From Casetext: Smarter Legal Research

Antonick v. Electronic Arts Inc.

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF CALIFORNIA
Jan 22, 2014
No. C 11-1543 CRB (N.D. Cal. Jan. 22, 2014)

Opinion

No. C 11-1543 CRB

01-22-2014

ROBIN ANTONICK, Plaintiff, v. ELECTRONIC ARTS INC., Defendant.


ORDER GRANTING EA'S AMENDED

RENEWED PHASE II MOTION FOR

JUDGMENT AS A MATTER OF

LAW, OR, ALTERNATIVELY, FOR A

NEW TRIAL

Following a two-phase trial in which the jury found EA breached its contract with Plaintiff Robin Antonick, the parties filed several post-trial motions, which are now before the Court. Defendant Electronic Arts ("EA") moves for judgment as a matter of law as to both phases of the trial or, in the alternative, for a new trial. See generally dkts. 443, 540. Antonick moves for prejudgment interest on the royalties he is owed as a result of the Phase II verdict, see dkt. 528, and for entry of final judgment as to his fraud claim, see dkt. 542. Upon consideration of the motions, the oppositions thereto, and the entire record of the case, the Court DENIES EA's motion for judgment as a matter of law as to Phase I, GRANTS EA judgment as a matter of law as to Phase II, and conditionally GRANTS EA's motion for a new trial of Phase II.

I. BACKGROUND

In 1984, EA hired Antonick to write source code for a video game to be called "Football." After EA negotiated with John Madden to use his name and likeness for the game, Antonick and EA entered into a new contract to develop a game for the Apple II computer called John Madden Football. Trial Ex. 15 (1986 Contract). The Contract also gave Antonick the right to royalties on "Derivative Works," defined as "any computer software program or electronic game which . . . constitutes a derivative work of the Work within the meaning of the United States Copyright law." Id. § 1.03.

In 2011, Antonick brought this suit against EA alleging breach of contract and fraud based on EA's failure to pay him royalties on allegedly derivative works. See generally Compl. (dkt. 1). Following a motion to dismiss and three motions for summary judgment, the Court held a jury trial in two phases. In Phase I, the jury was asked to decide whether the statute of limitations barred Antonick's claims. At the close of evidence, the jury found that Antonick proved by a preponderance of the evidence that before November 21, 2005, he did not discover, and did not know of facts that would cause a reasonable person to suspect, that EA had allegedly breached the 1986 Contract or made the allegedly fraudulent statements. See Verdict (dkt. 441).

In Phase II, the same jury was asked to determine whether Antonick proved that there are substantial similarities between Sega Madden and Apple II Madden with respect to the expression in the source code of (1) field width or (2) plays and formations. Then, for any Sega Madden game for which the jury found substantial similarities in either of those elements, the jury was asked to determine whether Antonick proved that that particular Sega Madden game is virtually identical to Apple II Madden when considering the games as a whole. In response to Question 1, the jury found that Antonick did not prove that there are substantial similarities in the expression of field width in the source code, but did prove there are substantial similarities between the expression of the source code for plays and formations. See Verdict (dkt. 516) at 1. In Question 2, the jury found that each of the seven Sega Madden games at issue were "virtually identical" to Antonick's version. Id. at 2. By this verdict, the jury found that the Sega Madden games are derivative works under the 1986 Contract, and that EA breached the contract by failing to pay Antonick royalties on their sales.

For purposes of the jury instructions and verdict form, the Court instructed the jury that it would "refer to John Madden Football for the Apple II as 'Apple II Madden,' and . . . to all of the versions of John Madden Football for the Sega Genesis at issue in this case as 'Sega Madden.'" See Jury Instructions at 4-5.

EA now moves for judgment as a matter of law as to both phases of the trial under Federal Rule of Civil Procedure 50(b), or, in the alternative, for a new trial under Federal Rule of Civil Procedure 59. Antonick moves for prejudgment interest and for entry of final judgment as to his fraud claim, which the Court dismissed prior to Phase II.

II. LEGAL STANDARDS

A. Motion for Judgment as a Matter of Law under Federal Rule of Civil Procedure 50(b)

"Judgment as a matter of law is appropriate when the evidence, construed in the light most favorable to the nonmoving party, permits only one reasonable conclusion, which is contrary to the jury's verdict." Hagen v. City of Eugene, 736 F.3d 1251, 1256 (9th Cir. 2013) (quoting Omega Envtl., Inc. v. Gilbarco, Inc., 127 F.3d 1157, 1161 (9th Cir. 1997)). The court may grant a motion for judgment as a matter of law only where "there is no legally sufficient basis for a reasonable jury to find for [the non-moving] party on that issue." Krechman v. Cnty. of Riverside, 723 F.3d 1104, 1109-10 (9th Cir. 2013) (internal quotation marks omitted); see also, e.g., Byrd v. Maricopa Cnty. Sheriff's Dep't, 629 F.3d 1135, 1138 (9th Cir. 2011) ("[A] motion for judgment as a matter of law is properly granted only if no reasonable juror could find in the non-moving party's favor.") (internal quotation marks omitted).

If, however, "there is such relevant evidence as reasonable minds might accept as adequate to support the jury's conclusion," the motion should be denied. Hagen, 736 F.3d at 1256 (internal quotation marks omitted); see also Barnard v. Theobald, 721 F.3d 1069, 1075 (9th Cir. 2013). When considering a motion for judgment as a matter of law, the court may not make credibility determinations, weigh the evidence, or substitute its own view of the evidence for the jury's. See, e.g., Krechman, 723 F.3d at 1110.

