On December 27 Judge Alsup set forth a questionable alternative approach to the determination of patent damages in the case, particularly in light of the prior Sun/Google negotiations in 2006. (See 657 [PDF; Text]) We immediately questioned the approach the judge was suggesting and suggested that it would not be received well by Google.

The parties have now filed the responses requested by Judge Alsup, and as we had anticipated, Google is vehement in its objection to this alternative approach. Not surprisingly, Oracle's response is "well, it's okay, but it could be tweaked even more in our favor."

Jump To Comments

In analyzing the parties’ submissions on reasonable royalty issues, the judge would like to have the benefit of counsel’s guidance on the extent to which the following approach would be proper under appellate law. Significantly, this approach would not place the burden on Oracle to allocate among the items in the $100 million offer by Sun in 2006. Here are the steps in the possible line of reasoning:

1. Through its econometric and conjoint statistical analyses, Oracle claims to be able to spread the 2011 value of Android across various features. This methodology, standing alone, has not been challenged by Google. These analyses allow Oracle to opine on a value for a particular feature (such as processing time) in the 2008–2011 marketplace and, in turn, opine on the value of that feature to Android (meaning to Google) in 2008–2011. This, of course, is a stand-alone value independent of the 2006 offer.

2. The 2008–2011 value is today an arguable indicator of the value Google would have placed on that feature in 2006. True, it is not a precise value because it derives from data as yet unknown in 2006, but it might be a rough indicator of how vital that feature was expected to turn out so long as the marketplace events, as they eventually unfolded, were reasonably predictable in 2006. For example, if in 2011 we now know that feature X has proven to be extremely important to consumers, then the argument would be that the parties in 2006 could reasonably have expected that feature to turn out approximately this way. This expectation then, in turn, would have informed their 2006 negotiation. Again, so far, this line of reasoning has nothing to do with the $100 million offer.

3. The value of a feature must be apportioned among all of the know-how inputs that enabled it. That a license to a particular patent claim in suit is now needed to practice a feature does not justify attributing the entire value of the feature to that patent claim, for other know-how may also be required to practice the feature, such as licenses from other competitors and Google’s own independent know-how contribution to developing that feature. Again, the fact that a license to practice an Oracle patent claim is essential to the feature does not justify appropriating the full market value of the feature to that claim. This is a question of apportionment but it differs from the issue of apportionment of the 2006 package.

2

4. Thus, in the 2006 hypothetical negotiation, both sides would be informed, the argument would go, as to the expected importance of a feature but would also be informed as to the relative contribution of the claimed invention in comparison to all other know-how needed to enable the particular feature. Only the expected percentage contribution of the claimed invention to the overall expected value would be on the negotiating table.

5. Finally, under this line of analysis, the relevance of the $100 million offer in 2006 would be defensive, meaning Google would be entitled to rebut by showing that Sun would not have extracted the vast sums now suggested by Dr. Cockburn because Sun was willing to license, not just the claimed inventions, but the entire package for $100 million. If Google wishes to argue further that only a small percentage of the $100 million should be attributed to the claimed inventions, then it would be Google’s burden to allocate the $100 million between the claimed inventions versus the rest of the 2006 package, subject to adjustments for fragmentation and so on.

Under this line of reasoning, to repeat, there would be no need for Oracle, in order to meet its initial burden, to allocate the $100 million among the thousands of items in the 2006 package (or even to address the 2006 offer). Oracle would, on the other hand, have to allocate the value of a feature as between the claimed invention and all other know-how contributing to that feature.

Oracle has already measured the contribution of the patents and copyrights in suit to specific attributes of Android, such as improved application launch speed, enlarged memory, and an increased number of available applications. Oracle has also measured the incremental effect on mobile advertising revenue as a result of those specific enhancements. Using these ex post measurements to inform the ex ante hypothetical negotiation is supported by substantial appellate law under the circumstances presented here.

The one step to which Oracle objects is Step 3, the apportionment of contribution among inputs. Oracle claims Prof. Cockburn has already isolated the contribution of the patent and copyrights, a point that Google has continued to challenge. Oracle further challenges this step on its legal merits. Oracle goes so far as to make the bold statement:

Accordingly, the value of each feature is entirely the value of the patents and copyrights. No further apportionment of the features at issue to “other know-how contributions” is necessary or appropriate. [emphasis added]

I also can't pass up the following paragraph from the Oracle response:

Requiring allocation of the relative contributions of all the various “know-how inputs” that might be viewed as enabling a particular feature would also be unworkable. In 2007, Congress specifically considered an amendment to the patent laws that would have required essentially what the Alternate Approach proposes here, namely that, “[t]he court shall exclude from the analysis the economic value properly attributable to the prior art, and other features or improvements, whether or not themselves patented, that contribute economic value to the infringing product or process.” Patent Reform Acts of 2007, 110 H.R. 1908, section 5(a)(2); 110 S. 1145, section 5(a)(2). In response, then- Chief Judge Michel of the Federal Circuit wrote to Senators Leahy and Hatch that he was “unaware of any convincing demonstration of the need” for such a requirement, and that it was “a massive undertaking for which courts are ill-equipped” which would cause courts to “be inundated with massive amounts of data, requiring extra weeks of trial in nearly every case.” (Letter from Chief Judge Paul Michel, May 3, 2007, available at http://frwebgate.access.gpo.gov/cgibin/ getdoc.cgi?dbname=110_senate_hearings&docid=f:37760.pdf, at p. 277–78.) The amendment did not pass the Senate.

Of course, Oracle loves the idea of being able to "peek ahead" to Android's success. This is the most controversial aspect of the judge's proposal, and as we will see, it is an approach that Google believes flies in the face of settled law. In fairness, Oracle appears to agree with Google's position that such post-time-of-initial-infringement evidence can only be considered to measure whether a royalty is reasonable, not what the royalty should be.

How does Google feel about this alternative approach? (681 [PDF; Text]) Have a barf bag handy.

The short answer is Google has found no appellate case law supporting this approach. In particular, there is no basis in appellate law for using the actual market value of the claimed inventions in 2008-11 as a direct proxy for how reasonable parties would have valued those inventions in 2006. In fact, the Federal Circuit has made clear that, where there is good evidence of what the parties anticipated at the time of the hypothetical negotiation, it is of no moment whether those expectations were actually realized. To the extent the Federal Circuit has allowed hypothetical negotiators to “peek into the future,” it has been as a check on the reasonableness of conclusions reached through other means, not to arrive at those conclusions in the first instance. In this case in particular it would make no sense to calculate the present values of features as a proxy for their values in 2006. Unlike in many infringement cases, Google and Sun actually engaged in negotiations around the time of alleged first infringement for a technology partnership that would have included an intellectual property package containing all the asserted patents and copyrights. Although the parties did not fix a price for any of the individual patent claims or copyrighted material now at issue, Sun must have understood all the components of the package it was offering—its patent portfolio, copyrights, the JAVA trademark, other know-how, and cash payments to replace potential lost revenue—and the general value of those components. As Sun’s successor, there is no reason why, in preparing its damages expert reports, Oracle could not have apportioned that package to account for each of its components.

