These consolidated cases arise from a copyright dispute between General Universal Systems, Inc. ("GUS") and HAL, Inc. GUS sued HAL, two of its officers, and several of HAL's customers claiming that HAL infringed its copyright in a freight packaging software system, misappropriated related trade secrets, violated the Lanham Act, and breached a contract with Joe Lopez. The district court dismissed GUS's copyright, Lanham Act, and trade secret claims on summary judgment and, following a jury trial, granted HAL judgment as a matter of law on the contract claim. GUS subsequently filed a second suit against several of HAL's customers, which the district court dismissed on the basis of collateral estoppel. The district court also awarded costs and attorneys' fees to HAL as the prevailing party on the copyright claim. GUS has appealed each of the district court's determinations.

For the reasons we will explain, we AFFIRM in part, REVERSE in part, and REMAND to the district court for proceedings consistent with this opinion.

* The facts and procedural history of this nine-year-old dispute are long and, in some instances, in sharp dispute. With the exception of the contract claim, the underlying claims were each dismissed on summary judgment, and we view those facts in the light most favorable to GUS, the non-movant.1 GUS's contract claim comes to us in a slightly different posture — HAL was awarded judgment as a matter of law after GUS prevailed in a jury trial — and we present the facts related to the contract issue in the light most favorable to GUS.2

* In 1979, GUS developed a software system — the CHAMPION PACKER Computer Software Program — for one of its clients, Joe Lopez. The CHAMPION PACKER program was a tracking system designed for use in the freight forwarding and shipping industry. GUS licensed the software to Lopez, retaining all rights to any improvements to the program, and eventually obtained a copyright registration in 1981.

Sometime later, Lopez created a derivative version of CHAMPION PACKER by converting it from the BASIC 4 programming language to the COBOL language, and began selling his version, LOPEZ COBOL, as a replacement for CHAMPION PACKER.3 In 1992, he formed a venture called HAL, Inc. with Ernest Parkin and Joe Herrin to develop and market a new freight software system that incorporated bar coding technology. Under their agreement, Lopez was to contribute LOPEZ COBOL for use in developing the new system; Parkin was to provide the system design and programming expertise; and Herrin was to supply industry expertise. Lopez, however, was detained in a Mexican jail for seven months during the initial stages of the software development project, and he was ousted from the company without recompense in March 1993. HAL continued to work on the software system and eventually began selling a freight tracking software system called "MEPAW."

Lopez claimed that MEPAW was an unauthorized copy of LOPEZ COBOL and that Parkin and Herrin had breached their obligation to compensate him for providing the LOPEZ COBOL system. Under threat of litigation from GUS, Lopez assigned his contract claim to GUS and agreed to assist GUS with a copyright infringement suit against HAL, Parkin, and Herrin. GUS filed its first suit against HAL on May 23, 1995 ("GUS 1"), raising various claims under federal and state law. The focus of the case, however, was GUS's claim that HAL had infringed its copyright in the CHAMPION PACKER system by copying the LOPEZ COBOL system. GUS sought damages, injunctive relief, and attorneys' fees.

On November 17, 1997, HAL filed a motion for partial summary judgment on the issue of copyright infringement, arguing that GUS could not establish that MEPAW copied the nonliteral elements of either LOPEZ COBOL or CHAMPION PACKER. In particular, HAL faulted GUS for failing to conduct an Altai analysis, the methodology used to analyze claims of nonliteral copying of software, asserting that without this analysis, GUS could not show that the MEPAW source code copied protectable elements of either CHAMPION PACKER or LOPEZ COBOL. In response, GUS argued that Altai did not apply to all of its infringement claims, such as its claims that HAL copied source code. GUS asserted that Altai applied only to claims that nonliteral elements of the software were copied. HAL in turn challenged GUS's source code infringement claim, arguing both that GUS had agreed that Altai analysis would be used to analyze the infringement claims and that GUS had never before put forth a theory of literal infringement.

A few months later, on February 13, 1998, HAL filed a second motion for partial summary judgment, this time focusing on the issue of copyrightability. HAL argued that GUS could not prove that it owned a valid copyright in the software because its copyright registration covered only data entry screens and record layout samples and not the software as a whole. HAL also noted that GUS had produced no evidence describing their software system as it existed in 1983, the date when Lopez allegedly copied it, because there was no extant copy of the 1983 version of GUS's CHAMPION PACKER software.4 Without such evidence, GUS could not prove that MEPAW copied its software system. GUS responded by pointing to its copyright registration for the freight forwarding software, which, it asserted, was prima facie evidence of its ownership of a copyright in the software. GUS also reminded the court that it owned LOPEZ COBOL, the system that HAL's MEPAW software copied.

On February 18, 1998, GUS filed its own motion for partial summary judgment in which it argued that it could prove as a matter of law that HAL had infringed its copyright in the freight forwarding software. Specifically, GUS claimed that it could prove that HAL had directly or literally copied portions of LOPEZ COBOL, noting in particular that the MEPAW system used fields, record layouts, and actual executable code taken directly from LOPEZ COBOL. As proof, GUS attached four exhibits which it asserted were examples of direct copying: (1) a database layout listing the layout of fields used by the MEPAW and LOPEZ COBOL systems; (2) a directory list comparing data entry fields used by the two software systems; (3) a print-out of a program from the MEPAW system indicating that it was created on October 25, 1983, by Lopez; and (4) invoices generated by the two systems. HAL challenged each of these exhibits, arguing that none established improper copying, and renewed its claim that GUS could not prevail under the Altai analysis.

A Magistrate Judge recommended granting HAL's first motion and dismissing the claims of nonliteral copyright infringement, but recommended denying HAL's second motion on copyrightability. The Magistrate Judge also recommended denying GUS's motion on literal copying, concluding that there were genuine issues of material fact precluding summary disposition.