B. Motion for a New Trial Under Federal Rule of Civil Procedure 59

The court "may grant a new trial if, having given full respect to the jury's findings, the judge on the entire evidence is left with the definite and firm conviction that a mistake has been committed . . . ." Tortu v. Las Vegas Metro. Police Dep't, 556 F.3d 1075, 1087-88 (9th Cir. 2009) (internal quotation marks and alteration omitted); see Fed. R. Civ. P. 59(a)(1)(A) (providing that district courts may grant a motion for a new trial "on all or some of the issues . . . after a jury trial, for any reason for which a new trial has heretofore been granted in an action at law in federal court."). "The primary basis for granting a new trial is that the jury's verdict was against the clear weight of the evidence." Computer Access Tech. Corp. v. Catalyst Enters., Inc., 273 F. Supp. 2d 1063, 1065 (N.D. Cal. 2003) (citing Landes Constr. Co., Inc. v. Royal Bank of Canada, 833 F.2d 1365, 1371 (9th Cir. 1987)). In contrast to its role with respect to a motion for judgment as a matter of law, when considering a motion for a new trial, the court "can weigh the evidence and assess the credibility of witnesses, and need not view the evidence from the perspective most favorable to the prevailing party." Landes Constr. Co., 833 F.2d at 1371 (citations omitted).

III. DISCUSSION

EA seeks judgment as a matter of law on Phase I, arguing that (1) the doctrine of judicial estoppel should bar Antonick from changing his position as to when he first became aware that he might have a claim against EA, and (2) no reasonable jury could have found that Antonick was not on notice of his claim before November 2005. In its motion for judgment as a matter of law on Phase II, EA contends that Antonick failed to prove (1) that there are substantial similarities in the expression in the source code plays and formations between Apple II Madden and Sega Madden, or (2) that any Sega Madden game is virtually identical to Apple II Madden when each game is considered as a whole.

In response to each motion, Antonick contends that EA waived many of its Rule 50(b) arguments by failing to raise them in its Rule 50(a) motion, and that, even if preserved, they should be rejected on the merits because substantial evidence supports the Phase I and Phase II verdicts.

The Court has considered Antonick's waiver arguments and concludes that EA adequately preserved the arguments this Order addresses.

Also before the Court is EA's motion to strike the segments of the trial testimony of Antonick's expert, Michael Barr. See dkt. 502. The Court has considered EA's motion to strike portions of Barr's testimony in which he opined about matters outside of his initial report and rebuttal report, and finds that the statements to which EA objects were sufficiently disclosed in his report or his deposition. The motion to strike is therefore DENIED.

A. EA Is Not Entitled to Judgment as a Matter of Law on Phase I

1. Judicial Estoppel Doctrine Does Not Apply

The issue raised by EA's judicial estoppel motion is whether Antonick impermissibly changed his position at trial regarding when he became aware of a possible claim against EA for purposes of the discovery rule such that (1) his trial testimony was clearly inconsistent with his prior position; (2) the prior position persuaded the Court to deny EA's motions to dismiss and for summary judgment; and (3) if not estopped from asserting the inconsistent position, Antonick would derive an unfair advantage and EA would face an unfair disadvantage. See Ah Quin v. Cnty. of Kauai Dep't of Transp., 733 F.3d 267, 270-71 (9th Cir. 2013).

Quoting the California Supreme Court, the Ninth Circuit explained that under the discovery rule:

[T]he plaintiff discovers the cause of action when he at least suspects a factual basis, as opposed to a legal theory, for its elements, even if he lacks knowledge thereof—when, simply put, he at least suspects that someone has done something wrong to him, wrong being used, not in any technical sense, but rather in accordance with its lay understanding. He has reason to discover the cause of action when he has reason at least to suspect a factual basis for its elements. He has reason to suspect when he has notice or information of circumstances to put a reasonable person on inquiry; he need not know the specific facts necessary to establish the cause of action; rather, he may seek to learn such facts through the process contemplated by pretrial discovery; but, within the applicable limitations period, he must indeed seek to learn the facts necessary to bring the cause of action in the first place—he cannot wait for them to find him and sit on his rights; he must go find them himself if he can and file suit if he does.

In his Complaint, Antonick alleged that

Only with the extensive publicity surrounding Electronic Arts' Madden NFL 20th Anniversary celebrations did Antonick become aware that Electronic Arts had continued to create derivative works from his work and considered its current software to have derived from Antonick's intellectual property. Specifically, in its publicity materials surrounding the 20th Anniversary[,] Electronic Arts, to Antonick's surprise[,] traced its current software back to his software[,] not the version developed by Park Place.
Compl. ¶ 92. Also in 2009, at the suggestion of an acquaintance, Antonick viewed a CNBC interview of Electronic Arts founder Trip Hawkins, in which Hawkins traced the Madden franchise back to Antonick's version by observing that the first version "took four years" to create. Id. ¶ 93. The interview
caused Antonick to do some additional research during which he came across the website of Park Place co-founder Troy Lyndon. Ex. 19, Response to Defendant's First Set of Interrogatories at 18. On the website, Lyndon credited Hilleman with helping him develop the 1990 Sega Genesis version of Madden and referred to the "countless hours" Hilleman spent assisting Jim Simmons to develop the game. Ex. 15, Troy Lyndon Website at 2. At this point, Plaintiff was stunned. He knew that Hilleman had also spent "countless hours" working with him on the 1989 version of Madden. Hilleman's significant involvement in the Park Place version contradicted EA's assurances that that version had been independently developed without any reference to Antonick's work. Ex. 19, Response to Defendant's First Set of Interrogatories at 18; see also Ex. 1, Antonick Dep. at 243-44.
All of these facts in the aggregate caused Antonick to suspect, for the first time in July 2009, that EA may not have been completely forthright with him about its past royalty obligations.
See 1st MSJ Opp'n (dkt. 125) at 9.

At trial, Antonick testified that the moment "that the big red flag . . . went off" that EA may have used his intellectual property was when he viewed the Troy Lyndon website. Tr. 276:6-8. Antonick pointed to three facts he learned for the first time from the website: (1) Lyndon stated that Park Place developed Sega Madden in less than six months, which Lyndon considered "record time;" (2) the programmer of Sega Madden, Jim Simmons, was a high school friend of Lyndon's, who had never programmed a professional game; and (3) producer Richard Hilleman spent "countless hours" with programmer Jim Simmons making the game more realistic. Tr. at 275:6-25. In particular, the statement that Hilleman spent "countless hours" with Simmons indicated to Antonick that the first Sega Madden was not developed in a clean room, amounting to a violation of his intellectual property rights. Tr. at 275:18-25-276:1-5. As for viewing the twentieth anniversary materials and the CNBC interview, Antonick testified at trial that these two events did not make him suspicious that he had a claim against EA because neither indicated that EA had used his intellectual property. Rather, they merely connected the later versions to his game, and, as Antonick acknowledged, he did not own the "John Madden" name. Tr. 274-19-25.