...any upper limit set during the parties’ actual negotiations around the time of first infringement—when the success of Android was uncertain—would be significantly lower than the present value of the entire Android platform, which would be the starting point under the approach contemplated in the Order.

More important than Google's horror at this approach is the fact, uncontroverted by Oracle, that:

It would also violate federal law by substituting Google’s profits for Oracle’s damages—an approach disavowed by Congress in 1946, when it amended the patent statute specifically to eliminate an infringer’s profits as a measure of damages, and further condemned by Georgia-Pacific, the grandfather of reasonable-royalty case law.

Google also debunks the "peek forward" cases cited by Oracle, except to the extent that they simply permit such post-negotiation information to be used "to support the reasonableness of a royalty calculation based on facts known at the time [of initial infringement]." The important point is that, yes, post-negotiation information can be considered as one of the fifteen Georgia-Pacific factors to determine a reasonable royalty, but it cannot be the sole factor or even the primary factor. Such post-negotiation information can be used to test for reasonableness, not as a basis for determining the royalty in the first place.

Google also highlights the fact that this approach would allow Oracle to substitute Google's profits for Oracle's actual damages, an approach specifically rejected by both case law and Congress. (See Section B of the Google response.)

Finally, Google points out that apportionment, which Oracle claims to have done but which Google continues to question, is only problematic when one tries to make the determination based on the infringer's profits, which is not permitted by law, versus the value to the patent holder, a further reason for not using the alternative approach.

One can only hope that Judge Alsup, now having been informed by the arguments of the parties, recognizes that the alternative approach is not only wrong, it is unworkable.

**********

Docket

681 - Filed and Effective: 01/05/2012
Google Response to Trial Plan
Document Text: Response re 657 Order GOOGLE'S RESPONSE TO THE COURT'S DECEMBER 27, 2011 REQUEST FOR FURTHER BRIEFING ON DAMAGES EXPERT ISSUES by Google Inc.. (Van Nest, Robert)

682 - Filed and Effective: 01/05/2012
Oracle Response to Trial Plan
Document Text: RESPONSE to ORACLE AMERICA, INC.S RESPONSE TO COURTS DECEMBER 27, 2011, REQUEST FOR FURTHER BREIFING ON DAMAGES (DKT. NO. 657) by Oracle America, Inc. (Holtzman, Steven)

**************

Documents

681

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION

ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.

Case No. 3:10-cv-03561-WHA

GOOGLE’S RESPONSE TO THE
COURT’S DECEMBER 27, 2011
REQUEST FOR FURTHER BRIEFING
ON DAMAGES EXPERT ISSUES

Dept. Courtroom 8, 19th Floor
Judge: Hon. William Alsup

TABLE OF CONTENTS

Page

I. INTRODUCTION ...... 1

II. ARGUMENT ...... 3

A. Under clearly established Federal Circuit case law, a reasonable
royalty must be based on the parties’ reasonable expectations at the
time of alleged first infringement, not on subsequently occurring
market facts after infringement ...... 3

B. The approach in the Order is incompatible with federal statutory and
case law ...... 6

C. Because it was uncertain in 2006 whether Android would succeed or
fail, the actual market value of Android in 2008-11 is not a reasonable
proxy for the expected market value of Android in 2006 ...... 7

D. The approach in the Order would not simplify Oracle’s apportionment task ...... 8

E. Using the approach in the Order would lead Oracle to further inflate its
damages ...... 9

F. Other comments on the Order ...... 10

i

TABLE OF AUTHORITIES

Page(s)

Federal Cases

Fromson v. Western Litho Plate and Supply Co.
853 F.2d 1568 (Fed. Cir. 1988) ...... 5

Georgia-Pacific Corp. v. United States Plywood Corp.
243 F. Supp. 500 (S.D.N.Y. 1965) ...... passim

Hanson v. Alpine Valley Ski Area, Inc.
718 F.2d 1075 (Fed. Cir. 1983) ...... 3, 4

Integra Lifesciences I, Ltd. v. Merck KGaA
331 F.3d 860 (Fed. Cir. 2003) ...... 3

Interactive Pictures Corp. v. Infinite Pictures, Inc.
274 F.3d 1371 (Fed. Cir. 2001) ...... 3, 4, 6, 7

Lucent Technologies v. Gateway, Inc.
580 F.3d 1301 (Fed. Cir. 2009) ...... 3, 5

Radio Steel & Mfg. Co. v. MTD Prods. Inc.
788 F.2d 1554 (Fed. Cir. 1986) ...... 4, 6

Riles v. Shell Exploration & Prod. Co.
298 F.3d 1302 (Fed. Cir. 2002) ...... 3

Stickle v. Heublein, Inc.
716 F.2d 1550 (Fed. Cir. 1983) ...... 7

Trans-World Mfg. Corp. v. Al Nyman & Sons, Inc.
750 F.2d 1552 (Fed. Cir. 1984) ...... 5, 7

Federal Statutes

35 U.S.C. § 284 ...... 6

ii

Congress did not intend to aid a patentee in solving his problem of
proving the quantum of his damages by enabling him to substitute the quantum
of the infringer’s profits … the two concepts are basically different; they cannot
be treated as equivalent.

— Georgia-Pacific Corp. v. United States Plywood Corp., 243 F. Supp.
500, 519 (S.D.N.Y. 1965).

I. INTRODUCTION

In a December 27, 2011 order (“Order”), the Court asked the parties to submit briefing on a possible approach to calculating a reasonable royalty. That approach would involve:

1. Attempting to calculate the value of certain features of Android (e.g., processing speed) in the 2008-11 market, rather than at the time of alleged first infringement in 2006;
2. Using the 2008-11 value of that feature as a stand-in for the importance of that feature to Google at the time of alleged first infringement in 2006, “so long as the marketplace events, as they eventually unfolded, were reasonably predictable in 2006”;
3. Apportioning the 2008-11 value of that feature among all the know-how enabling that feature, including the value of the patent claims and copyrights in suit, in order to calculate the value of the patent claims and copyrights in the 2008-11 market;
4. Using these 2008-11 values of the components of the relevant features to set the parameters of the 2006 hypothetical negotiation; and
5. Limiting the use of the parties’ actual 2006 negotiations to Google’s defense case, where Google would have the burden of allocating the value of Sun’s 2006 demands to Google among the claimed inventions and the other components of Sun’s IP portfolio.

Order at 2-3.

After setting forth this possible approach, the Court asked the parties to address the issues whether (1) appellate law allows patent damages to be calculated using this sort of forwardlooking approach; (2) Oracle would have the burden to apportion the values of given features between the claimed inventions and other contributions; and (3) Oracle would have no obligation to allocate Sun’s actual 2006 demands to Google between the claimed inventions and the other components of Sun’s IP portfolio, although Google would be free to do so defensively.

