Robin Antonick v. Electronic Arts, Inc.

[[COURTLISTENER_SUBOPINION {"id":"4101140","type":"010combined","part":"opinion","author":"Hurwitz","source_field":"html_with_citations"}]]
OPINION
HURWITZ, Circuit Judge:
In this case, the plaintiff claimed copyright infringement. But the contents of the copyrighted work and the allegedly infringing works were never introduced into evidence. The district court held that the claim failed as a matter of law. We agree, and affirm.
I. Background
Robin Antonick developed the computer code for the original
John Madden Football
game for the Apple II computer (“Apple II Madden”). Electronic Arts, Inc. (“EA”) released Apple II Madden in 1988. Apple II Madden, the first football video game with 11 players on each side, was an instant hit, the best seller of any sports video game of its time. Antonick subsequently programmed Madden games for the Commodore 64 and IBM-compatible computers (“IBM Madden”).
In 1989, Antonick began working for EA on Madden games for the Nintendo and Sega Genesis entertainment systems. But in August 1990, EA told him to stop— Nintendo was becoming obsolete, and EA had decided on a new direction for the Sega game, hiring Park Place Productions to create a version with “more of an arcade style.” In November 1990, EA released its first version of Sega Madden. In late 1991 or early 1992, EA released Antonick’s last Madden game, an update of IBM Madden.
Each year from 1992 to 1996, EA released Madden games for Sega Genesis and Super Nintendo (“Super Nintendo Madden”). The Madden games have remained incredibly lucrative, selling millions of copies and even attracting a loyal fan base among professional football players.
*1065
Antonick’s 1986 contract with EA defined “a custom computer software program known as John Madden Football” designed for the “Apple [II] Family of Computers” as the “Work,” and provided that Antonick would receive royalties on any “Derivative Work,” 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.” Antonick received no royalties for Sega Madden or Super Nintendo Madden, which EA assured him were not Derivative Works.
In 2011, Antonick brought this diversity action against EA, seeking contract damages in the form of unpaid royalties for Sega Madden and Super Nintendo Madden. The district court bifurcated the trial. In Phase I, the jury found that the statute of limitations did not bar Antonick’s claims. Phase II involved the merits of Antonick’s claims. Antonick produced evidence that Park Place was rushed and inadequately staffed, and argued that it copied his code to meét 'the demanding deadline for the first Sega Madden. Anton-ick’s expert, Michael Barr, opined that Sega Madden was substantially similar to certain elements of Apple II Madden. In particular, Barr opined that the games had similar formations, plays, play numberings, and player ratings; a similar, disproportionately wide, field; a similar eight-point directional system; and similar variable names, including variables that misspelled “scrimmage.”
1
But neither the source code for Apple II Madden—the “Work”—nor the source code of any allegedly infringing works were introduced into evidence. Nor were images of the games at issue introduced.
2
Nonetheless, the jury found that the Sega Madden games were Derivative Works under the 1986 contract. The district court then granted judgment as a matter- of. law (“JMOL”) to EA, holding that Antonick had not provided sufficient evidence of copyright infringement, because neither the source code used for Apple II Madden nor Sega Madden was in evidence.
II. Discussion
A. The Sega Claims
Although this is a contract case, because'royalties are available to Antonick under the 1986 contract only for a derivative work of Apple II Madden “within the meaning of the United States copyright law,” he had to prove copyright infringement to prevail on his contract claims. Antonick was thus required to prove that EA “copied protected elements of the work.”
Jada Toys, Inc. v. Mattel, Inc.,
518 F.3d 628, 636
(9th Cir. 2008) (citations omitted). “Absent direct evidence of copying, proof of infringement involves fact-based showings that the defendant had ‘access’ to the plaintiffs work and that the two works are ‘substantially similar.’ ”
Three Boys Music Corp. v. Bolton,
212 F.3d 477, 481
(9th Cir. 2000) (citation omitted).