The district court reviewed the record de novo, concluded that there was no plain error in the Magistrate Judge's Memorandum and Recommendation (M&R), and adopted it as its own. The court departed from the Magistrate Judge's recommendation with regard to GUS's summary judgment motion on literal copyright infringement. The court concluded that GUS had failed to raise a genuine issue of material fact on the issue and granted HAL summary judgment on the claim.

Having concluded that there was no copyright infringement, the court soon dismissed GUS's related Lanham Act, conversion, misappropriation, and trade secret claims. The sole remaining claim — Lopez's breach of contract claim against Parkin and Herrin — was tried to a jury in September 2000. After a three day trial, the jury found that Parkin and Herrin had breached their agreement with Lopez and awarded him $250,000 in damages plus $106,000 in attorneys' fees. However, the court also awarded HAL, Herrin, and Parkin a total of $448,928.73 in costs, attorneys' fees, and expenses incurred in successfully defending the copyright cause of action.

Approximately one year later, in September 2001, the court modified this judgment. First, the court concluded that Larry Lee, GUS's first attorney who withdrew as counsel-of-record in late 1997, was entitled to receive one-third of Lopez's judgment as attorneys' fees for work performed prior to his withdrawal as counsel. Lee had filed a complaint-in-intervention in January 1998, but GUS had failed to respond or deny the allegations in that complaint. Second, on September 26, 2001, the court set aside the jury verdict altogether and granted HAL, Herrin, and Parkin judgment as a matter of law on the contract claim. The court concluded that the evidence presented was legally insufficient to support the jury's verdict because Lopez had admitted, in a prior bankruptcy proceeding, that he did not own any interest in HAL.

HAL, Herrin, and Parkin filed motions seeking to enforce their judgment, including a motion for turnover relief and a motion to require GUS to post a $65,000 bond for appellate attorneys' fees. In June 2002, the district court granted HAL's motion requiring the posting of a $65,000 bond. The court later granted HAL's motion for turnover and sale of GUS's Lopez COBOL software, and we denied GUS's motion for an emergency stay.

In April 2001, nearly two years after the district court dismissed GUS's copyright claims on summary judgment, GUS filed a second, parallel suit against three of HAL's customers: Boaz Export Crating Co., Cargo Crating Co., and Dixie Box & Crating of Texas, Inc.5 GUS argued that, as purchasers of HAL's infringing software, Boaz had violated GUS's copyright in LOPEZ COBOL. In response, Boaz argued that GUS was collaterally estopped by the decision in the first suit from relitigating the copyright claim and moved to dismiss.

While this motion was pending, Parkin allegedly admitted in a hearing that HAL copied portions of LOPEZ COBOL. Without bringing this information to the court's attention, GUS responded to Boaz's motion, arguing that LOPEZ COBOL was not at issue in the prior suit and that collateral estoppel could not apply. The district court disagreed and granted Boaz's motion to dismiss.

Shortly thereafter, GUS filed a Rule 60(b) motion in the original suit against HAL. Pointing to Parkin's admission of copying, GUS argued that the judgment was procured through misconduct. The district court rejected this argument, finding neither perjury nor misconduct on Parkin's part. GUS appeals the court's rejection of its Rule 60(b) motion, and also argues that Parkin's alleged perjury should bar the application of collateral estoppel in the Boaz suit.

GUS's primary focus on appeal is the court's grant of summary judgment on the copyright issue. GUS also protests the dismissal of its trade secret, misappropriation, and Lanham Act claims; the award of fees to HAL and Boaz based on their successful defense against GUS's copyright charges; the court's grant of judgment as a matter of law on the contract claim; the order requiring the posting of a bond, and the turnover order.

We will address each of these issues in turn.

At the heart of GUS's appeal is its claim that the district court erred in granting HAL summary judgment on the issue of copyright infringement. GUS protests the court's decision on a number of grounds. First, GUS disputes the court's dismissal of its copyright claims under Altai, arguing that this doctrine does not apply to claims of source code copying. Second, GUS argues that the court impermissibly granted summary judgment on the issue of literal source code copying even though the Magistrate Judge admitted that there were genuine issues of material fact on this issue. Third, GUS contends that the court imposed an improper legal standard by holding that literal infringement requires that the two computer programs be virtually identical. Finally, GUS protests several other related decisions, such as the court's award of fees to HAL and its turnover order.

* We review the district court's grant of summary judgment de novo.6 Summary judgment should be granted only if there is "no genuine issue as to any material fact" and the moving party is entitled to judgment as a matter of law.7 In determining whether there is a genuine issue of material fact, we review the evidence and the inferences to be drawn therefrom in the light most favorable to the non-moving party.8 The moving party bears the initial burden of demonstrating that summary judgment is appropriate. If the moving party meets this burden, then the "nonmovant must go beyond the pleadings and designate specific facts showing that there is a genuine issue for trial."9 In conducting our review of the summary judgment grant, however, we examine only the judgment of the district court and are not bound by its rationale. We may affirm on any grounds argued below and supported by the record.10

To establish a prima facie case of copyright infringement, a copyright owner must prove "(1) ownership of a valid copyright, and (2) copying [by the defendant] of constituent elements of the work that are original."11 A certificate of registration, if timely obtained, is prima facie evidence both that a copyright is valid and that the registrant owns the copyright. There is no contention that GUS does not own a valid copyright in CHAMPION PACKER and LOPEZ COBOL. Rather, the issue is whether GUS provided sufficient evidence of copying to survive summary judgment under the second part of the infringement test.