The short answer is Google has found no appellate case law supporting this approach. In particular, there is no basis in appellate law for using the actual market value of the claimed inventions in 2008-11 as a direct proxy for how reasonable parties would have valued those

1

inventions in 2006. In fact, the Federal Circuit has made clear that, where there is good evidence of what the parties anticipated at the time of the hypothetical negotiation, it is of no moment whether those expectations were actually realized. To the extent the Federal Circuit has allowed hypothetical negotiators to “peek into the future,” it has been as a check on the reasonableness of conclusions reached through other means, not to arrive at those conclusions in the first instance.

In this case in particular it would make no sense to calculate the present values of features as a proxy for their values in 2006. Unlike in many infringement cases, Google and Sun actually engaged in negotiations around the time of alleged first infringement for a technology partnership that would have included an intellectual property package containing all the asserted patents and copyrights. Although the parties did not fix a price for any of the individual patent claims or copyrighted material now at issue, Sun must have understood all the components of the package it was offering—its patent portfolio, copyrights, the JAVA trademark, other know-how, and cash payments to replace potential lost revenue—and the general value of those components. As Sun’s successor, there is no reason why, in preparing its damages expert reports, Oracle could not have apportioned that package to account for each of its components.

Equally important, the parties’ real-world negotiations placed a defined upper limit on the aggregate value of the complete Sun package. One possible upper limit, which Oracle’s expert used in his most recent report, was Sun’s $100 million demand to Google in March 2006. Another possible upper limit, as Google has urged, was the subsequent $28 million demand Sun later made in June 2006 for the same package. But regardless of which figure is used, any upper limit set during the parties’ actual negotiations around the time of first infringement—when the success of Android was uncertain—would be significantly lower than the present value of the entire Android platform, which would be the starting point under the approach contemplated in the Order. Calculating Oracle’s damages as a percentage of the current value of Android would almost certainly have the effect of inflating damages even beyond the “vast sums,” Order at 3, Oracle put forth in its previous damages reports. It would also violate federal law by substituting Google’s profits for Oracle’s damages—an approach disavowed by Congress in 1946, when it amended the patent statute specifically to eliminate an infringer’s profits as a measure of

2

damages, and further condemned by Georgia-Pacific, the grandfather of reasonable-royalty case law. See 243 F. Supp. at 519, 521-22, 525. And it would not make the experts’ apportionment task any easier, since the myriad types of know-how that enable the relevant features of Android were not a subject of discovery in this case.

The Court’s initial instinct in the July 22, 2011 order was right. The reasonable royalty analysis should be based on the value of the patents and copyrights in suit at the time of the hypothetical negotiation in 2006, and the correct way, both logically and legally, to attack that problem is by taking a close look at the parties’ actual 2006 negotiations.

II. ARGUMENT

A. Under clearly established Federal Circuit case law, a reasonable royalty must be
based on the parties’ reasonable expectations at the time of alleged first
infringement, not on subsequently occurring market facts after infringement.

For the past decade, the Federal Circuit has consistently ruled that a reasonable royalty in a patent case must be determined based on the parties’ “sales expectations at the time when infringement begins … as opposed to an after-the-fact counting of actual sales.” Interactive Pictures Corp. v. Infinite Pictures, Inc., 274 F.3d 1371, 1373 (Fed. Cir. 2001) (emphasis added).¹ Based on this rule, Oracle cannot use the present-day value of Android features—which, after all, can be measured only by calculations akin to “an after-the-fact counting of actual sales”—as a stand-in for the value the parties would have expected from those features in mid-2006.

The Federal Circuit’s Interactive Pictures decision explains that expectations—not after the fact actual numbers—are what form the basis for a hypothetical negotiation. There, the plaintiff’s damages expert relied on an annual sales projection prepared by the defendant two months before infringement began. Id. at 1384. The defendant disavowed its own projection as

__________________________________

¹ See also, e.g., Lucent Technologies v. Gateway, Inc., 580 F.3d 1301 (2009) (“The hypothetical negotiation tries, as best as possible, to recreate the ex ante licensing negotiation scenario and to describe the resulting agreement.”); Integra Lifesciences I, Ltd. v. Merck KGaA, 331 F.3d 860, 869 (Fed. Cir. 2003) (reversing jury verdict for lack of substantial evidence where royalty determination set at time following first infringement); Riles v. Shell Exploration & Prod. Co., 298 F.3d 1302, 1311 (Fed. Cir. 2002) (determination of a reasonable royalty “contemplates a hypothetical negotiation between the patentee and the infringer at a time before the infringement began.”); Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075, 1079 (Fed. Cir. 1983) (the “key element in setting a reasonable royalty … is the necessity for return to the date when the infringement began”).

3

“speculative,” pointing out that its actual sales had fallen short of expectations. Id. The Federal Circuit disagreed, rejecting the use of hindsight:

[We have] endorsed the conceptual framework of a hypothetical negotiation between the patentee and the infringer as a means for determining a reasonable royalty. When that framework is employed, the negotiation must be hypothesized as of the time infringement began. . . . In this case, the 1996 business plan and its projections for future sales were prepared by [defendant] two months before infringement began. Thus, rather than being outdated for purposes of the hypothetical negotiation, those projections would have been available to [defendant] at the time of the hypothetical negotiation. The fact that [defendant] did not subsequently meet those projections is irrelevant to [defendant’s] state of mind at the time of the hypothetical negotiation. Nor does [defendant’s] subsequent failure to meet its projections imply that they were grossly excessive or based only on speculation and guesswork. Instead, [defendant’s] subsequent failure to meet its projections may simply illustrate the element of approximation and uncertainty inherent in future projections.

Id. at 1384-85 (emphases added). Interactive Pictures made clear that the controlling evidence for purposes of setting a royalty rate must be the parties’ reasonable expectations at the time of first infringement. It explicitly rejected the notion that prior Federal Circuit case law

require[d] that estimates of sales revenues, as referenced in a hypothetical negotiation at the time infringement began, must later bear a close relation to actual sales revenue. Such a proposition would essentially eviscerate the rule that recognizes sales expectations at the time when infringement begins as a basis for a royalty base as opposed to an after-the-fact counting of actual sales.

Id. (emphasis added).

This is consistent with the Federal Circuit’s treatment of this question since the court’s formation. In Hanson, its first case to consider the issue, the defendant appealed a verdict awarding a very large royalty. See 718 F.2d 1075, 1077 (Fed. Cir. 1983). The defendant argued that the royalty was too high to be reasonable, because it would make it impossible for the defendant to turn a profit. The Federal Circuit rejected that argument, explaining that the key issue was what the parties would have agreed to at the time of first infringement, regardless of whether that royalty was one the defendant would have agreed to after the fact. A reasonable royalty should “be determined not on the basis of a hindsight evaluation of what actually happened, but on the basis of what the parties to the hypothetical license negotiations would have considered at the time of the negotiation.” Id. at 1081 (emphasis added); see also Radio Steel & Mfg. Co. v. MTD Prods. Inc., 788 F.2d 1554, 1557 (Fed. Cir. 1986).