In support of its judicial estoppel motion, EA argues that Antonick's admission at trial that he learned nothing new from the anniversary materials or the CNBC interview is plainly inconsistent with his prior position that these two events spurred him to begin research that led to his discovery of his potential claim. In EA's view, Antonick changed his position to gain an advantage at trial: anticipating that EA would prove that neither the anniversary materials nor the CNBC interview provided Antonick with information in 2009 that he did not have before the statute of limitations ran, Antonick focused on his viewing of the Troy Lyndon website—rather than the anniversary materials or CNBC interview—as the "aha" moment that "knocked [him] off his feet." See Tr. 275.

The Court concludes that Antonick's trial testimony is not "clearly inconsistent" with his earlier position. See Ah Quin, 733 F.3d at 270. Prior to trial, Antonick represented that the anniversary materials and CNBC "interview seemed inconsistent with EA's earlier promises—and caused Antonick to do some additional research, during which he came across" Lyndon's website. See 1st MSJ Opp'n at 9. When viewing the information on Lyndon's website, Antonick "was stunned." Id. This version of events is not inconsistent with his trial testimony that it was not until viewing Lyndon's website that he believed the clean room concept was violated during Sega Madden's creation. Tr. 276:6-8.

While Antonick admitted at trial that he knew since 1990 that every game subsequent to Apple II Madden shared the Madden name, it was not until 2009 that he heard the acknowledgment by EA and Hawkins that the Madden series began with his game. That new fact caused Antonick to investigate further and led him to the Lyndon website. Accordingly, the Court does not find that Antonick's positions are inconsistent and will not grant EA's motion for judgment as a matter of law as to Phase I on the basis of judicial estoppel.

2. There Is Sufficient Evidence for a Reasonable Jury to Find Antonick's Claims Are Not Barred by the Statute of Limitations

In further support of its motion for judgment as a matter of law, EA contends that "[e]ven if one were to take Antonick's trial testimony at face-value, no reasonable jury could conclude that Antonick did not know about the relevant facts on [Troy Lyndon's] website before 2005," Phase I JMOL at 20, and "a reasonable person would have suspected the factual basis of his claim before November 2005," id. at 18. In EA's view, as soon as Antonick knew—no later than 1990—that all the games after his were called John Madden Football, that the first Sega Madden took less than a year to create while his took four, and that Richard Hilleman was the producer of both his game and the first Sega Madden, the statute of limitations began to run. Whether he learned additional information from the Lyndon website in 2009 is irrelevant if he was already on notice.

The Court cannot conclude that no reasonable juror viewing the testimony and evidence at trial could have found that prior to November 21, 2005, Antonick did not know, or know of facts that would cause a reasonable person to suspect, that he had a claim against EA. Even though reasonable minds may differ as to what a reasonable person would have suspected, there was sufficient evidence to support the jury's finding. See Johnson v. Paradise Valley Unified Sch. Dist., 251 F.3d 1222, 1227 (9th Cir. 2001) ("Substantial evidence is evidence adequate to support the jury's conclusion, even if it is also possible to draw a contrary conclusion from the same evidence.") (citation omitted); see also Lewis v. Musicians Union, AFM Local 6, C-90-3099 MHP, 1993 WL 356903, at *3 (N.D. Cal. Sept. 7, 1993) ("It is not up to the court to second-guess the jury on its conclusions. While reasonable minds may differ as to what constitutes 'reasonable' efforts, there was evidence to support the jury's decision.").

For these reasons, the Court denies EA's motion for judgment as a matter of law as to Phase I.

B. EA is Entitled to Judgment as a Matter of Law on Phase II

The Court now turns to EA's motion for judgment as matter of law as to Phase II. In this phase, the jury was asked two questions to determine whether (1) EA copied the expression of any protected element in any of the Sega Madden games at issue, and (2) if so, whether that copying made Sega Madden an infringing work of Apple II Madden.

To this end, Question 1 asked whether Antonick proved that there are substantial similarities between Sega Madden and Apple II Madden with respect to the two elements of Apple II Madden that the Court previously held were protectable: the expression in Antonick's source code of "field width" and of "plays and formations." See Order on 3d MSJ at 32. The jury found that for each version of Sega Madden at issue, Antonick proved that there are substantial similarities with respect to the expression in Antonick's source code of "plays and formations," but did not prove that there are substantial similarities with respect to the expression in Antonick's source code of "field width." See Phase II Verdict (dkt. 516) at 1.

The Sega Madden games at issue are: John Madden Football, John Madden Football '92, two versions of John Madden Football '93, John Madden Football '94, Madden NFL '95, and Madden NFL '96. See Phase II Verdict at 2. The Court refers to six games created after the first John Madden Football for Sega in 1990 as the "subsequent" Sega Madden games.

The jury's task did not end there, because while it found that EA copied some of Antonick's code, not all copying of protected expression is copyright infringement. See, e.g., Newton v. Diamond, 388 F.3d 1189, 1193 (9th Cir. 2004) ("The principle that trivial copying does not constitute actionable infringement has long been a part of copyright law."). "For an unauthorized use of a copyrighted work to be actionable, the use must be significant enough to constitute infringement." Id. at 93.

To determine whether the copying of the source code that expressed "plays and formations" amounts to infringement, Question 2 required the jury to compare "the works to determine whether, as a whole, they are sufficiently similar to support a finding of illicit copying." See Mattel, Inc. v. MGA Entm't, Inc., 616 F.3d 904, 913-14 (9th Cir. 2010); Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1442-43 (9th Cir. 1994). This second step, called the intrinsic test, "which examines an ordinary person's subjective impressions of the similarities between two works, is exclusively the province of the jury." Funky Films, Inc. v. Time Warner Entm't Co., L.P., 462 F.3d 1072, 1077 (9th Cir. 2006) (citation omitted).