To prove actionable copying under the second prong, a plaintiff must make two showings. First, the plaintiff must, as a factual matter, prove that the defendant "actually used the copyrighted material to create his own work."12 A plaintiff may make this showing either with proof of direct evidence of copying or through circumstantial evidence demonstrating both (1) that the defendant had access to the copyrighted work and (2) that the two works are "probatively" similar.13 The access element is satisfied if the person who created the allegedly infringing work had a reasonable opportunity to view the copyrighted work.14 The second element — probative similarity — requires a showing that the works, "when compared as a whole, are adequately similar to establish appropriation."15 In some cases, factual copying may be proven without a showing of access "[i]f the two works are so strikingly similar as to preclude the possibility of independent creation."16

If the plaintiff demonstrates factual copying, he must next demonstrate that the copying is legally actionable by showing that the allegedly infringing work is substantially similar to protectable elements of the infringed work. "[A] side-by-side comparison must be made between the original and the copy to determine whether a layman would view the two works as `substantially similar.'"17 Typically, the question whether two works are substantially similar should be left to the ultimate factfinder, but "summary judgment may be appropriate if the court can conclude, after viewing the evidence and drawing inferences in a manner most favorable to the nonmoving party, that no reasonable juror could find substantial similarity of ideas and expression."18

It is settled that computer programs are entitled to copyright protection.19 This protection extends not only to the "literal" elements of computer software — the source code and object code20 — but also to a program's nonliteral elements, including its structure, sequence, organization, user interface, screen displays, and menu structures.21 To assess a claim of software infringement, we have generally endorsed the "abstraction-filtration-comparison" test first outlined by the Second Circuit in Altai and refined by the Tenth Circuit in Gates Rubber Co. v. Bando Chemical Industries, Ltd.22 The Altai test deploys a three-step procedure to assess whether protectable expression has been improperly copied. First, at the abstraction stage, the court "dissect[s] the allegedly copied program's structure and isolate[s] each level of abstraction contained within it." Second, the court filters out unprotectable expression by examining the structural components at each level of abstraction to determine whether they can be protected by copyright. Copyright protection does not extend to ideas, processes, facts, elements dictated by considerations of efficiency, elements required by factors external to the program itself, or items taken from the public domain.23 With these nonprotectable elements filtered out, there remains a "core of protectable expression,"24 and we must then "determine whether the defendant's program misappropriated substantial portions of the plaintiff's program."25

GUS's first argument on appeal focuses on the district court's use of the Altai test. GUS asserts that Altai applies only to claims that an infringer has copied a computer program's nonliteral elements, like its structure, sequence, organization, and user interface, and not to claims that source code was copied.26 GUS urges, therefore, that the court erred in relying on Altai to dismiss its source code copying claims.

Although we have generally endorsed the Altai test to evaluate claims that nonliteral software elements were copied, we have not explicitly addressed whether that test should be used to evaluate charges that a program's source or object code was copied. We need not answer this question today, however, because contrary to GUS's claim, the district court did not use the Altai test to decide any source code copying claims. Rather, the court used Altai only in its decision to dismiss claims that nonliteral elements were copied. As the court stated in its order, "[a]bsent a valid Altai analysis to establish unlawful copying of the non-literal aspects of the computer programs, Defendants are entitled to summary judgment."27

Nor did the district court err in dismissing these claims of nonliteral copying. On filing this suit, GUS alleged that HAL copied nonliteral aspects of CHAMPION PACKER and LOPEZ COBOL. When HAL filed a summary judgment motion on this issue, arguing essentially that GUS could not prove that MEPAW copied any nonliteral elements, GUS responded by claiming that source code was copied. However, nowhere in its response did GUS present evidence disputing HAL's assertion that summary judgment was appropriate on the issue of the copying of nonliteral program elements. More to the point, nowhere in its submission did GUS complete the Altai analysis necessary to evaluate claims that a program's nonliteral elements were copied. Without this analysis, there was no evidence in the record to support GUS's claims that nonliteral elements were copied, and the district court properly dismissed those claims.

We find no error in the district court's use of Altai or in its decision granting HAL summary judgment on the issue of nonliteral infringement.

Turning from Altai, we next address GUS's arguments that the court applied an overly restrictive test for evaluating its claim that source code was copied. GUS presents two distinct arguments. First, GUS urges that there were genuine issues of material fact. Second, GUS argues that the court imposed an overly rigid standard for literal copying by requiring that the infringed program be virtually identical, rather than substantially similar, to the copyrighted program.

We reject GUS's arguments and conclude that summary judgment was appropriate.

* To evaluate GUS's claim, some background is necessary. Shortly after HAL filed its summary judgment motion on the nonliteral copying issue, GUS filed its own motion for summary judgment on the issue of source code copying. In this motion, GUS argued for the first time that HAL "directly" copied LOPEZ COBOL's source code. GUS attached four exhibits as evidence of this copying: (1) a printout that purported to be a "database layout" listing the names of fields used by the MEPAW and LOPEZ COBOL systems; (2) a directory list giving the names of data entry fields used by the two software systems; (3) a print-out of a program contained in MEPAW that was created on October 25, 1983, by Jose Lopez; and (4) print-outs of invoices generated by the LOPEZ COBOL and MEPAW systems. GUS argued that the first three of these exhibits constituted evidence of "direct, line by line copying." For the fourth exhibit, GUS argued that the layouts of the MEPAW and LOPEZ COBOL invoices were almost exactly the same, suggesting that the source code which generated them must also be strikingly similar.

HAL filed a lengthy response to GUS's motion, challenging each of these exhibits and accusing GUS of misrepresenting the facts. Specifically, HAL attached an affidavit from Parkin indicating that the first exhibit was a database layout that Lopez gave to Defendant Al Parkin to enhance; the second exhibit was not used in MEPAW; the third exhibit was a program Parkin obtained while he worked for the General Services Administration and was not authored by Lopez; and the fourth exhibit was a generic invoice of a type found in every packing system. HAL also attached several affidavits from other parties, some of which attested to the substantial differences between the two software packages while others disputed GUS's contention that HAL had access to the CHAMPION and LOPEZ COBOL programs. HAL ended its response with a charge that there was no evidence supporting GUS's infringement claims and with a request that the court dismiss the action in its entirety. GUS did not reply to HAL's motion.