4

Oracle may point to a series of Federal Circuit cases allowing hypothetical negotiators to “peek into the future.” But these cases simply affirm the uncontroversial Georgia-Pacific principle that post-negotiation information can be used to support the reasonableness of a royalty calculation based on facts known at the time. After all, one of the 15 non-exclusive factors in the Georgia-Pacific analysis is “[t]he established profitability of the product made under the patent; its commercial success; and its current popularity.” 318 F. Supp. at 1120.

For example, in Trans-World Mfg. Corp. v. Al Nyman & Sons, Inc., 750 F.2d 1552 (Fed. Cir. 1984) the Federal Circuit reversed because the district court had refused to allow plaintiff to introduce any evidence of defendant’s actual profits as part of its reasonable royalty case. See 750 F.2d at 1566. It cited factor 8 of the Georgia-Pacific test, noting that “[e]vidence of the infringer’s actual profits generally is admissible as probative of his anticipated profits.” Id. at 1568. But it also “express[ed] no opinion concerning the weight, if any, to be given such evidence or any conditions that might properly be imposed upon its admission; we indicate only that we do not think the district court should have excluded it.” Id. Likewise, though Fromson v. Western Litho Plate and Supply Co., 853 F.2d 1568 (Fed. Cir. 1988), noted in dictum that the reasonable royalty calculation “permits and often requires a court to look to events and facts that occurred thereafter and that could not have been known to or predicted by the hypothesized negotiators,” id. at 1575, the court did not go beyond the limited Georgia-Pacific rule allowing consideration of post-negotiation profitability as one of more than a dozen factors bearing on whether a royalty is reasonable. Recently, in its discussion of Georgia-Pacific factor 11 concerning “[t]he extent to which the infringer has made use of the patent, Georgia-Pacific, 318 F. Supp. at 1120,” the Federal Circuit noted that “[c]onsideration of evidence of usage after infringement started can, under appropriate circumstances, be helpful to the jury and the court in assessing whether a royalty is reasonable.” See Lucent, 580 F.3d at 1333-34 (emphasis added). Again, however, the court emphasized that the goal was to “recreate the ex ante licensing negotiation scenario,” id. at 1325, and that evidence of actual use was only one piece of evidence that could be used to support the reasonableness of a royalty analysis, id. at 1334 (noting that evidence of post-infringement usage is one piece of evidence, along with “sales projections

5

based on past sales, consumer surveys, focus group testing, and other sources,” that could help to estimate what the hypothetical negotiators would have believed about future usage).

But none of these cases hold—and the Federal Circuit has never suggested—that a court could simply substitute post-negotiation market facts for the results of a hypothetical negotiation—or, as here, the parties’ valuation during an actual negotiation—as the basis of a royalty determination, as the approach in the Order suggests doing.

B. The approach in the Order is incompatible with federal statutory and case law.

In addition to the absence of support for the approach in the Order in appellate case law, that approach is also incompatible with both federal statutory and case law in important respects. First, by substituting the 2008-11 actual market value of Android features (which necessarily would be measured as a percentage of Google’s actual Android profits) for the 2006 hypothetical value of those features, the approach would effectively permit a disgorgement remedy that is unauthorized by law. In 1946, Congress amended the patent-damages statute, 35 U.S.C. § 284, with the specific purpose of eliminating an infringer’s profits as a measure of damages. See Georgia-Pacific, 243 F. Supp. at 525-26 (purpose of the 1946 amendments was “eliminating the infringer’s profits as an independent measure of the patent owner’s recovery”). Accordingly, the Georgia-Pacific court noted that “Congress did not intend to aid a patentee in solving his problem of proving the quantum of his damages by enabling him to substitute the quantum of the infringer’s profits for the quantum of the patentee’s actual damages.” Id. at 519. Similarly, in Interactive Pictures, the Federal Circuit explained that basing a reasonable royalty on post hoc outcomes “would essentially eviscerate the rule that recognizes sales expectations at the time when infringement begins as a basis for a royalty base as opposed to an after-the-fact counting of actual sales.” Interactive Pictures, 274 F.3d at 1385; see also Radio Steel, 788 F.2d 1554 (Fed. Cir. 1986) (holding that a reasonable royalty is based not on the infringer’s profit, but on the result of a hypothetical negotiation at the time of infringement). The proposed approach in the Order would substitute Google’s actual 2008-11 results for what was anticipated in 2006, and then shift the burden to Google to persuade the jury not to rely on Google’s actual success as the basis for determining a reasonable royalty. This is exactly what all the above authority rejects.

6

Second, the Federal Circuit has never used proof of actual outcomes as an end in itself; it always has used such evidence as a check on experts’ claims of reasonableness. See Interactive Pictures, 274 F.3d at 1385; Trans-World, 750 F.2d at 1568. Put somewhat differently, the Federal Circuit has made clear that if there is good evidence of what the parties anticipated at the time of the hypothetical negotiation, then it does not matter whether those hopes were realized. Interactive Pictures explicitly affirmed the use of “projections [that] would have been available to [defendant] at the time of the hypothetical negotiation,” while also holding that “[t]he fact that [defendant] did not subsequently meet those projections is irrelevant to [defendant’s] state of mind at the time of the hypothetical negotiation.” 274 F.3d at 1384-85.

In this case, there is substantial, credible evidence of the parties’ expectations at the time of alleged first infringement—namely, the months-long series of back-and-forth negotiations, including the exchange of formal draft contracts, between Sun and Google in 2006. Those negotiations are the proper basis of a reasonable royalty calculation under Federal Circuit law. See Stickle v. Heublein, Inc., 716 F.2d 1550, 1556-61 (Fed. Cir. 1983) (reversing district court’s damages award in part because “notably absent from [the district court’s] findings is any consideration of the actual negotiations between the parties”). By contrast, the approach in the Order would use future market facts not as a check on the parties’ reasonable expectations, but as the basis for measuring the expectations. This would be damages based on “an after-the-fact counting of actual sales”—exactly what Interactive Pictures prohibits. 274 F.3d at 1385.

C. Because it was uncertain in 2006 whether Android would succeed or fail, the actual
market value of Android in 2008-11 is not a reasonable proxy for the expected
market value of Android in 2006.

As the Court noted in its Order, even if the approach set forth in that Order were viable as a matter of law, it would not make sense to use the actual 2008-11 value of Android as a proxy for the expected 2006 value of Android unless “the marketplace events, as they eventually unfolded, were reasonably predictable in 2006.” Order at 2. But the market events of the past three years were far from predictable at the time of the hypothetical negotiation.

In 2006, the smartphone market was in its infancy; none of the current dominant players, including Apple’s iPhone, had been introduced. Moreover, prior to developing and releasing

7

Android, Google had no track record in any facet of the smartphone market. It had never built or released a smartphone operating system or a mobile applications framework, let alone a fullstack operating environment like Android. Although the parties negotiated and the alleged infringement began in 2006, the actual architecture and feature set of Android was undefined at that point. Android was not even announced to the public until late 2007, and the first Android phone not released until late 2008—two and a half years after the hypothetical negotiation.