The works must be compared as a whole because the relevant inquiry is "whether a substantial portion of the protectable material in the plaintiff's work was appropriated—not whether a substantial portion of defendant's work was derived from plaintiff's work." Newton, 388 F.3d at 1195. The sheer amount of protected information copied from plaintiff's work is not necessarily dispositive. Rather, courts are to focus on the significance of the protected expression to plaintiff's entire work. See InDyne, Inc. v. Abacus Tech. Corp., 876 F. Supp. 2d 1278, 1284 (M.D. Fla. 2012), aff'd, 513 F. App'x 858 (11th Cir. 2013) (explaining that the relevant inquiry for the second prong of the infringement inquiry is "whether the elements of [plaintiff's] software copied are protected expression and 'of such importance to the copied work that the appropriation is actionable.'"). If the copying of the protected expression is so extensive that it renders the plaintiff's work and the allegedly infringing work virtually identical, then the copying is actionable and the resulting work infringing.

Here, because the Court determined that Apple II Madden was entitled to thin—rather than broad—copyright protection, a work would be infringing only if an ordinary reasonable observer comparing Apple II Madden as a whole to Sega Madden as a whole would consider the works virtually identical. See Phase II Order at 3-4. The jury found all seven versions of Sega Madden at issue virtually identical to Apple II Madden, indicating its conclusion that all seven versions of Sega Madden at issue are infringing works, and therefore derivative works under the 1986 Contract for which Antonick is owed royalties.

In its order on EA's third motion for summary judgment, the Court used analytic dissection to determine if any of the allegedly similar elements of Apple II Madden and Sega Madden are protected by copyright and concluded that only two of ten elements were protectable. See Order on 3d MSJ at 15-32. The Court then filtered out the unprotectable elements to define the scope of copyright protection to afford Apple II Madden: broad or thin. Phase II Pretrial Order at 2. "Due to the narrow range of possible expression for a football video game and the fact that only two of the ten similar elements are protectable, the Court conclude[d] that Antonick's work is entitled to only thin protection." Id. at 3.

The Court, having reviewed the trial record, concludes that EA is entitled to judgment as to Question 2. Even construing the evidence in the light most favorable to Antonick, there is no legally sufficient basis for the jury's verdict that any of the Sega Madden games as a whole are virtually identical to Apple II Madden as a whole. The Court now addresses the evidentiary deficiencies as to Question 2.

1. The Jury Had No Evidence of Sega Madden as a Whole

In Question 2, the jury was asked to conduct the second prong of the infringement inquiry: whether Sega Madden as a whole is virtually identical to Apple II Madden as a whole. The Ninth Circuit has not defined "virtually identical," but some district courts have attempted to provide guidance as to the standard. See, e.g., Merch. Transaction Sys., Inc. v. Nelcela, Inc., No. CV02-1954-PHX- MHM, 2009 WL 723001, at *17 (D. Ariz. Mar. 18, 2009) (defining virtual identity as "differ[ing] from one another by no more than a trivial degree"); Berkla v. Corel Corp., 66 F. Supp. 2d 1129, 1143-44 (E.D. Cal. 1999) ("Clearly, the plain meaning of the words ["virtually identical"] will not allow any significant dissimilarities," but "will be satisfied at some point short of photographic reproduction.").

As the court in Berkla observed, "it is difficult to pin down a definition more precise than the narrow spectrum" between having no significant dissimilarities but being short of a photographic reproduction. Id. Indeed, because the intrinsic test calls for an ordinary observer's subjective comparison, virtual identity must be in the eye of the beholder. See id. (Whether two works are virtually identical is a "decision made on subjective observation."). For the same reason a court cannot strictly define what it means for works to be "virtually identical," a jury must be permitted to make its own subjective observation of the works.

However, in this case, the jury had no evidence of Apple II Madden or Sega Madden as a whole to enable it to make this subjective comparison. By failing to offer evidence of the games in their entirety, Antonick's proof on Question 2 was insufficient.

In response to EA's motion, Antonick does not point to any evidence of the works "as a whole." His opposition to EA's motion for judgment as a matter of law discusses each element in Apple II Madden, and recounts the testimony showing their similarity to their counterparts in Sega Madden. See Phase II JMOL Opp'n (dkt. 545) at 21-23. Similarly, in closing, counsel addressed the question posed to the jury as to whether or not Apple II Madden "as a whole is virtually identical to the Sega Madden game," and stated:

[T]hey both use a version of player ratings. They both use a system where you reach decision points and make a decision on what should happen in the game. They both have the same plays, the same sequencing, the same order; same plays pretty much to a T. They both have the same field size. Same formations. Same directional control system.
Tr. 2078:19-25-2079:1-4.

Comparing the works element by element is plainly counter to Ninth Circuit law. See Rodesh v. Disctronics, Inc., 8 F.3d 29, at *2 (9th Cir. 1993) ("Unlike the objective, extrinsic test, in the intrinsic test 'analytic dissection' is forbidden. The whole idea is to view the work as a whole, not to break it apart into pieces for individual scrutiny.") (internal citations omitted). If the jury's comparison of the works could be accomplished by merely comparing each of the elements head-to-head without considering the significance of the protected expression copied, "there would be no point to the extrinsic test, or to distinguishing ideas from expression" and a copyright holder could impermissibly rely on similarities in expression resulting from unprotectable elements. Apple Computer, 35 F.3d at 1446.