On the issue of substantial similarity, a Magistrate Judge concluded that the four exhibits GUS submitted were inadequate to sustain a finding as a matter of law that MEPAW and LOPEZ COBOL were substantially similar. She concluded, however, that a genuine issue of material fact existed as to whether the software systems were substantially similar. Neither party objected to these findings.

The district court conducted a de novo review of the record, concluded that there was no plain error in the Magistrate Judge's Memorandum and Recommendation, and adopted the report as its own. However, the court employed a different analysis of the infringement issue. The court noted, first, that to prove literal copying, a plaintiff must show identical or virtually identical copying of substantial portions of a program. Second, the court stated that such identical copying would not generally be present when the two programs are written in a different programming language. The court concluded that GUS had failed to present evidence raising a genuine issue of material fact, denied GUS's summary judgment motion, and granted summary judgment to HAL.

We are ultimately persuaded that summary dismissal was appropriate because of GUS's failure to adduce evidence to support its charges of copying and substantial similarity.28

As a preliminary matter, we reject GUS's contention that the district court's decision to grant summary judgment was procedurally improper — that the court granted summary judgment to HAL sua sponte, without a motion for summary judgment by HAL before it. In response to GUS's summary judgment motion, HAL asserted that it was entitled to a "finding[ ] that the plaintiffs are unable to prove both non-literal copying under the Altai analysis, and direct copying." Citing GUS's failure to adduce any evidence supporting its claims of source code copying, HAL charged that GUS had no evidence of copying and urged the district court to dismiss the entire case. While not labeled as a cross-motion for summary dismissal, HAL sought summary judgment on the issue of literal copying and requested the court to act accordingly. The court's action was not improper.29

We are persuaded that summary judgment was appropriate because GUS presented no evidence supporting its claims of literal copyright infringement. At trial, GUS would bear the ultimate burden of proving that MEPAW copied LOPEZ COBOL. In response to HAL's charge that no evidence supported GUS's claim, GUS was required to come forward with some evidence supporting the essential elements of its claim. GUS failed to do so.30 The four isolated pieces of evidence GUS produced fail on the most basic level. To prevail on a claim of source code copyright infringement, GUS would have to prove that MEPAW's source code is substantially similar to LOPEZ COBOL's source code. To do so, "a side-by-side comparison must be made between the original and the copy."31 GUS, however, failed to attach any of its own source code to its summary judgment motion or to compare MEPAW source code to LOPEZ COBOL source code, despite its conclusory assertions that the four exhibits were evidence of direct copying.32 Without providing its own source code for comparison, GUS did not satisfy the requirement that the infringed and infringing work be compared side-by-side. Perhaps there was relevant LOPEZ COBOL source code buried deep in the record somewhere,33 and perhaps the district court could have waded through the record to find code that looked similar to the exhibits GUS attached. But the court was not required to do so. In response to HAL's motion, the burden was on GUS to prove that it could support its claims. It did not meet this burden.

GUS's difficulty stems, ultimately, from the muddled nature of its infringement claims. As the district court noted, when GUS first filed its infringement suit, it claimed only that HAL misappropriated nonliteral elements of CHAMPION PACKER, like its structure, sequence, and organization. GUS broadened its claims in its summary judgment motion, asserting that HAL directly copied source code from CHAMPION. The exhibits that GUS attached as evidence of source code copying, however, did not reflect the change in the nature of its claims. Two of the exhibits — the database layout and directory listing — are at best evidence that HAL copied nonliteral elements from LOPEZ COBOL; they do not provide evidence of source code copying.34 Likewise, the fourth exhibit — a comparison of invoice layouts — is a nonliteral element of the software programs; GUS's conclusory remark that the "similarity of layout in the invoices ... highlights the fact that the source code which generated both invoices is strikingly similar" is both factually questionable and legally insufficient to state a claim of source code copying. GUS had the MEPAW and LOPEZ COBOL source codes at its disposal; it should have supported its assertions with tangible references to these materials rather than with empty and conclusory statements. "[S]uch conclusory, unsupported assertions are insufficient to defeat a motion for summary judgment."35

Only the third exhibit contains any source code. It is a print-out of a software program GUS claims Lopez wrote on October 25, 1983. GUS's expert, Dr. Nassar, asserts that he found this program in the MEPAW system and that a "strikingly similar" copy exists in LOPEZ COBOL. GUS, however, fails to provide the copy from LOPEZ COBOL's source code for comparison. Nor does GUS refute HAL's denial that Lopez wrote the program and that HAL obtained properly from other channels.

In short, GUS presented insufficient evidence of source code copying to survive summary judgment. When GUS filed its motion for summary judgment, it did not need to marshal all its facts to support its infringement claims. But when HAL responded with an allegation that GUS could produce no evidence on basic elements of its claims, GUS was required to come forward with tangible evidence. It failed to do so, and summary judgment for HAL was appropriate.

We affirm the dismissal of GUS's claims of source code copying.36

After dismissing GUS's copyright claims, the district court awarded HAL attorneys' fees as the prevailing party. GUS contests the award of fees, arguing in particular that because it brought and litigated its copyright claims in good faith, an award of fees was inappropriate.

The decision to impose costs lies in the sound discretion of the district court,37 reviewable only for an abuse of discretion.38 This circuit adheres to the practice that such fees are "discretionary but routinely awarded."39 Under this approach, we cannot agree that the district court abused its discretion in awarding HAL fees. The court carefully considered GUS's conduct during the course of the litigation — conduct which in some cases bordered on the overzealous — and noted the tremendous burden that GUS's zealous representation placed upon HAL. GUS provides no persuasive reason why the award should be upset, and we therefore affirm the award of fees.40

GUS next disputes the district court's dismissal of its Lanham Act claims. GUS argued to the district court that HAL, by copying and marketing GUS's software as HAL's own, engaged in "reverse palming off" in violation of § 43(a) of the Lanham Act.41 The district court rejected GUS's claim, relying heavily on its earlier conclusion that GUS failed to present evidence of actionable copying. GUS contends on appeal that the court's decision was error for two reasons: first, because there is substantial evidence that MEPAW is a copy of LOPEZ COBOL that has only been modified slightly, and second, because a Lanham Act claim does not require the Altai-type analysis that the district court used to dismiss GUS's copyright claims. We find neither argument persuasive.