During this long lag time, there was no guarantee that Android would succeed at all, and substantial doubt throughout the industry and within Sun in particular. As Sun’s lead negotiator Vineet Gupta admitted at his deposition, “[A] lot of our customers did not expect Android would work, and they wanted to continue working with Java. So we didn't see it as a threat at all at that time [in 2008].” July 26, 2011 Gupta Dep. at 136:6-13. In 2008, Sun’s Chief Technology Officer James Gosling disparaged Android as “a bag of code” with “no business plan, no phones, no nothing … nobody is actually doing anything, nobody is actually shipping anything.” Trial Ex. 3104 (http://www.youtube.com/watch?v=thsklMITu0I). Similarly, even on Android’s release, most market analysts expected it to acquire at most 5-10% of the market, not over 40%, as it has done. Leonard Report at 39 & n.148 (citing analyst reports concluding that “no one is expecting Android to be a major success overnight,” with analysts such as J Gold Associates and Gartner respectively predicting that Android would achieve only a 5% or 10% market share in three years). Even as late as April 2010, Sun employees were dismissing projections of modest success by Android—which projections are now dwarfed by actual adoption of Android—based on initial Android sales. Trial Ex. 2229.

D. The approach in the Order would not simplify Oracle’s apportionment task.

Adopting the approach in the Order would not simplify the experts’ analysis, to the extent it is even possible. In Georgia-Pacific, the court repeatedly explained that Congress’s primary reason for abolishing an infringer’s profits as a basis for patent damages was that apportioning such profits had proven “insoluble.” 243 F. Supp. at 521-22. As the court observed,

if [damages are] to be measured by the amount of the infringer’s profits, it would require the ascertainment of those profits. This in turn would necessitate an accounting for profits . . . which would be open to all the criticisms which were

8

leveled at such proceedings [including] the often insuperable problem of apportioning the infringer’s total profits on the sale of a product between the patented and nonpatented features.

Id. at 525 (emphasis added).

Here, using the proposed framework, Oracle first would have to define the total 2008-11 value of Android, then isolate the percentage of that value provided by the patented features. Then Oracle would have to separate out “other know how [that] may also be required to practice the feature, such as licenses from other competitors and Google’s own independent know-how contribution to developing that feature.” Order at 2. Critically, unlike in the case of the Sun IP package, Oracle does not necessarily know every piece of Google engineering and know-how that contributed to the relevant Android features. That know-how was not the subject of discovery in this case. Further, the present value of Android and its features also reflects many other factors besides enabling intellectual property and Google engineering—at least including the strength of Google’s brand, its marketing efforts, and its relationships with its partners and customers—which are also unrelated to the claimed inventions and must be subtracted away.

Even after all of these discounts were applied, the end result would be only the maximum possible market value of the claimed inventions—not the result of a hypothetical negotiation. Obviously, a party that licenses technology generally does not end up paying the maximum market value of that technology, especially where it has alternative solutions (as Google did). It is unclear how an expert could account for this negotiating reality, particularly since the Court already rejected Oracle’s attempt to do so via the byzantine Nash Bargaining Solution.

By contrast, there is no reason Oracle could not have calculated, at least approximately, the value of the various components of the Sun IP package. Oracle knows what those components are; it understands how Sun valued them, both internally and in licensing; and it understands their functional importance to the various Java technologies. It cannot claim that it lacked the means to conduct the correct, legally mandated apportionment.

E. Using the approach in the Order would lead Oracle to further inflate its damages.

As discussed already, during the parties’ real-world negotiations in 2006, they exchanged various proposals, each of which set an upper limit for the value of Sun’s intellectual property.

9

Oracle prefers Sun’s initial demand of $100 million, while Google thinks it makes more sense to use Sun’s final demand of $28 million. Whatever the right baseline might be, both of these numbers account to some extent for the parties’ actual 2006 expectations about the success of Android. Accordingly, either figure is far more realistic as a measure of what a hypothetical negotiation would have produced and thus far more legally sound. Basing a hypothetical negotiation on the present market value of the Android platform, which would be the starting point for valuing individual Android features under the Order, would only spur Oracle to take yet another shot at posting the biggest possible—and still unsupportable—damages number.

F. Other comments on the Order.

First, if the Court does adopt the approach set forth in the Order, Google agrees it would be Oracle’s burden to separate out the various “know-how inputs” that enabled each of the allegedly patented features. Order at 2:21-28. Indeed, this point highlights yet another analytical defect in Dr. Cockburn’s analysis, which mistakenly calculated the value of the accused features of Android, which are enabled by many components other than the claimed inventions, rather than the value of the allegedly infringed intellectual property. Google believes the evidence will show that the real value of each of the features at issue is the result of Google’s work—most obviously the source code written by Google engineers.

Second, for the reasons set forth in its brief in response to the Court’s December 6, 2011 tentative order, Google reaffirms that the second expert report of Dr. Cockburn should be largely stricken and that Oracle should not get a third chance to draft a plausible damages report.

Third, with respect to the econometric and conjoint analyses Oracle used to apportion the current market value of Android, Google did not challenge those analyses under Daubert only because Google had a limited number of motions in limine available. Google requests permission to file Daubert motions challenging the admissibility of those analyses.

Dated: January 5, 2012

KEKER & VAN NEST LLP

By: /s/ Robert A. Van Nest
ROBERT A. VAN NEST
Attorneys for Defendant
GOOGLE INC.

10

682

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
SAN FRANCISCO DIVISION

ORACLE AMERICA, INC.
Plaintiff,
v.
GOOGLE, INC.
Defendant.

Case No. CV 10-03561 WHA

ORACLE AMERICA, INC.’S RESPONSE
TO COURT’S DECEMBER 27, 2011,
REQUEST FOR FURTHER BREIFING
ON DAMAGES (DKT. NO. 657)

Dept.: Courtroom 8, 19th Floor
Judge: Honorable William H. Alsup

Oracle America submits this response to the Court’s December 27, 2011 request for further briefing (the “Request”), seeking comment on questions regarding an alternative approach (the “Alternate Approach”) to reasonable royalty damages in this case.

I. INTRODUCTION

For the reasons set forth in its previous briefing and during oral argument on December 21, Oracle believes that Prof. Cockburn’s existing analysis is a sound approach to calculating damages in this case. The Alternate Approach also could be both feasible and appropriate under applicable appellate law, subject to at least one important caveat. Oracle has already measured the contribution of the patents and copyrights in suit to specific attributes of Android, such as improved application launch speed, enlarged memory, and an increased number of available applications. Oracle has also measured the incremental effect on mobile advertising revenue as a result of those specific enhancements. Using these ex post measurements to inform the ex ante hypothetical negotiation is supported by substantial appellate law under the circumstances presented here.

