAS source code which were not authorized under the license agreement. The SAS claim rests upon the same actions alleged and found to have occurred under the breach of contract claims. Courts have long noted that a licensee infringes a copyright by exceeding his license. See, e.g., Gilliam v. American Broadcasting Companies, Inc., 538 F.2d 14, 20-21 (2d Cir.1976); National Bank of Commerce v. Shaklee Corporation, 503 F. Supp. 533, 544 (W.D.Texas 1980). Thus, use of the SAS program on a non-designated CPU, the making of unauthorized copies and use of the SAS program after the license agreement was properly terminated all constitute a form of copyright infringement and the SAS motion for summary judgment on this issue is granted. Once again, the Court must note that its finding on these issues is binding only after proof that there is a valid copyright in SAS 79.5.

There can no longer exist any doubt that computer programs are copyrightable. See, e.g., H.R. No. 1476, 94 Cong., 2nd Sess. (1976) at 51, reprinted in [1976] U.S.Code Cong. & Admin.News 5659, 5664.^[4] Even S & H admits that "but for the contractual requirement that no copyright protection could be obtained for the versions of SAS, up to and including SAS 76.2, the authors of the SAS program were perfectly free to register SAS with the copyright office." S & H Memorandum in Support of Motion to Dismiss, Pages 28-29. However, S & H alleges that, as a matter of law, INDAS cannot be construed as a "copy" of SAS 79.5. S & H relies on 17 U.S.C. § 102(b) which states that copyright protection does not extend to "any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work." Thus, even assuming all the allegations of SAS are true, S & H argues that it has used only a very limited portion of the SAS scientific ideas and procedures.

If the Court assumes all of the SAS allegations are true, SAS would be entitled to recovery. As the CONTU report stated, "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."^[5] On this issue, the instant case is analogous to the case of Apple Computer, Inc. v. Formula Intern., Inc., supra, wherein the Court stated:

[Plaintiff] seeks here not to protect ideas (i.e., making the machine perform particular functions) but rather to protect their particular expressions of those ideas in the form of specific programs. And [defendant] wants the privilege of using and marketing those expressions without having to invest the millions of dollars and thousands of manpower hours necessary to develop them.

562 F. Supp. at 783.^[6] Similarly, at least one case relied upon by S & H indicates that *423 evidence of direct copying will constitute a copyright violation. As the Court noted in Synercom Technology, Inc. v. University Computing Co., 462 F. Supp. 1003, 1013 (M.D.Tex.1978):

[I]t is as clear an infringement to translate a computer program from, for example, FORTRAN to ALGOL, as it is to translate a novel or play from English to French. In each case the substance of the expression (if one may speak in such contradictory language) is the same between original and copy, with only the external manifestation of the expression changing. Likewise, it would probably be a violation to take a detailed description of a particular problem solution, such as a flow chart or step-by-step set of prose instructions, written in human language, and program such a description in computer language.

The Synercom court distinguishes: (1) literary translations from (2) the preparation of a computer program in any language from a mere general description of the problem to be solved, because the preparation of a new computer program requires "substantial imagination, creativity, independent thought, and exercise of discretion, and the resulting program can in no way be said to be merely a copy or version of the problem statement." Id. However, if a party adopted an already prepared computer program and made very simple changes in the program involving a mere rearrangement of the existing expression to disguise its copying, there would certainly be an absence of worthwhile imagination, creativity and independent thought. Such "simple changes" would involve even less independent thought than translating a novel or play from English to French. The copyright "cannot be limited literally to the text, else a plagiarist would escape by immaterial variations." Nichols v. Universal Pictures Co., 45 F.2d 119, 121 (2d Cir.1930).

Balancing 17 U.S.C. § 102(b) against the thrust of immaterial variations the fundamental question concerns what level of similarity short of verbatim copying will constitute copyright infringement.^[7] Judge Learned Hand stated, "Obviously, no principle can be stated as to when an imitator has gone beyond the `idea,' and has borrowed its `expression.' Decisions must therefore inevitably be ad hoc." Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487, 489 (2d Cir.1960).

In light of the foregoing discussion, one of the ultimate issues in this lawsuit becomes what, if anything, was directly copied from SAS to INDAS or was altered too insignificantly to constitute independent thought or creativity by S & H. Both parties have submitted numerous reports to the Court regarding specific alleged similarities between SAS 79.5 and INDAS. At this point, there can be no doubt that a genuine issue of material fact exists as to any similarities between SAS 79.5 and INDAS making the issue inappropriate for disposition on summary judgment.

For the reasons stated in the accompanying Memorandum, it is hereby ORDERED:

1. SAS 76.2 is in the public domain pursuant to the contract entered into by SAS with the United States Department of Agriculture.

2. Summary judgment for SAS is granted on its claims for breach of contract presented in Count II of the SAS Complaint regarding: (a) S & H use of the licensed program on a non-designated CPU, (b) S & *424 H creation of unauthorized copies of the licensed program and (c) S & H use of the licensed program beyond the lawful termination of the license agreement.

3. Contingent upon a finding at trial that SAS has a valid copyright in SAS 79.5, summary judgment for SAS is granted on its claims for copyright infringement based upon: (a) S & H use of the licensed program on a non-designated CPU, (b) S & H creation of unauthorized copies of the licensed program and (c) S & H use of the licensed program beyond the lawful termination of the license agreement.

In all other respects the S & H Motion to Dismiss and the SAS Motion for Partial Summary Judgment are denied.

[1] "A trade secret may consist of any formula, pattern, device or compilation of information which is used in one's business, and which gives him an opportunity to obtain an advantage over competitors who do not know or use it."

[2] The Court finds inappropriate the reliance of S & H on the case of McElroy v. Boise Cascade Corporation, 632 S.W.2d 127 (Tenn.App.1982). The McElroy court only addressed the issue of negligent representations without analysis or comment as to intentional misrepresentation.

[3] See also Restatement (Second) of Contracts § 202 (1979): "(3) Unless a different intention is manifested, (a) where language has a generally prevailing meaning, it is interpreted in accordance with that meaning; ..."

[4] For a general background supporting the copyrightability of computer programs see, e.g., Williams Electronics, Inc. v. Artic International, Inc., 685 F.2d 870 (3d Cir.1982); Apple Computer, Inc. v. Formula International, Inc., 562 F. Supp. 775 (C.D.Cal.1983); Tandy Corp. v. Personal Microcomputers, Inc., 524 F. Supp. 171 (N.D.Cal.1981).

[5] CONTU is an acronym for the National Commission on New Technological Uses of Copyright Works. The Commission was established by Congress in 1974 to consider major public policy questions of concern to Congress in the copyright field.

[6] The Court is well aware that the Apple case involved alleged copying of a Read Only Memory silicon chip. However, the different mode of expression of the computer program does not make the legal principles enunciated inapplicable to the instant case. As the legislative history indicates:

Under the bill it makes no difference what the form, manner or medium of fixation may be whether it is in words, numbers, notes, sounds, pictures or any other graphic of symbolic indicia, whether embodied in a physical object in written, printed, photographic, sculptural, punch, magnetic, or other stable form, and whether it is capable for perception directly or by means of any machine or device "now known or later developed." H.R. No. 1476, 94th Cong., 2nd Sess. (1976) at 52, reprinted in [1976] U.S.Code Cong. & Admin.News 5659, 5665.

[7] For a general discussion of "comprehensive nonliteral similarity" see 3 Nimmer on Copyright § 13.03[A], at 13-17 (1982).