GUS's first argument depends entirely upon its copyright argument, which we have rejected. The district court committed no error in dismissing GUS's copyright infringement claims. GUS's first Lanham Act argument must fail as well.

GUS's second argument is more problematic. In just one sentence, GUS asserts that a Lanham Act claim does not require that a defendant prove "substantial similarity" through the type of Altai-analysis that the district court demanded in this case. GUS offers no argument or explanation on this point, and cites no authority for its statement. Failing to adequately brief this contention, GUS has waived it.42

Even were we inclined to flesh out GUS's argument, its Lanham Act claim would face the Supreme Court's recent decision in Dastar Corp. v. Twentieth Century Fox.43 Dastar revolved around a 1949 television series called "Crusader in Europe," which was based on General Eisenhower's written account of the Allied campaign in Europe. Twentieth Century Fox owned the exclusive rights in the television series, but it failed to renew its copyright registration, and the copyright expired in 1977, leaving the series in the public domain. In 1995, Dastar purchased copies of the original, public domain television series, and then copied and edited them. In particular, Dastar substituted a new opening sequence, credit page, and final closing, and made various other similar changes. Fox and several licensees brought suit, alleging, inter alia, that Dastar's sale of its version of the series without proper attribution to the Crusade series constituted "reverse passing off" in violation of § 43(a) of the Lanham Act.

The Court rejected the plaintiffs' claim, concluding that the term "origin" in § 43(a) applies only to "the producer of the tangible goods that are offered for sale, and not to the author of any idea, concept, or communication embodied in those goods."44 In so doing, the Court carefully distinguished Lanham Act claims from copyright claims; the former "were not designed to protect originality or creativity," while the latter were. The Court concluded that claims of false authorship and reverse passing off, when raised to protect an author's interest in the intellectual content of communicative products, were not actionable under § 43(a) and should instead be pursued under copyright law. "To hold otherwise would be akin to finding that § 43(a) created a species of perpetual patent and copyright, which Congress may not do."45

While there are some differences between Dastar and the situation at hand, we find Dastar's reasoning controlling. GUS has not accused HAL of taking tangible copies of its software, removing its trademarks, and selling them as its own. Rather, GUS asserts that HAL copied the ideas, concepts, structures, and sequences embodied in its copyrighted work. In sum and substance, GUS's claim is simply a claim that HAL has infringed its copyright in LOPEZ COBOL. Dastar makes clear that such claims are not actionable under § 43(a).

The district court's grant of summary judgment to HAL on GUS's Lanham Act claims is affirmed.

GUS next contests the district court's dismissal of its state law claim of trade secret theft. Shortly before trial, HAL filed a summary judgment motion on this issue, arguing that GUS could not prove all of the essential elements of its claim. The district court agreed, disposing of GUS's claim on two alternative grounds. First, the court concluded that Texas law requires that the claimed trade secret have been discovered through improper means. The court found no improper means in this case because the HAL Defendants obtained access to the LOPEZ COBOL System pursuant to a written agreement with Lopez. Second, the court observed that Texas law requires that a person asserting a trade secret must take reasonable precautions to protect the claimed trade secret. The court found that the undisputed evidence showed that GUS had failed to do so. Based on these two alternative grounds, the court concluded that LOPEZ COBOL, "was not properly protected as a trade secret."

GUS argues that the court applied an outmoded and overly restrictive test for trade secret misappropriation. We agree.

* We review a district court's interpretation of state law de novo.46 To state a claim for trade secret misappropriation under Texas law, a plaintiff must (1) establish that a trade secret existed; (2) demonstrate that the trade secret was acquired by the defendant through a breach of a confidential relationship or discovered by improper means; and (3) show that the defendant used the trade secret without authorization from the plaintiff.47 Only the first and second factors of this test are at issue in this appeal. "The existence of a trade secret is properly considered a question of fact to be decided by the judge or jury as fact-finder."48

The district court first concluded that GUS could not prevail under the first prong of the trade secret theft test. The court held that Texas law requires that a person asserting a trade secret take reasonable precautions to protect it.49 The district court concluded that Lopez did not take reasonable measures in this case, and concluded that LOPEZ COBOL was not properly protected as a trade secret.

The Texas Supreme Court recently clarified that to determine whether there is a trade secret protected from disclosure or use, a court must examine six relevant but nonexclusive criteria: (1) the extent to which the information is known outside the business; (2) the extent to which it is known by employees and others involved in the business; (3) the extent of measures taken to safeguard the secrecy of the information; (4) the value of the information to him and to his competitors; (5) the amount of effort or money expended in developing the information; and (6) the ease or difficulty with which the information could be properly acquired or duplicated by others.50 The court expressly held that "the party claiming a trade secret should not be required to satisfy all six factors because trade secrets do not fit neatly into each factor every time."51 Determining whether any given piece of information is entitled to trade secret protection, then, is a contextual inquiry which must evaluate a number of factors.

The district court did not engage in this broad inquiry, but instead focused solely on Lopez's alleged failure to take "reasonable precautions" to protect LOPEZ COBOL, pointing to Lopez having allowed Parkin to copy LOPEZ COBOL onto a personal computer and take it with him to Arizona. The court, however, overlooked the fact that Lopez and Parkin were engaged in a joint venture to exploit that very software. There was also uncontroverted evidence that Lopez carefully secured his software system from parties other than Herrin and Parkin — that is, that Lopez took reasonable precautions to protect LOPEZ COBOL from persons other than those selected by Lopez to have access for limited purposes.