The one aspect of the Alternate Approach with which Oracle disagrees is Step 3, which would require that Oracle to further apportion the value of the relevant features among all of the know-how inputs that enabled them. Such a requirement is unnecessary as a factual matter because Prof. Cockburn has already isolated the specific contributions of the patents and copyrights in suit to Android, and has already analyzed the absence of viable non-infringing alternatives. Imposing such a requirement would also be legally improper because the Federal Circuit has never required a plaintiff to measure the value of the non-infringing elements of an accused product or feature. And such a requirement would be unworkable because of the complementarity problems that both parties’ economic experts have emphasized.

With the exception of the additional apportionment in Step 3, the Alternate Approach would provide an economically sound method of informing the amount of reasonable royalty damages. Under this approach, Oracle need not rely on the 2006 negotiations as the basis for its reasonable royalty claims. However, the 2006 negotiations may remain relevant, not only for potential defensive use by Google, but by either party to inform the nature and amount of the overall value at stake in the hypothetical negotiation.

1

II. RESPONSE TO ALTERNATIVE APPROACH AND QUESTIONS

STEP 1 – Measure the value to Android of specific features enabled by the patents and copyrights
in suit, as of 2008-2011.

Oracle agrees that this first step is an appropriate element of the reasonable royalty calculation. See Georgia-Pacific Co. v. United States Plywood Corp., 318 F. Supp. 1116, 1120 (S.D.N.Y. 1970) (Factors 9, 11, 13). As Google’s expert, Dr. Leonard, wrote last month: “Under a sound economic approach, the dollar amount of the reasonable royalty award should reflect the incremental dollar value of the patented technology to the defendant as compared to the next best alternative.” (Bailey, Leonard & Lopez, “Apportionment Treats The Symptom, Not The Disease,” Law360, Dec. 9, 2011, http://www.law360.com/ip/articles/289544 (“Leonard Law 360 Article”); see also R. Higgins & D. Martin, “The Economics Of The Entire Market Value Rule: As Applied to Complex Products” at p. 5 (2011), http://ssrn.com/abstract=1961276 (“An efficient royalty is equal to the marginal contribution of specific IP to the net value of the product or component whose creation the IP enables.”).)

As the Request notes, Oracle has measured the effect of the patents and copyrights in suit on various attributes of Android, such as application launch time. Although the Request states that Oracle has measured the value of certain Android features, it would be more accurate to say that Oracle has measured the incremental performance benefit from Google’s use of the patents and copyrights in suit. For example, Oracle did not simply measure how speed increases Google’s market share in the abstract, but instead measured the delta in market share attributable to the greater speed and memory provided by the patents compared to the characteristics Android would have had in the absence of the patented functionality, and the delta in market share attributable to the larger number of applications enabled by the copyrights.¹ Those deltas would have reduced demand for Android and thereby shifted usage to other smartphone platforms, which generate some – but not as much – advertising revenue for Google. Thus, in carrying out his apportionment analysis, Prof. Cockburn measured only the incremental readily quantifiable revenue associated with the patents and copyrights in suit.

____________________________________

¹ As discussed below with regard to Step 3, because Prof. Cockburn’s analysis is supported by, among other things, evidence demonstrating the unavailability of non-infringing alternatives, his evaluation of the incremental benefit of the patents and copyrights in suit specifically takes into account the value of the next-best alternative (i.e., relying on other mobile platforms for mobile advertising revenue). See Cockburn Report at ¶¶ 3, 16, 244, 387–416.

2

STEP 2 – Use data from 2008 to 2011 as an indicator of anticipated value in 2006.

The Alternate Approach notes that the evidence of actual value from 2008 to 2011 is an “arguable indicator of the value Google would have placed on that feature in 2006.” Oracle agrees. The Court asks “to what extent are hypothetical negotiators in 2006 allowed to peek into the future to see how events unfolded through 2011?” The Federal Circuit has repeatedly held that the jury may in appropriate circumstances consider evidence that came into existence only after the actual negotiation. See, e.g., Lucent Techs, Inc. v. Gateway, Inc., 580 F.3d 1301, 1333–34 (Fed. Cir. 2009) (“Consideration of evidence of usage after infringement started can, under appropriate circumstances, be helpful to the jury and the court in assessing whether a royalty is reasonable. Usage (or similar) data may provide information that the parties would frequently have estimated during the negotiation.”); Sinclair Refining Co. v. Jenkins Petroleum Process Co., 289 U.S. 689, 698 (1933) (factual developments occurring after the date of the hypothetical negotiation are a “book of wisdom that courts may not neglect”). In addition, Oracle and Google agree the jury should be instructed that:

In this trial, you have heard evidence of things that happened after the alleged infringement first began, including evidence of sales of devices incorporating Android and profits made by Google related to Android. That evidence can be considered only to the extent that it sheds light on what the parties would have anticipated at the time of the hypothetical negotiation. You may not limit or increase the royalty based on the actual profits Google made.

(Dkt. No. 539 at 160) (Disputed Jury Instruction No. 56) (emphasis added).

Evidence of the value consumers place on the patented functionality and copyrighted expression as of 2011 corroborates what Google itself anticipated in 2006, as shown by the contemporaneous evidence discussed by Prof. Cockburn. See Cockburn Report, ¶¶ 252–63. This is when “peeking into the future” is strongly supported by appellate law. The parties could have quantified the anticipated importance of speed, memory, security, and availability of applications, relying on data similar to the kinds of data relied on by Oracle’s experts.

STEP 3 – Apportion the value of the relevant features among all of the know-how inputs that
enabled it.

The Alternate Approach appears to contemplate that Oracle should be required to go beyond what Prof. Cockburn has already done and measure the relative contribution of all of the other “know-

3

how inputs” that enable a particular feature, in addition to the benefit provided by the patents and copyrights in suit. In the Request, the Court asks “[i]s it correct that the burden would be on Oracle to apportion the value of a feature as between the claimed invention versus all other know-how contributions to that feature?” As discussed above, Prof. Cockburn based his valuation of the patents and copyrights in suit on the understanding that the value to be measured is the delta between an Android with and without the patents and copyrights, by measuring, for example, exactly how many seconds faster an application opens with the patent enabled than without. If the Court is asking whether Oracle should be required to further apportion that delta among different inputs, Oracle respectfully submits that such a requirement is unnecessary, contrary to law, and unworkable.