“The Ninth Circuit employs a two-part test for determining whether one work is substantially similar to another.”
Benay v. Warner Bros. Entm’t, Inc.,
607 F.3d 620, 624
(9th Cir. 2010) (citation omitted).
[A plaintiff] must prove
both
substantial similarity under the “extrinsic test” and substantial similarity under the “intrin
*1066
sic test.” The “extrinsic test” is an objective comparison of specific expressive elements. The “intrinsic test” is a subjective comparison that focuses on whether the ordinary, reasonable audience would find the works substantially similar in the total concept and feel of the works.
Id.
(citations and question marks omitted). The district court granted JMOL to EA under the “intrinsic test” because “the jury had no evidence of Apple II Madden or Sega Madden as a whole to enable it to make this subjective comparison.”
•  The district court was correct. An-tonick’s claims rest on the contention that the source code of the Sega Madden games infringed on the source code for Apple II Madden. But, none of the source code was in evidence. The jury therefore could not compare the works to determine substantial similarity. See Seiler
v. Lucas-film, Ltd.,
808 F.2d 1316, 1319
(9th Cir. 1987) (“There can be no proof of ‘substantial similarity
5
and thus of, copyright infringement unless Seiler’s works are juxtaposed with Lucas’ and their contents compared.”) (applying the best evidence rule in a copyright action);
id.
(“[Pjroof of the infringement claim consists of the works alleged to be infringed.”);
accord Airframe Sys., Inc. v. L-3 Commc’ns Corp.,
668 F.3d 100, 107
(1st Cir. 2011) (“Having presented no evidence sufficient to prove the content of its registered source code versions, Airframe cannot show that any of its registered works is substantially similar to the allegedly infringing M3 program.”);
Gen. Universal Sys., Inc. v. Lee,
379 F.3d 131, 146
(5th Cir. 2004) (per curiam) (“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.”). And, absent evidence of the copyrighted work and the allegedly infringing works, the record is insufficient to allow appellate review of the jury’s verdict.
See, e.g., Olson v. Nat’l Broad. Co.,
855 F.2d 1446
, 1448, 1451 (9th Cir. 1988) (granting JMOL to copyright defendant because no reasonable jury could have found substantial,similarity);
cf. Shaw v. Lindheim,
919 F.2d 1353, 1355
(9th Cir. 1990) (“We have frequently affirmed summary judgment in favor of copyright defendants on the issue of substantial similarity.”).
Antonick argues there was no need to introduce the source code (or even the games at issue), because Park Place had access to the Apple II Madden code and a motive to copy it, and his expert and others testified to the similarity of the works. These arguments fail for three reasons.
First, the evidence at most demonstrates access and a possible motive to copy; it does not establish that -the “protected portions of the works are substantially similar.”
Jada Toys,
518 F.3d at 637
.
3
Access alone cannot establish copyright infringement.
Shaw,
919 F.2d at 1361
.
*1067
Second, our law is clear that expert testimony cannot satisfy a plaintiffs burden of proof under the intrinsic test, which “depend[s] on the response of the ordinary reasonable person.”
Brown Bag Software v. Symantec Corp.,
960 F.2d 1465, 1475
(9th Cir. 1992) (quoting
Sid & Marty Krofft Television Prods., Inc. v. McDonald’s Corp.,
562 F.2d 1157
, 1164 (9th Cir. 1977)).
4
Barr’s testimony may have been enough to establish substantial similarity under the extrinsic test, but it cannot satisfy Antonick’s burden of production under the intrinsic test.
Id.
Third, the lay testimony was about how the games appeared, not how they were coded—and Antonick does not assert a copyright interest in Apple II Madden’s audiovisual appearance, only in its coding.