Antonick asserts that "a jury may appropriately undertake a 'qualitative' evaluation of similarities," rather than the quantity of copied material. See Phase II JMOL Opp'n at 21. He even quotes a book written by EA's expert for the proposition that "it is not 'the percentage of code that determines whether the [copying] is substantial' because 'code can be critical' to a program's operation but 'consist of a relatively small percentage of total lines of code.'" Id. at 21 n.24 (citing dkt. 325-13 at 331). The Court does not dispute the proposition that infringement can result from the significance of the expression copied rather than the sheer quantity. See, e.g., Newton, 388 F.3d at 1195 ("Substantiality is measured by considering the qualitative and quantitative significance of the copied portion in relation to the plaintiff's work as a whole.") (citing Worth v. Selchow & Righter Co., 827 F.2d 569, 570 n.1 (9th Cir. 1987)). However, this proposition does not substitute for sufficient evidence to allow the jury to place the protected expression—Antonick's source code for plays and formations—in the context of Apple II Madden as a whole.

See also, e.g., Greenspan v. Random House, Inc., 859 F. Supp. 2d 206, 218 (D. Mass. 2012), aff'd, No. 12-1594, 2012 WL 5188792 (1st Cir. Oct. 16, 2012) (concluding that "the five sentences that convey these ideas are quantitatively and qualitatively insubstantial in the context of [plaintiff's work] as a whole. Any copying claimed based on [the protectable elements] simply was not so extensive that an ordinary observer could conclude that the defendants unlawfully appropriated the plaintiff's original expressions.").

2. The Jury Had No Evidence that Any Subsequent Sega Madden Game was Virtually Identical to Apple II Madden

The record is similarly devoid of evidence supporting the jury's conclusion that the six Sega Madden games subsequent to 1990 are virtually identical to Apple II Madden. Antonick's evidence on the subsequent games relied on the proposition that if (a) the first Sega Madden is virtually identical to Apple II Madden and (b) all seven Sega Madden games were virtually identical to each other, then (c) all of the Sega Madden games were infringing works of Apple II Madden. Even assuming that this approach was valid, the record contains no legally sufficient basis for the jury to find that all of the Sega Madden games are virtually identical to each other.

No reasonable jury could have concluded that the seven Sega Madden games were virtually identical to each other. The only information about the subsequent games as a whole was Barr's opinion that each Sega Madden version changed "just a few things" each year—specifically by (1) changing player ratings, (2) adding plays, (3) adding features such as instant reply, (4) improving the look of the game's graphical characters—but were "essentially the same software program." Tr. 1310:19-25-1311:1-5. Without the opportunity to view each of the versions of Sega Madden, the jury had no basis for evaluating whether the changes Barr addressed altered each subsequent game such that they should not be considered one and the same for purposes of the intrinsic test.

Barr also testified that the source code of the subsequent games showed that they "had more plays" than 1990 Sega Madden, and specifically that the 1993 Sega Madden "actually added additional plays" from Apple II Madden. Tr. 1400:8-15.

In an apparent attempt to demonstrate that the later Sega Madden games were even more similar to Apple II Madden, counsel argued in closing that Barr "found no evidence that [EA] took plays out across the games he looked at," Tr. 2081:1-3. Even if it were relevant that EA did not delete plays in subsequent games, counsel's representation is belied by the record. When asked if he "ever [found] any evidence that plays were taken out of games," Barr testified that, he "didn't go specifically looking for game by game whether individual plays were taken out. . . . There were more plays in the later games." Tr. 1400:16-21.

Barr's opinion that all seven Sega Madden games are "essentially the same" as a whole cannot substitute for the jury's subjective comparison of each of the seven Sega Madden games as a whole to Apple II Madden as a whole. Because the intrinsic test requires the perspective of an ordinary, reasonable observer, Funky Films, Inc., 462 F.3d at 1077, expert testimony is not admissible evidence of similarity for purposes of the intrinsic test. See, e.g., Olson v. Nat'l Broad. Co., Inc., 855 F.2d 1446, 1448-49 (9th Cir. 1988) (stating that expert testimony is appropriate under the extrinsic test, but not under the intrinsic test); Express, LLC v. Fetish Grp., Inc., 424 F. Supp. 2d 1211, 1228 (C.D. Cal. 2006) ("While expert testimony is generally appropriate in conducting the extrinsic test, expert testimony may not be considered in conducting the intrinsic test.") (internal citation omitted); Trust Co. Bank v. Putman Publ'g Grp., Inc., No. CV 87 07393 AHS(JRX), 1988 WL 62755, at *6 (C.D. Cal. Jan. 4, 1988) ("Expert testimony is inadmissible on this intrinsic test.").

See also Computer Associates Int'l, Inc. v. Altai, Inc., 982 F.2d 693, 713 (2d Cir. 1992) ("[E]xpert testimony may be used to assist the fact finder in ascertaining whether the defendant had copied any part of the plaintiff's work. . . . However, once some amount of copying has been established, it remains solely for the trier of fact to determine whether the copying was 'illicit' . . . . Since the test for illicit copying is based upon the response of ordinary lay observers, expert testimony is thus 'irrelevant' and not permitted.") (citations omitted).

Just as Barr's expert opinion that the games are "essentially the same" does not relieve Antonick from providing evidence of each Sega Madden game as a whole, counsel's argument that all the Sega Madden games can be considered one and the same falls short providing evidence to allow the jury to compare the works as a whole. Counsel urged the jury to find that the Sega Madden games are virtually identical to each other if: "For all practical purposes, is it the same? . . . For the practical points that matter, is it the same thing?" Tr. 2079:7-9; see also Tr. 2079:16 (Sega Madden games "are all in the same family."). Certainly, at some level, all seven Sega Madden games could be considered "the same thing" for "practical purposes"—indeed, they are all football video games, all football games for the Sega Genesis, and all John Madden Football games in particular. Those similarities among them, however, are not evidence that they are virtually identical to each other, because copyright law protects only similarity in protectable expression, not similarities in unprotectable ideas. If counsel's approach were permitted, a copyright holder could rely on a jury's finding that one work is infringing in order to prevail as to any number of other works, which would eviscerate plaintiff's burden of proving infringement as to each work.