Requiring allocation of features among all “know-how inputs” is unnecessary in this case because Oracle has already isolated the incremental benefit of the patents and copyrights in suit to the relevant features, and measured the increased advertising dollars that Google has generated as a result. Based on an extensive array of documentary evidence, admissions, and technical analysis, Prof. Cockburn has opined that there are no adequate non-infringing alternatives. Indeed, Google has essentially conceded that there were no non-infringing alternatives to these patents and copyrights other than to abandon Java altogether. See, e.g., Dec. 21, 2011 Hearing Tr. at 56:21–57:5 (“[Mr. Purcell:] The main [noninfringing alternative] would be just not using the Java programming language and a Java virtual machine or a Dalvik virtual machine, I should say, at all; programming Android from the get-go in another programming language, such as C, C++, Objective C, the language that the iPhone is programmed in.”)²; Opening Expert Report of Owen Astrachan at ¶130 (“Once Google decided to provide the ability for developers to write applications using the Java programming language, compatibility and interoperability with the existing body of software, tools, and knowledge about the

_____________________________________

² Abandoning Java was also not a viable option. Contrary to Google’s representation at the hearing (see id. at 57:6–58:6), the 2010 Lindholm e-mail is far from the only business document that contradicts counsel’s (and Google’s damages experts’) insistence that Google at most had “a slight preference for Java” at the time the infringement began. As Prof. Cockburn has observed, Google’s documents in 2005 and 2006 are replete with references to the “critical” need for Java and a license to Sun’s intellectual property, and to the fact – as Andy Rubin explained to Google co-founder Larry Page in October 2005 – that “the alternatives are sub-optimal” such that Google decided to “[d]o Java anyway and defend our decision, perhaps making enemies along the way”. See Cockburn Report at ¶¶ 406–16, 447–51.

4

Java APIs was an external factor constraining Android’s options. This essentially required Google to include the APIs at issue.”) Oracle’s engineers have conducted tests using industry benchmarks and rebuilt Android phones to measure specific effects, on speed and memory, of the patents-in-suit alone. See Cockburn Report ¶¶ 252–59, 264–67; Kemerer Decl.; Birger Summary. Through econometric analyses of smartphone purchase data and conjoint surveys, Profs. Cockburn and Shugan determined how many Android users would choose a non-Android phone if the incremental benefit of those particular patents were lost, and the phone’s performance diminished accordingly. See Cockburn Report ¶¶ 268–81. Prof. Cockburn’s conclusion is supported by both the affirmative conclusions of Oracle’s technical experts and by Google’s own experts’ conclusions. Google’s technical and damages experts largely assert that the available alternatives are using another programming language, removing the virtual machine, or removing the patented functionality altogether, not that Google had some specific workaround to provide the same functionality. Accordingly, the value of each feature is entirely the value of the patents and copyrights. No further apportionment of the features at issue to “other know-how contributions” is necessary or appropriate.

Even if some further allocation were needed (which it is not) to arrive at a value of the patents and copyrights in suit to Android, it would be unprecedented to require Oracle to arrive at that contribution by valuing the contribution of other know-how inputs. The Federal Circuit has never required a plaintiff to measure the value of the non-infringing elements of an accused product or feature. To the contrary, in Finjan, the court held that a jury could appropriately find that the infringed patents contributed a “substantial fraction” of the value of the accused product, even though there was neither evidence nor expert testimony concerning the value of the other know-how inputs that made the accused product useful and valuable to consumers. Finjan Inc. v. Secure Computing Corp., 626 F.3d 1197, 1211 (Fed. Cir. 2010). Other cases are in accord, and the Federal Circuit has, including recently, upheld damages on this basis, including under the entire market value rule. Bose Corp. v. JBL, Inc., 274 F.3d 1354, 1361 (Fed. Cir. 2001) (damages based on the value of loudspeaker systems that include accused ports allowable where patent was, among other benefits, integral to performance); Funai Elec. Co., Ltd. v. Daewoo Elecs. Corp., 616 F.3d 1357, 1375–76 (Fed. Cir. 2010) (damages award upheld where there was “general industry demand for smaller, cheaper, faster, and more reliable VCRs, and

5

Funai presented evidence that the patented technology furthers these goals”). The apportionment analysis that Prof. Cockburn undertook to value the incremental contribution of the patented technology to Android’s market share is precisely in line with, and in some respects more rigorous than, existing law.

Requiring allocation of the relative contributions of all the various “know-how inputs” that might be viewed as enabling a particular feature would also be unworkable. In 2007, Congress specifically considered an amendment to the patent laws that would have required essentially what the Alternate Approach proposes here, namely that, “[t]he court shall exclude from the analysis the economic value properly attributable to the prior art, and other features or improvements, whether or not themselves patented, that contribute economic value to the infringing product or process.” Patent Reform Acts of 2007, 110 H.R. 1908, section 5(a)(2); 110 S. 1145, section 5(a)(2). In response, then- Chief Judge Michel of the Federal Circuit wrote to Senators Leahy and Hatch that he was “unaware of any convincing demonstration of the need” for such a requirement, and that it was “a massive undertaking for which courts are ill-equipped” which would cause courts to “be inundated with massive amounts of data, requiring extra weeks of trial in nearly every case.” (Letter from Chief Judge Paul Michel, May 3, 2007, available at http://frwebgate.access.gpo.gov/cgibin/ getdoc.cgi?dbname=110_senate_hearings&docid=f:37760.pdf, at p. 277–78.) The amendment did not pass the Senate.³ STEP 4 – Hypothetical negotiation in 2006 between Sun and Google in which “[o]nly the expected
percentage contribution of the claimed invention to the overall expected value would be on the
negotiating table.”

Oracle agrees that the 2006 hypothetical negotiation can be viewed as a negotiation over the contribution of the claimed inventions to the value expected by the parties. Oracle also agrees that the

_______________________________

³ It is well-settled among experts in the field, including Google’s expert here, that there is no sensible way to allocate value among the numerous, often synergistic technical components of a product, device, or feature. See, e.g., Leonard Law 360 Article (“When the combined use of two or more assets is worth more than their individual use, there is no unique way to apportion the overall value of the product among the assets. Unless a particular apportionment scheme was specified by legislation, substantial legal ambiguity would be created, and courts, juries, and parties would bear a heavy litigation burden. Moreover, any mechanical rule to apportion the synergies among the various assets needed to create the synergies would be arbitrary.”).

6

hypothetical negotiation would concern the “overall expected value.” Limiting the negotiation to only anticipated incremental gains to Google in the form of near-term mobile advertising revenues, however, would substantially understate the value on the table. Google and Sun both expected, and expected to share, substantial strategic benefits from the success of a compatible Android.

As demonstrated by Google’s contemporaneous documents, Google expected to obtain important strategic benefits from a successful Android platform that was fast, stable, and attractive to application developers. See Cockburn Report at ¶¶ 252–63; 343–86 (reviewing contemporaneous evidence). Those other expected benefits included protecting Google’s non-Android mobile business (from which Google earns billions) and gaining access to new markets enabled by Android (e.g., links between Android and Google’s “Google+” social network service, and Google’s music and digital content services). Google is now seeking to protect that immense strategic value through steps such as its $12 billion acquisition of Motorola Mobility, which Google’s CEO has specifically described as designed to protect Android. Similarly, in the 2006 negotiations, both Sun and Google expected that Sun would receive a portion of the strategic value of a compatible Android by licensing commercial implementations of the platform and providing related services, which Sun called Project Armstrong. Cockburn Report ¶¶ 192, 200, 231, 288, 291, 296–301. Sun estimated that the revenues for Project Armstrong would exceed $600 million in the first three years alone, far in excess of any measure of the $100 million “starting point” license fee. Id. ¶ 298.