Antonick argues that copying was shown by testimony of Michael Kawahara, an Apple II Madden assistant producer. When asked whether he recognized any of the plays in Sega Madden from Apple II Madden, Kawahara answered affirmatively, stating that “[it] was—well, since the interface was—well, it was the same as we used in the Apple II. It was very easy to look at all of the plays in the Genesis version and they looked identical ,.. to the original Apple II version.” This comment, however, does not establish that the source code for the two games were substantially similar. Kawahara had no programming responsibilities for Apple II Madden; did not understand the Apple II Madden code; did not see the Sega Madden code; and admitted that he had no knowledge about differences in the games’ codes.
Antonick also cites a statement by Richard Hilleman, an EA representative, that it was “possible” he had told an interviewer that “the Sega game took the system’s approach from Mr. Antonick’s game and just simply put a different aesthetic on top of it.” But, an “approach” is an idea that cannot be copyrighted—only its expression in code is protectable—and Sega Madden could have used Apple II Madden’s “approach” to football video games without violating the copyright laws.
See Bikram’s Yoga Coll. of India, L.P. v. Evolation Yoga, LLC,
803 F.3d 1032, 1037-38
(9th Cir. 2015) (“[R]ecognizing this vital distinction between ideas' and expression, courts have routinely held that the copyright for a work describing how to perform a process does not extend to the process itself.”).
Finally, Antonick argues that EA’s post-verdict Rule 50(b) motion for JMOL regarding, the intrinsic test should not have been considered because the pre-verdict Rule 50(a) motion argued only that the evidence was insufficient to show substantial similarity between the two elements of the code that the district court had ruled protectable, rather than discussing similarity of the works as a whole.
See EEOC v. Go Daddy Software, Inc.,
581 F.3d 951, 961
(9th Cir. 2009) (“Because it is a renewed motion, a proper post-verdict Rule 50(b) motion is limited to the grounds asserted in the pre-deliberation Rule 50(a)
*1068
motion.”)- But, both motions argued that the failure to place the source code in evidence was fatal to Antonick’s claim that EA had copied his work. That preserved the argument. ■
Id.
(“Rule 50(b) ‘may be satisfied by an ambiguous or inartfully made motion’ under Rule 50(a).”) (quoting
Reeves v. Teuscher,
881 F.2d 1495, 1498
(9th Cir. 1989)).
B. The Super Nintendo Claims
Antonick sought royalties for the Super Nintendo Madden games under Amendment 1 to his contract, which provided for royalties for derivative works for platforms in the “Same Microprocessor Family”, as the Apple II. The Amendment defined “Microprocessor Family” as “a single microprocessor and all related microprocessors that utilize the same instruction set and have the same instruction and data word size.”
Antonick’s expert, Garry Kitchen, testified that the Apple II used the 6502 processor, which had an 8-bit data word size and 56 instructions of up to 3 bytes in length. The Super Nintendo used the more advanced 5A22 processor, which in its native mode used a 16-bit data word size and at least 92 instructions of up to 4 bytes in length; but which could also act as a 6502 processor for purposes of backwards compatibility. Kitchen testified that this backwards compatibility meant that the microprocessors were in the same “family,” as the industry uses the term. He acknowledged that Super Nintendo Madden was designed to use that system’s advanced capabilities, utilizing a larger instruction set, longer data word sizes, and longer instructions than Apple II Madden.
After Kitchen’s testimony, the district court dismissed the Super Nintendo claims, holding that the Super Nintendo was not in the same Microprocessor Family as the Apple II under the contractual definition, because it used a larger instruction set, instruction size, and data size. Antonick argues that the two processors were in the same family as a practical matter because they
could
use the same instruction set, instruction size, and data size. Antonick’s factual premise may well be correct. But we deal here with contract interpretation, and the word “could” is not in the contractual definition.. Instead, the contract requires that, to be in the same family, two processors must “utilize the
same
instruction set and have the
same
instruction and data word size.” The Apple II and Super Nintendo processors have different instruction sizes and data word sizes.
5
The district court therefore did not err in dismissing the Super Nintendo derivative work claims.