See, e.g., Mattel, Inc. v. MGA Entm't, Inc., 616 F.3d at 916-17 ("[W]hen works of art share an idea, they'll often be 'similar' in the layman's sense of the term. For example, the stuffed, cuddly dinosaurs at issue in Aliotti [v. R. Dakin & Co., 831 F.2d 898, 901 (9th Cir. 1987)], were similar in that they were all stuffed, cuddly dinosaurs-but that's not the sort of similarity we look for in copyright law. 'Substantial similarity' for copyright infringement requires a similarity of expression, not ideas. See id. The key question always is: Are the works substantially similar beyond the fact that they depict the same idea?").

Finally, Antonick's evidence as to the later games as a whole cannot rely on the stipulation that each Sega Madden game "used code from" the prior one. Tr. 1340:19-1341:22. In closing, counsel argued that the stipulation "show[s] that [EA] agrees that those games, year to year, shared the code that we talked about. They share code. The code drops down, it goes from game to game." Tr. 2080:14-17. But some shared code among Sega Madden games is not the issue. The stipulation offers no evidence about how much code was passed on to each subsequent game, or whether that code included Antonick's expression of source code as to plays and formations.

For these reasons, the Court concludes that the record contains no evidence from which a reasonable juror could conclude that Apple II Madden and any of the seven Sega Madden games are virtually identical when compared as a whole. Consequently, Antonick did not prove any of the Sega Madden games are infringing works, and EA is entitled to judgment as a matter of law.

C. In the Alternative, EA is Entitled to a New Trial of Phase II

In the alternative to its Motion for Judgment as a Matter of Law, EA seeks a new trial of Phase II. For the same reasons the Court grants EA's Judgment as a Matter of Law, see Section III.B supra, the Court concludes that the jury's verdict on Question 2 is against the clear weight of the evidence. Pursuant to Federal Rule of Civil Procedure 50(c)(1), the Court conditionally grants EA's motion for a new trial.

Under Rule 50(c)(1), "[i]If the court grants a renewed motion for judgment as a matter of law, it must also conditionally rule on any motion for a new trial by determining whether a new trial should be granted if the judgment is later vacated or reversed. The court must state the grounds for conditionally granting or denying the motion for a new trial."

In addition, the Court concludes that EA is entitled to a new trial based on the jury's finding on Question 1. Antonick's expert, Michael Barr, used flawed methods to support his opinion that EA copied Antonick's protectable expression. Specifically, Barr (1) used reversed-engineered binary data to produce and compare visual representation of the plays in Apple II Madden and Sega Madden, rather than comparing the source code itself, and (2) relied on similarities in unprotectable elements to suggest copying of protectable expression. As such, the jury's finding that there are substantial similarities between Sega Madden and Apple II Madden with respect to the expression in the source code of plays and formations is against the clear weight of the evidence.

Barr explained that for the purpose of his analysis, he had fragments of source code for the play editor, which included assembly language code for the formations, but not for the player instructions. Tr. 1375:5-9, 1431:22-24, 1432:5-7. For the player instructions, he had only "binary data." Tr. 1375:1-3; see also Tr. 1437:16-25-1438:8-9. The source code for the play editor "generated the binary data files that are used in the program." Tr. 1432:22-24. Because "comparing the source code and the binary requires a lot of human effort and programmer effort, Barr "undertook actually to automate the process to a certain extent, to generate a visual comparison of what the formations would look like and what the instructions to the players would look like." Tr. 1377:16-19.

First, Barr reverse-engineered the binary data for plays to perform his comparison. For the formations in the fragments of play editor source code available, he "extracted" "the binary [play] data file containing those formations." Tr. 1445:18-25. A programmer on Barr's staff wrote "software to read in the data from both games, the play data, and generate as best we could visual imagines comparing the two plays." Tr. 1377:24-25-1378:1-2. These "visual images" showed "where do each of the players appear relative to the line of scrimmage and what do they do after the formation?" Tr. 1378:2-3. Presenting the play diagrams for ten offensive plays and ten defensive plays side-by-side in Exhibit 476, Barr concluded that the plays were "effectively the same." Tr. 1383:19-14. Barr's analysis is deficient in several respects.

Reverse-engineering binary data does not necessarily produce the source code that created it. Barr admitted that "two experienced experts trying to approximate original source code through the process of decompiling binary code will produce different approximations of the original source code." Tr. 1436. In other words, "the process of decompiling binary code does not re-create the original source code." Tr. 1437; see also Syntek Semiconductor Co., Ltd. v. Microchip Tech. Inc., 307 F.3d 775, 779 (9th Cir. 2002) ("[S]ource code created by decompiling object code will not necessarily be identical to the source code that was compiled to create the object code.").

1. Barr's Comparison Of Visual Images Generated From Binary Data Does Not Provide Evidence Of Source Code Copying

Antonick's burden was to prove substantial similarities between the expression of the plays and formations in the Apple II Madden source code and the expression of the plays and formations Sega Madden source code. What Barr presented, however, was a visual representation of the code for the plays and formations in Apple II Madden and its counterpart play in Sega Madden—in his words, "the output for that program that [Barr and his staff] wrote to generate what the plays would have looked like from the underlying data in both games." Tr. 1378:23-25. By comparing the output, rather than the source code that creates it, Barr impermissibly compared the plays themselves, which, as this Court has already held and the law makes clear, are not protectable. See Order on 3d MSJ at 31, see also, e.g., 17 U.S.C. § 102(b) (2006) ("In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery."); Nimmer on Copyright, § 13.03A1 ("[C]opyright does not protect against the borrowing of abstract ideas contained in the copyrighted work. . . . if the only similarity between plaintiff's and defendant's works is that of the abstract idea, there is an absence of substantial similarity and hence, no infringement results."). Because the protected expression here was source code in Apple II Madden, a visual depiction of the result of that source code, versus its counterpart in Sega Madden, is not evidence of substantial similarity.