The value of these strategic benefits must also be accounted for in the reasonable royalty, subject to the apportionment analysis in Step 1. As the Federal Circuit has held, Oracle may seek a reasonable royalty based on both Google’s and Sun’s expected business advantage from using the patents in suit. See Powell v. Home Depot U.S.A., Inc., --- F.3d ----, 2011 WL 5519820, at *11–12 (Fed. Cir. Nov. 14, 2011) (“It is settled law that an infringer’s net profit margin is not the ceiling by which a reasonable royalty is capped.”) (emphasis removed); Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538, 1554–55 (Fed. Cir. 1995) (en banc) (affirming royalty based on patentee’s anticipated exploitation of patent rather than infringer’s expected gains); Monsanto Co. v. Ralph, 382 F.3d 1374, 1384 (Fed. Cir. 2004) (“the law does not require that an infringer be permitted to make a profit”). Nor are damages capped at the patentee’s expected profits. Powell, 2011 WL 5519820, at *11–12 (Fed.

7

Cir. Nov. 11, 2011) (rejecting argument that a patentee’s damages should be capped at his expected profits from selling the patented product to the infringer). Moreover, a reasonable royalty must factor in strategic considerations that would have been important to the parties. See Monsanto Co. v. McFarling, 488 F.3d 973, 979–80 (Fed. Cir. 2007).⁴

If the Alternate Approach were to consider only Google’s anticipated advertising revenue gains, while disregarding all other strategic benefits to Google and making no provision for Sun’s expressed interest in licensing its intellectual property only on terms that would allow it to earn hundreds of millions of dollars by sharing in the success and proliferation of a compatible Android, it would significantly understate the appropriate measure of Oracle’s damages. This would be inconsistent with “Congress’ overriding purpose of affording patent owners complete compensation.” See General Motors Corp. v. Devex. Corp., 461 U.S. 648, 655 (1983).

STEP 5 – Limit the relevance of Sun’s $100 million offer in 2006 to Google’s defensive argument
that it is a ceiling on the reasonable royalty; Google alone bears the burden of apportioning if it
chooses to make that argument.

The Court asks: “Is it correct that although Google might raise the $100 million offer by way of defense, Oracle would have no duty in its case in chief, if it used the above approach to allocate the $100 million as between the claimed invention versus the many other thousands of items in the 2006 package?” Oracle agrees that under the Alternate Approach, it would not need to rely on or allocate the 2006 starting point license.

Oracle does not understand the Alternate Approach to preclude Oracle from referring to the 2006 negotiations in its offensive case, but would object if it did. Limiting evaluation of the 2006 offer to defensive use in which the $100 million offer is merely apportioned down – failing to take into account that the purpose of the offer was to enable Sun to build substantial value on top of and

_____________________________

⁴ In McFarling, Monsanto customarily charged a $6.50 license for each bag of Round-up Ready soybean seeds, but its license forbade farmers from saving and replanting seeds. Instead, the standard license required farmers in succeeding years to purchase additional seed from authorized distributor seed companies, which were also Monsanto licensees. The seed companies then paid Monsanto an additional royalty, which amounted to another $19 to $22. 488 F.3d at 979–80. The Federal Circuit held that a reasonable royalty could not be limited to the customary $6.50, because such a license and royalty would deprive Monsanto of the benefits (including non-monetary benefits) of its “reasonable commercial strategy” of having seed companies cooperate to promote Monsanto’s product and control its distribution. Id. at 980–81 (affirming jury verdict awarding royalty of $40 per bag of seed).

8

alongside a jointly controlled and compatible Android – would be inconsistent with the facts and exclude significant elements of value from the reasonable royalty. As the Court has previously noted, should the jury find that Google’s actual infringement was incompatible and caused fragmentation, any “ceiling” should be adjusted upward from the starting point. (Dkt. No. 230 at 8–9; Dkt. No. 642 at 5– 6.)

Both parties should be free to cite the 2006 negotiations to the jury. See Univ. of Colorado Found., Inc. v. Am. Cyanamid Co., 216 F. Supp. 2d 1188, 1197 (D. Colo. 2002) (“A hypothetical negotiation should take into account the actual facts as they occurred in the matter both before and after the hypothetical negotiations would occur.”), aff’d, 342 F.3d 1298 (Fed. Cir. 2003). If either party chooses to do so, it must explain any adjustments, including apportionment, necessary to make those 2006 discussions sufficiently comparable to Google’s actual infringement to inform the hypothetical negotiation. Lucent, 580 F.3d at 1327–28 (requiring that party citing other license agreements to support hypothetical royalty demonstrate how those other agreements are comparable).

The license fees discussed in the 2006 negotiation should not, however, be used as an absolute cap on the total value of the hypothetical license. In Finjan, for example, the court rejected the defendant’s argument that the jury had erred when it awarded damages of $9.18 million for infringement of just three of its patents. On appeal, the defendant objected that the verdict was improper because the plaintiff had previously licensed its entire patent portfolio to Microsoft for only $8 million. Finjan, 626 F.3d at 1211–12. The Federal Circuit rejected defendant’s argument, commenting that the plaintiff had appropriately “noted multiple differences” between the portfolio license and the hypothetical negotiation, including differences in the competitive relationship between the parties and the fact that the plaintiff had received “significant intangible value” from the portfolio license. Id.

III. CONCLUSION

For the reasons stated in response to the tentative order and during argument on December 21, 2011, Oracle respectfully contends that Prof. Cockburn’s existing analysis is sound in its entirety, satisfying the level of precision and scientific rigor required by the appellate courts. Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 86 (1st Cir. 1998) (district court should not apply

9

“a standard of scientific certainty to the . . . testimony beyond that which Daubert envisions,” which turned his role from “that of gatekeeper to that of armed guard”).

Nonetheless, Prof. Cockburn could, based on analysis already included in his expert report, estimate reasonable royalties for the patents and copyrights in suit using the Alternate Approach, applied as described above. Indeed, Oracle believes that the Alternate Approach provides a useful framework for the jury’s assessment of a reasonable royalty in this case. Presenting and analyzing the evidence of the overall incremental value provided by the specific patents and copyrights in suit, consistent with the infringer’s expectations at the time infringement began and informed by the contemporaneous evidence viewed as a whole, is a sensible method of approaching the award of a reasonable royalty. However, the Alternate Approach would not comply with appellate law on reasonable royalty damages if applied in a manner that would (1) require that Oracle value all “know how inputs” to specific features of Android enhanced by the infringement, where Oracle has already isolated the incremental benefit of the patents and copyrights in suit; or (2) render irrelevant the parties’ 2006 negotiations except to the extent that Google can use them to argue that the 2006 offer represents the upper bound of any royalty.

Dated: January 5, 2012

BOIES, SCHILLER & FLEXNER LLP

By: /s/ Steven C. Holtzman
Steven C. Holtzman

Attorneys for Plaintiff
ORACLE AMERICA, INC.

10