Further factual development may shed light on whether trade secret protection is appropriate. We conclude, therefore, that a genuine dispute of material fact precludes determination whether LOPEZ COBOL was properly protected as a trade secret.

Summary judgment may yet have been appropriate if the district court's second basis for judgment independently bars GUS's claim. As an alternative basis for dismissing GUS's trade secret claim, the court concluded that GUS could not satisfy the second prong of the trade secret claim. Specifically, the court concluded that HAL did not acquire LOPEZ COBOL through improper means.

However, Texas trade secret law does not impose liability only when improper means are used. Under Texas law, there is liability for trade secret misappropriation if either "(a) he discovers the secret by improper means, or (b) his disclosure or use [of the trade secret] constitutes a breach of confidence reposed in him by the other in disclosing the secret to him."52 The district court did not consider whether there was a confidential relationship among Lopez, Perkins, and Herrin that imposed upon them a duty not to use the trade secret. There may have been an appropriate confidential relationship: although the facts available in the record are unclear, it appears that Lopez, Perkins, and Herrin may have formed a partnership to exploit the LOPEZ COBOL system.53 Under Texas law, a partnership can be considered a confidential relationship, and participants in a joint venture are often held to owe duties to one another.54

HAL, for its part, vigorously asserts that there was no confidential relationship among the parties.55 HAL also contends that the district court reviewed the entire record and found that HAL's use of LOPEZ COBOL was unrestricted, an assertion which finds no support in the text of the district court's opinion. In the end, HAL's argument highlights the presence of disputed material facts.

The district court erred in not considering whether GUS might prevail by demonstrating a breach of a confidential relationship. On the present record, there is a genuine issue of material fact and summary judgment was inappropriate.

We reverse the district court's grant of summary judgment in favor of HAL on GUS's claim of trade secret theft.

GUS's next set of arguments focuses on its contract claims.56 Unlike its copyright claims, GUS tried its contractual allegations to a jury, which ruled in GUS's favor. Over two years after the jury returned its verdict, the district court granted HAL judgment as a matter of law. GUS appeals this decision, and also appeals a decision by the district court to limit the manner in which GUS could prove contractual damages.

Finding no reversible error on either issue, we affirm.

* GUS first contends that the court abused its discretion in excluding evidence of HAL's sales or the value of the market for HAL's software at the time of trial. Specifically, the court refused to allow GUS to question defense witnesses regarding statements on HAL's website that the market for HAL's software was one billion dollars per year. The court also refused to allow GUS to introduce evidence that Parkin copied LOPEZ COBOL to create MEPAW, HAL's primary asset. GUS also contends that the court abused its discretion in instructing the jury that Lopez's damages were to be measured at the time of HAL's breach, rather than at the time of trial, and in refusing to charge the jury that Lopez might be entitled to specific performance.

* We examine first the court's contested evidentiary rulings. We review a district court's exclusion of evidence for abuse of discretion.57

As a preliminary matter, we reject GUS's contention that the court's refusal to allow evidence of the value of the market for HAL's software was error. This evidence was irrelevant to determining the value of HAL either at the time of the breach or at trial. Similarly, we find no error in the court's exclusion of evidence that HAL copied LOPEZ COBOL to create MEPAW. This is admittedly a closer question because evidence that HAL used LOPEZ COBOL might serve as evidence of the value of his contribution to the joint venture or possibly even the value of HAL at the inception of the partnership. However, the issue before the court was solely whether HAL breached its agreement with Lopez, not whether HAL copied LOPEZ COBOL or whether Lopez was entitled to restitution for the value of his contribution. Given the limited focus of the trial and the nature of GUS's breach of contract claim, we find no abuse of discretion in the district court's ruling.

A somewhat more difficult question is raised by GUS's twin assertions that Lopez's contractual damages should have been measured as of the time of trial, rather than the time of breach, and that Lopez should have received specific performance instead of damages. HAL responds that GUS's claim for damages measured as of trial is nothing more than a request for specific performance, which is disfavored under Texas law and not merited in this case.58 The district court rejected GUS's request for specific performance and limited GUS's recovery to damages calculated as of the time of the breach. The jury, in turn, awarded Lopez $250,000, a sum which corresponded to his testimony regarding the amount he expended to create LOPEZ COBOL pursuant to his contract with Parkin and Herrin.

We review jury instructions for abuse of discretion.59 We ask "`whether the court's charge, as a whole, is a correct statement of the law and whether it clearly instructs jurors as to the principles of the law applicable to the factual issues confronting them.'"60 We review the district court's interpretation of Texas law de novo.61 Where a party argues on appeal that the district court erred in refusing to give a proffered jury instruction, that party must "show as a threshold matter that the proposed instruction correctly stated the law."62

We find no error in the district court's decision rejecting GUS's request of specific performance. Under Texas law, specific performance is an equitable remedy that is normally available only when the complaining party cannot be fully compensated through the legal remedy of damages or when damages may not be accurately ascertained.63 When one party breaches an agreement to provide stock, Texas courts have consistently held that the damages remedy is adequate, and GUS offers no argument why specific performance was warranted under the facts of this case.64 We reject this contention.