Permitting Antonick to rely on the visual depiction resulting from his source code to show illicit copying would be tantamount to granting him ownership of the particular play itself. Copyright law recognizes that the same function or implementation—here, a football play—can be carried out by different authors in different ways. The Copyright Act "confers ownership only over the specific way in which the author wrote out his version. Others are free to write their own implementation to accomplish the identical function, for, importantly, ideas, concepts and functions cannot be monopolized by copyright." Oracle Am., Inc. v. Google Inc., 872 F. Supp. 2d 974, 989, 997-98 (N.D. Cal. 2012) ("[T]here might be a myriad of ways in which a programmer may . . . express the idea embodied in a given subroutine."). In accordance with this basic principle, Antonick's ownership is limited to the expression of the plays and formations in his source code, not the plays and formations themselves.

Barr admitted the printed playbook that was packaged with Apple II Madden included twenty plays similar to the twenty he compared. Tr. 1446:7-11. His analysis may be evidence that EA created plays that were visually similar to those in the printed playbook, but EA was free to do so because the printed playbook was not protected by copyright. See Jury Instr. (dkt. 509) at 5.

For this reason, Barr's opinion, based on "generat[ing] a visual comparison of what the formations would look like and what the instructions to the players would look like," Tr. 1377: 17-19, is not evidence of substantial similarity in the relevant source code. Antonick offered no other source code comparison as to the expression of plays and formations.

Antonick responds that presenting side-by-side printouts of source code would have been meaningless to the jury without the assistance of expert testimony, particularly because the codes were in different programming languages. See Phase II JMOL Opp'n at 11. This contention is baseless, as Barr did just that to compare the source code of a play in Sega Madden 1995 with its counterpart play in Super Nintendo Madden 1995. Tr. 1398:16-25-1399:1-24. Despite the difference in assembly languages used in those two games, Barr was able to explain to the jury why he concluded that the expressions of the play in the source code were identical. Tr. 1398:16-25-1399:1-17. The problem is not that Barr offered expert testimony about similarity in the source code, but rather that he opined about the similarity of the codes' output.
--------

2. Similarities In Unprotectable Elements Are Not Evidence That Protectable Elements Were Copied

When asked for his "opinion as to the similarity between Mr. Antonick's play formations and Mr. Simmons's play formations," Barr testified that in his opinion, "the formations are for all practical purposes the same." Tr. 1383:2-3. As for "the similarity between Mr. Antonick's initial player movement and Mr. Simmons' initial player movement," Barr opined that most of the plays are "very similar or effectively the same, but not all the plays." Tr. 1383:13-14. Regarding the "overall similarity" between the play numbers, play formations, and initial movement between Antonick's game and Simmons's, Barr "consider[ed] these plays in the Sega game to be basically a subset of the plays in the Antonick game." Tr. 1383: 19-25:1384:1-2.

Barr further testified that "the Sega Genesis game relied on the source code from the Antonick game." Tr. 1384:11-12. Barr's "multiple bases" for that opinion were the similarities in: (1) field width; (2) x and y coordinates of the field; (3) misspelled terms representing the line of scrimmage; (4) subroutine names, label names, and strings; and (5) players' formations and initial movements. Tr. 1384:13-25-13851-8. Explaining why he believed that Antonick's "source code was leveraged in making the Sega Genesis game, and not the playbook," Barr stated that it was because the playbooks did not indicate that the field should be 80 yards or any wider than a regulation field, that the variable x should represent field width and y field length, that the game should have a 0 to 7 plus 8 directional system, or particular phrases in the code, such as "double fl," "double se," or "short zone." Tr. 1484:2023-1485:1-19.

In light of this testimony, the Court finds that Barr relied on similarities in unprotectable elements to opine about the substantial similarity of the protectable elements. This approach is plainly contrary to Ninth Circuit law, and not evidence that would properly support the jury's finding. It is well-settled that "the party claiming infringement may place no reliance upon any similarity in expression resulting from unprotectable elements." Apple Computer, Inc., 35 F.3d at 1446 (internal quotation marks omitted). Apple Computer, Inc., 35 F.3d at 1446. The very purpose of comparing protected elements is to determine whether the particular elements of each game are substantially similar. Using unprotectable elements to infer that protectable elements were copied is not proper evidence of whether the protected elements themselves are substantially similar. See Brocade Commc'ns Sys., Inc. v. A10 Networks, Inc., 873 F. Supp. 2d 1192, 1218 (N.D. Cal. 2012) ("In determining whether 'copying' has been shown in the context of computer software, 'which ordinarily contains both copyrighted and unprotected or functional elements,' a court determines 'whether the protectable elements, standing alone, are substantially similar.'") (quoting Sony Computer Entm't, Inc. v. Connectix Corp., 203 F.3d 596, 599 (9th Cir. 2000)) (emphasis added).

In defense of Barr's testimony, Antonick misconstrues the role of unprotectable elements in the jury's consideration of the works as a whole. The unprotectable elements must be identified and the works considered as a whole so that the trier of fact can determine whether the works are virtually identical as a whole as a result of defendant's copying or due to the similarities in unprotectable expression—not, as Antonick posits, to allow the trier of fact to infer copying of protectable expression. See Shaw v. Lindheim, 809 F. Supp. 1393, 1403 (C.D. Cal. 1992) (explaining that the Supreme Court made clear that a lack of similarity between original protected expression is fatal to a plaintiff's copyright claim; in other words, substantial similarities between unprotected elements, like those presented by plaintiff's expert witness at trial in this case, cannot be used to find substantial similarities in the protectible expression of two works.") (citing Feist Publ'ns, Inc. v. Rural Tel. Serv. Co., Inc., 499 U.S. 340, 379-80 (1991)). Nevertheless, in closing, Antonick's counsel urged the jury to make this inference. See Tr. 2073:12-17 ("You have to have those same [directional controls] if you're using Mr. Antonick's play data. You have to have those same numbers or the data won't work. So the fact that they have that directional control shows you, it's evidence of the fact that they are using the same play data because it's not a coincidence.").

The jury's finding of substantial similarity as to Antonick's plays and formations source code in the later versions of Sega Madden is also against the clear weight of the evidence. When asked for his "opinion about whether each 1990 to 1996 Sega Madden game derived from or was based on the Antonick Apple II game," Barr testified that "overall," the games "that we have contain a set of plays that includes a subset of the plays in Mr. Antonick's games." Tr. 1401:18-25-1402:1-3. In addition to relying on the similarity of unprotectable plays, Barr's opinion—essentially that Sega Madden's plays included some plays from Apple II Madden as well as others—is not evidence that the expression of the plays and formations in these games' source code is substantially similar to Antonick's.