Antonick argues in the alternative that EA breached the contract by failing to give him the opportunity to develop the Super Nintendo Madden game. The contract provided that, if Antonick developed any Derivative Works for new microprocessor families, he would be entitled “to written notice and the opportunity to develop additional Derivative Works for the New Microprocessor Family.” Antonick developed a Madden game for the Apple II GS, which he argues, and EA does not appear to deny, is in the same microprocessor family as the Super Nintendo. As damages, he seeks royalties based on the actual Super Nintendo game sales.
The district court dismissed this claim because a fact-finder would have to speculate on whether Antonick could have developed such a work, how well an Antonick-developed Super Nintendo game would have sold, and what royalty rate the parties would have agreed upon; Antonick’s
*1069
expert report did not address these issues. The district court was correct; the jury could not have determined Antonick’s damages from the alleged breach to a “reasonable certainty.”
Sargon Enters., Inc. v. Univ. of S. Cal.,
55 Cal.4th 747
,
149 Cal.Rptr.3d 614
,
288 P.3d 1237, 1254
(2012). Moreover, even if the district court erred, there was no harm, because'Anton-ick’s failure to introduce any source code precluded a finding that Super Nintendo Madden was a Derivative Work.
C. The Development Aids
The contract also gave EA a license to create derivative works using certain tools designed by Antonick (“Development Aids”), and provided that the parties would “negotiate in good faith” for further licenses if EA wanted to use the aids to create non-derivative works. Antonick alleges that EA used the aids to create non-derivative works without seeking that license.
The district court dismissed this claim because Antonick offered no evidence of purported damages. Antonick did not show the value of similar licenses or the benefit that EA received from using the Development Aids. Instead, Antonick cited only the report of his damages expert, which simply made generic royalty calculations based on existing sales without explaining how those calculations were relevant to the Development Aid claim. The district court correctly kept this unsubstantiated claim from the jury.
See Amelco Elec. v. City of Thousand Oaks,
27 Cal.4th 228
,
115 Cal.Rptr.2d 900
,
38 P.3d 1120, 1130, 1132
(2002).
CONCLUSION
The judgment of the district court is AFFIRMED.
2
. EA showed the jury a video of Sega Madden, but the jury did not view a video of Apple II Madden.
3
. If the range of possible expression is narrow, the copyrighted work is entitled to thin protection, and a plaintiff must show virtual identity between the copyrighted work and the allegedly infringing work.
See Mattel, Inc.
v.
MGA Entm't, Inc.,
616 F.3d 904
, 913-14 (9th Cir. 2010), If the range of possible expression is wide, the work is entitled to broad protection and the plaintiff must show only substantial similarity.
Id.
The district court held that "[d]ue 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,” Antonick’s work was entitled only to thin .protection, requiring him to show virtual identity of the works as a whole. Antonick disputes that conclusion, arguing that he needed to show only substantial similarity. We need not resolve that dispute, because we conclude that Antonick, having presented insufficient evidence of the works as a whole, loses under either standard.
4
. Antonick’s contrary cases are from- other circuits; of the two Ninth Circuit cases he cites, one is a memorandum disposition holding expert testimony admissible in a case in which the relevant works themselves were in evidence,
Lucky Break Wishbone Corp. v. Sears Roebuck & Co., 373
Fed.Appx. 752, 755-56 (9th Cir, 2010), and the other concerned a special master’s report, not expert testimony,
Johnson Controls, Inc. v. Phoenix Control Sys., Inc.,
886 F.2d 1173
, 1176 (9th Cir. 1989).
Antonick is not alone in contending that experts should be allowed to help juries assess the holistic similarity of technical works such as computer programs,
See Brown Bag,
960 F.2d at 1478
(Sneed, J., concurring);
Comput. Assocs. Int’l, Inc. v. Altai, Inc.,
982 F.2d 693, 713
(2d Cir. 1992). But, given our precedents, that argument must be addressed to an en banc court.
5
. Arguably, the chips "utilize the same instruction set” because the 6502 uses a subset of the 5A22's instructions.