We also reject GUS's claim that its damages should have been calculated as of the time of trial rather than at the time of breach. "[T]he rule in Texas has long been that contract damages are measured at the time of breach, and not by the bargained-for goods' market gain as of the time of trial."65 Texas courts have applied this rule in cases involving a failure to provide stock. In Walden v. Affiliated Computer Services, Inc., for example, the Texas Court of Appeals held that damages in a breach of contract claim brought by stock option holders against a corporation should have been calculated based on the value of the corporation's stock on the first day after the corporation's breach, not the value of the stock at the time of trial.66 The court explained that, "[i]n a case involving a contract to deliver stock, the proper measure of damages for breach of that contract is the same as for other contracts: the difference between the price contracted to be paid and the value of the article at the time when it should have been delivered."67

The Texas Supreme Court reinforced this rule in Miga v. Jensen.68 Miga sought to recover damages for the value of the stock that Jensen refused to sell him in 1994 plus his lost profits. The only evidence of lost profits, however, was the increased market value of the stock at the time of trial. The jury awarded him damages based on that increase, but the Texas Supreme Court reversed. Reiterating the rule that contract damages are measured at the time of breach, the Court concluded that the damages Miga was awarded under the rubric "lost profits" was actually the market gain, to which he was not entitled. In this case, GUS is attempting to recover precisely what Miga held is not recoverable under Texas law.69 We find no error in the district court's decision limiting GUS's recovery to damages as of the time of the breach.

We next review the court's decision granting HAL judgment as a matter of law on the contract claim. The district court based its decision on two independent grounds. First, the court concluded that statements made by Lopez in a prior bankruptcy proceeding were judicial admissions that barred his contract claim.70 The court concluded that these statements constituted judicial admissions and revealed that Lopez had no ownership interest in HAL, Inc. Second, the court concluded that the evidence was legally and factually insufficient to support the jury's findings of breach of contract and damages.

GUS presents two arguments against the district court's invocation of judicial estoppel: first, that HAL waived the judicial estoppel argument by not raising it before the district court, and second, that Lopez's prior statements were not judicial admissions. There is a strong argument that the court abused its discretion in treating these statements as judicial admissions: the jury heard the statements Lopez made in his bankruptcy filing but accepted Lopez's explanation at trial, and neither HAL nor the court itself ever thought to treat these statements as judicial admissions until over two years after trial.

We need not decide that issue, however, because GUS fails to overcome the court's second basis for granting JMOL — that the evidence was legally insufficient to support the verdict. Indeed, GUS presents no argument on this issue at all: it focuses its entire argument on the court's invocation of judicial estoppel.71 By failing to advance arguments in the body of its brief supporting its claim on appeal, GUS has abandoned this issue.72 Since the insufficiency of the evidence provides an independent foundation for the court's order, and GUS fails to refute it, the order must stand.

Even were we to reach the issue, we find no error in the district court's conclusion on this issue. At trial, the only evidence of damages presented was Lopez's testimony that he spent $250,000 in the development of LOPEZ COBOL. The jury clearly seized upon this figure and granted him damages in that amount. However, while this testimony may have been evidence of his investment in the venture, it did not speak to the value of the company itself at the time of breach, and it thus provided no basis for the award of damages for breach of contract. To recover on its contract claim, GUS was required to prove damages. To do so, it had to introduce evidence showing the value of the stock that HAL wrongfully withheld from Lopez. It could have done so in many ways — by proving the market value of that stock, or, since this was a closely corporation, by proving the market value of the assets of the company after deducting its liabilities73 — but it did not do so. Instead, it introduced evidence of Lopez's investment in HAL and in effect received restitution for that investment, a measure of damages neither pled nor argued nor merited under the facts of the case. This was insufficient, and JMOL was proper.

Finding no error, we affirm the district court's order granting HAL judgment as a matter of law.74

After losing its copyright claim against HAL, GUS filed suit against several of HAL's customers (collectively "Boaz") for copyright infringement based on their use of MEPAW. Boaz moved to dismiss the suit, arguing that the HAL case conclusively decided the issue of MEPAW's infringement and that collateral estoppel barred GUS from re-litigating the same issue. The district court agreed and dismissed the case.

Events that occurred in that proceeding, however, form the basis of several of GUS's claims on appeal. According to GUS, HAL made two admissions in the Boaz suit that MEPAW copied portions of LOPEZ COBOL. Specifically, GUS alleges that Parkin testified during a hearing that MEPAW copied certain elements of LOPEZ COBOL, a claim that Professor Davis, an expert hired by HAL, essentially reiterated in an expert report. Armed with these alleged admissions, GUS filed a Rule 60(b) motion for relief from judgment in the original HAL suit, seeking to overturn the original judgment. GUS argued that the admissions were evidence that HAL committed fraud and perjury in the first suit and that GUS was entitled to postjudgment relief. The district court denied GUS's motion, and GUS appeals that decision.

For the same basic reason, GUS challenges the district court's use of collateral estoppel in the Boaz suit, contending that the HAL suit does not merit collateral estoppel effect in light of Parkin's admission of copying. GUS also contests the court's decision granting Boaz attorneys' fees, arguing that the case was dismissed on a procedural technicality and that an award of fees is inappropriate.

* We first examine GUS's arguments concerning the denial of its Rule 60(b) motion. GUS contends that, given HAL's misconduct, relief is justified on two grounds: first, because HAL committed fraud and perjury within the meaning of Rule 60(b)(3), and second, because Parkin's and Davis's admissions constituted "newly discovered evidence" under Rule 60(b)(2). We find neither basis persuasive.

* Rule 60(b)(3) allows a district court to relieve a party from a final judgment if the adverse party committed fraud, misrepresentation, or other misconduct.75 "This subsection of the Rule is aimed at judgments which were unfairly obtained, not at those which are factually incorrect."76 To merit relief, the complaining party must "establish by clear and convincing evidence (1) that the adverse party engaged in fraud or other misconduct and (2) that this misconduct prevented the moving party from fully and fairly presenting his case."77 Misconduct may be shown by evidence that the opposing party withheld information called for by discovery78 or willfully committed perjury.79

Determining whether a party has made a sufficient showing to warrant relief lies in the sound discretion of the district court. We therefore review the court's decision only for an abuse of discretion.80 We find no error here.