For these reasons, the Court concludes that the jury's finding on Question 1 was against the clear weight of the evidence. Accordingly, EA is entitled to a new trial on the issue of whether there are substantial similarities between the expression of plays and formations in the Apple II Madden source code and the expression of plays and formations in Sega Madden.

D. Antonick's Non-Copyright Claims Dismissed Before Trial

Prior to Phase II, the Court granted EA's Third Motion in Limine, which sought to exclude evidence of ancillary contract breaches, see 6/26/13 Hr'g Tr. (dkt. 457) at 47, and held that Antonick failed to state a claim for fraud, see Phase II Pretrial Order (dkt. 460) at 5. The Court stated its intent to provide its reasoning for these rulings in a subsequent order, and does so now.

1. Antonick Did Not State Claims For Non-Copyright, Ancillary Contract Breaches

In the course of discovery, Antonick raised four additional breaches not alleged in his complaint, specifically, that EA failed to (1) put copyright notices on other Madden games that are Derivative Works under the Contract; (2) give Antonick a right of first refusal to develop other Madden games that would have been Derivative Works under the Contract; (3) maintain Antonick's confidential information (4) register Antonick's copyrights. See Pl.'s 3rd Am. Resp. to Rog. 1 (dkt. 294-11) at 5-6; Pl.'s. Resp. to Rog. 5 at 5-6. Prior to Phase II, the Court granted EA's motion in limine to exclude evidence of these so-called ancillary breaches, see 6/26/13 Hr'g Tr. at 47; see generally Mot. Limine 3 (dkt. 374), concluding that Antonick failed to state a claim for these ancillary breaches because he did not demonstrate separate damages. Instead, the sole damages named in his interrogatory responses and damage expert report were royalties for Derivative Works under Section V of the 1986 Contract.

Contrary to Antonick's position that nominal damages combined with disgorgement of EA's profits would serve as his damages for these ancillary breaches, the Court finds that royalties on EA's sales as provided in the 1986 Contract are the appropriate measure. Disgorgement of EA's profits on derivative works, is not only unnecessary to make Antonick whole, but it would have provided a windfall by awarding him more than he would have collected absent the breach. As for Antonick's "lost opportunity" claim for EA's alleged failure to give him the right of first refusal, awarding Antonick the profits EA earned on a game Antonick had no role in developing assumes that Antonick would have not only accepted EA's offer to develop the derivative work but also that his game would have generated the same profits as EA's version. This is highly speculative. Because the allegations are all variations on the claim that EA breached the 1986 Contract, and royalties are an appropriate remedy of a breach, there is no basis for awarding another form of damages.

2. Antonick Did Not State Claim For Fraud

Antonick's fraud claim does not survive independent of his breach of contract claim because he failed to allege that EA breached a duty independent of the contract or intentionally caused him harm beyond the contract breach. See Oracle Am., Inc. v. Serv. Key, LLC, No. C 12-00790 SBA, 2012 WL 6019580, at *9 n.5 (N.D. Cal. Dec. 3, 2012) ("Under California law, to maintain a fraud claim based on the same factual allegations as a breach of contract claim, a plaintiff must show that 'the duty that gives rise to tort liability is either completely independent of the contract or arises from conduct which is both intentional and intended to harm.'") (quoting Robinson Helicopter Co., Inc. v. Dana Corp., 34 Cal. 4th 989, 990 (2004)).

Instead, the allegedly fraudulent statements on which Antonick's claim relies were simply representations that EA was adhering to the contract and therefore cannot form the basis of a separate fraud claim. See JMP Sec. LLP v. Altair Nanotechnologies Inc., 880 F. Supp. 2d 1029, 1043 (N.D. Cal. 2012) (applying economic loss rule to dismiss fraud and negligent misrepresentation claims that "consist of nothing more than [defendant's] alleged failure to make good on its contractual promises."); Oracle USA, Inc. v. XL Global Servs., Inc., No. C 09-00537 MHP, 2009 WL 2084154, at *4 (N.D. Cal. July 13, 2009) ("[T]he fundamental rule in California is that no tort cause of action will lie where the breach of duty is nothing more than a violation of a promise which undermines the expectations of the parties to an agreement.").

Nor can Antonick's claim rely on the allegation that EA fraudulently concealed their wrongdoing by assuring him that Sega Madden was being developed without reference to his intellectual property. Because the jury found in Phase I that his claim is not barred by the statute of limitations, he cannot show reliance or damages arising from these statements.

IV. CONCLUSION

For the foregoing reasons, EA's Motion for Judgment as a Matter of Law re Phase I is DENIED. EA's Motion for Judgment as a Matter of Law re Phase II is GRANTED and its Motion for a New Trial is conditionally GRANTED. Antonick's Motion for Prejudgment Interest (dkt. 528) and Motion for Entry of Final Judgment as to his Fraud Claim (dkt. 542) are DENIED AS MOOT.

IT IS SO ORDERED.

________________

CHARLES R. BREYER

UNITED STATES DISTRICT JUDGE

Platt Elec. Supply, Inc. v. EOFF Elec., Inc., 522 F.3d 1049, 1057 (9th Cir. 2008) (quoting Norgart v. Upjohn, Co., 21 Cal. 4th 383, 397-98 (1999)).


Summaries of

Antonick v. Electronic Arts Inc.

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF CALIFORNIA
Jan 22, 2014
No. C 11-1543 CRB (N.D. Cal. Jan. 22, 2014)
Case details for

Antonick v. Electronic Arts Inc.

Case Details

Full title:ROBIN ANTONICK, Plaintiff, v. ELECTRONIC ARTS INC., Defendant.

Court:UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF CALIFORNIA

Date published: Jan 22, 2014

Citations

No. C 11-1543 CRB (N.D. Cal. Jan. 22, 2014)