GUS bases its argument on two isolated statements from the Boaz record, but neither is sufficient to demonstrate misconduct. First, during a hearing in the Boaz case, Parkin stated on cross-examination that "[s]ome of the data fields in the purchase order system was supplied by Mr. Lopez. I have never denied that. Didn't deny it in the other case." Second, Professor Davis stated in an expert report that "[i]t is my understanding that as one consequence of this agreement [between Parkin, Lopez, and Herrin], Mr. Parkin used some components of Lopez COBOL code in the creation of MEPAW, with the full knowledge, permission, and indeed the active cooperation of Mr. Lopez." GUS urges that these two statements reveal that HAL did indeed copy LOPEZ COBOL and that HAL's representations to the contrary in the original suit were false. We disagree.

First, GUS misunderstands what was at issue in the first case and why its copyright charges were dismissed. As we explained above, to prove copyright infringement, a plaintiff must demonstrate factual copying — that the defendant actually copied the software — and then actionable copying — that the allegedly infringing work is substantially similar to protected elements of the infringed work. GUS's copyright claims were dismissed because it failed to adduce evidence supporting the second prong of this test, not because GUS could not prove that HAL had actually copied MEPAW. It was assumed, for purposes of the summary judgment motion, that factual copying had occurred. Yet the evidence that GUS points to now — Parkin's statements — is at best evidence of factual copying, if it is even that. It says nothing about whether such copying was actionable. In short, it does nothing to unsettle the reasoning behind the district court's dismissal of GUS's copyright claims or to correct the fundamental deficiency in GUS's infringement argument.81 GUS had every opportunity in the prior case to prove that MEPAW was substantially similar to LOPEZ COBOL, but it failed to do so. It must live with that failure; it may not use a Rule 60(b) motion as an occasion to relitigate its case.

Moreover, GUS overstates the import of Parkin's and Davis's statements by taking them out of context and misrepresenting the conflict with the position taken by HAL in the original suit. GUS, for example, fails to report that Parkin also stated repeatedly in that hearing that HAL did not copy LOPEZ COBOL, that it only used information and expertise supplied by Lopez, and that Lopez consented to this use. All of these statements were consistent with statements made by HAL in the original litigation.82 Certainly, the district court, which had the benefit of observing Parkin's testimony and the testimony of other witnesses in the Boaz trial, concluded that there was no such conflict, and we cannot say the court abused its discretion in reaching that conclusion. To merit relief, GUS was required to prove misconduct with clear and convincing evidence. It failed to do so.

Under Rule 60(b)(2), a losing party may seek relief from a judgment because of "newly discovered evidence which by due diligence could not have been discovered in time to move for a new trial under Rule 59(b)." As with Rule 60(b)(3) motions, the decision whether to grant relief lies within the sound discretion of the court.83 "To succeed on a motion for relief from judgment based on newly discovered evidence, our law provides that a movant must demonstrate: (1) that it exercised due diligence in obtaining the information; and (2) that the evidence is material and controlling and clearly would have produced a different result if present before the original judgment."84 Moreover, "[t]he newly discovered evidence must be in existence at the time of trial and not discovered until after trial."

GUS has not made a sufficient showing.

First, as the district court noted, the newly discovered evidence was not available at the time of trial: Parkin's statement occurred in a hearing held on November 16, 2001, and Davis's report was prepared on November 1, 2001. These statements, then, were created after the entry of final judgment on September 27, 2001.

More importantly, GUS has not shown either that the evidence is material or that it would have produced a different result in the original case. Indeed, the precise opposite is apparent. As noted above, GUS's infringement claims were dismissed because GUS failed to provide evidence of substantial similarity. Parkin's and Davis's bare statements do nothing to unsettle this holding.

Finding no abuse of discretion, we affirm the district court's rejection of GUS's motion for relief from judgment.

For similar reasons, we also reject GUS's argument that the district court erred in invoking collateral estoppel to dismiss the Boaz suit. GUS argues, again, that the prior HAL judgment was procured by fraud and perjury which prevented GUS from fully and fairly presenting its case against HAL. GUS further argues that Parkin's and Davis's admissions constitute new evidence that distinguishes the factual basis of the HAL and BOAZ suits, making collateral estoppel inappropriate. However, GUS failed to present any of these arguments to the district court, and it has waived its right to present them here.86

GUS urges that we should nonetheless consider its argument and overrule the district court's finding of collateral estoppel in order to avoid a gross miscarriage of justice.87 GUS suggests — without argument or justification — that such an exception is appropriate in this case. We rejected above GUS's claim that its new evidence tainted the original proceeding.

Finding no error, we affirm the district court's invocation of collateral estoppel in the Boaz suit.

GUS next argues that the district court abused its discretion in awarding attorneys' fees to Boaz on its defense of the copyright claims. As we noted above, the decision to impose costs to the prevailing party in a copyright infringement action lies in the sound discretion of the district court.

We find no abuse of discretion in this case.

The district court carefully considered the issue in the Boaz case and applied the factors endorsed by the Supreme Court in Fogerty v. Fantasy.89 In Fogerty, the Supreme Court quoted with approval certain nonexclusive factors identified by the Third Circuit as particularly relevant to the purposes of a fee award under the Copyright Act. These factors include the frivolousness of the action, the party's motivation in bringing it, the objective unreasonableness in the factual and legal components of the case, and the need in particular circumstances to advance considerations of compensation and deterrence.90 All of the factors support an award of fees in this case. After losing its copyright suit against HAL, GUS turned around and filed suit against HAL's customers, raising on the same basic copying issues. To quote the district court, GUS's action was "little more than an attempt ... to get a second bite at the copyright apple."

We affirm the court's award of fees.

For the foregoing reasons, we affirm the decisions of the district court, reversing only the district court's dismissal of General Universal's trade secret claim in Case Number 01-21114. We remand that claim to the district court for further proceedings not inconsistent with this opinion. All outstanding motions are denied.