Wednesday, July 27, 2011

S3 Graphics deal appears increasingly unlikely to help HTC against Apple

In my previous posts on the dispute between Apple and HTC, most recently this one (in which I explained why Apple won't just settle for money like most other patent holders do), I consistently pointed out that HTC's best chance to force Apple into at least a partial cross-license would probably be S3 Graphics' assertions against Apple. HTC is in the process of acquiring S3 for approximately $300 million.

Over the last couple of days, some additional information concerning the ITC investigation of S3's complaint against Apple (investigation no. 337-TA-724) became available. Based on this new information, it seems increasingly unlikely to me that HTC can get much leverage out of that process. It looks like HTC is going to have to use some other patents than the four at issue in that proceeding in order to put Apple under serious pressure. Right now it seems that the validity of those four S3 patents is uncertain, and even the worst-case scenario for Apple may just come down to a need to equip its Macintosh computers with NVIDIA chips. The iPhone, iPad and iPod product lines aren't even affected at all. Against that background, Apple can apparently keep pursuing its multiple lawsuits against HTC.

In addition to the hard facts that cast doubt on the impact of S3's assertions against Apple, there are other signs of weakness, such as HTC's public averments of its willingness to negotiate (though it takes two to tango and unilateral press conferences rarely bring another party back to the negotiating table) and, which I actually consider much more telling, HTC's attempts to slow-roll the ITC's efforts to make the details of its initial determination on the S3 complaint public. The details of that determination greatly diminish its business impact.

Before I get into more detail, let's look at the overall battlefield between Apple and HTC/S3:


I presented that battlemap before. The overview of the current state of affairs is the same, but I have added some detail to the slideshow that shows the escalation of the dispute. Apple dropped two patents from its ITC complaint: one in February and another one in April (during the hearing). As a result, only four (not six) Apple patents were still in play when the Administrative Law Judge issued his initial determination.

The battlemap shows that HTC asserts a limited number of patents of its own, and those aren't overly impressive. Apple is already asserting many more, and it could easily add more anytime.

Therefore, HTC's potential leverage against Apple depends very much on whether S3 Graphics would be in a position to obtain an import ban against product lines representing a significant part of Apple's revenue base.

Conclusions of law (on S3's complaint) were finally made public

On July 25, 2011, the Administrative Law Judge in charge of investigation no. 337-TA-724 (ITC v. Apple) ordered the release of the conclusions of law from his final initial determination. I have uploaded the order, which contains those conclusions of law, to Scribd.

That document states that on July 22, Apple's lawyers requested "that the conclusions of law in the final initial determination be made public." This shows that Apple wanted the world to see that any leverage HTC might hope to get out of the S3 case is much less than most people thought after they heard that an import ban might be ordered against Apple.

Conversely, the order describes in detail how S3 tried to slow-roll the process in a way that apparently annoyed the judge. S3 was slow in proposing redactions. The proposed redactions were apparently overreaching. While there can be some confidential information in such documents, there's also a certain level of transparency. And the worst part is that S3 wanted to appeal to the Commission -- the six-member body at the top of the ITC -- against the judge's decision to release the conclusions of law. The judge, however, pointed out that he "has the discretion pursuant to [the relevant ITC rule] to decide whether information designated by a supplier is entitled to confidential treatment in an initial determination". Therefore, the judge denied S3's request for leave to appeal.

You can read the detailed conclusions of law in the Scribd document I mentioned (starting on page 5). Any references to "[t]he accused iDevices and iOS SDK" state that those products and technologies don't infringe. Still, a massive threat to its Macintosh business could also be a reason for Apple to settle. But that threat appears limited:

While the initial determination finds "[a]ll of the accused Mac OS X Devices identified in [a section that was not published]" to "literally infringe" various asserted patent claims, all asserted claims of the '087 patent and the '417 patent were deemed invalid for obviousness, and more importantly, Apple doesn't face a problem for any Mac OS X devices with the NVIDIA GPU:

"20. An implied license and the doctrine of patent exhaustion apply to those Mac OS X Devices incorporating the NVIDIA GPU that are protected by the NVIDIA License (MacBook, MacBook Air, and Mac mini)."

Patent exhaustion means that S3 already granted a license to NVIDIA and therefore can't assert the related patents against NVIDIA's customer Apple. I discussed the concept of patent exhaustion in connection with the Lodsys case, in which Apple also uses exhaustion as a defense. S3 argued (unsuccessfully so far) that NVIDIA's license didn't apply here:

"21. The evidence does not demonstrate that the NVIDIA Term Sheet was unenforceable for lack of substantial performance."

But S3 did convince the court that there's no patent exhaustion with respect to an "Intel License", which may mean that an Intel graphics chip (or an Intel multifunctional component that includes the functionality of a graphics chip) wasn't licensed at all or at least not to the particular patents NVIDIA asserts against Apple:

"22. The doctrine of patent exhaustion does not apply to those Accused Products protected by the Intel License."

What I conclude from this (though not with absolute certainty since I'd need to see the actual agreements) is that Apple can, in a worst-case scenario (in which the ITC reaffirms the initial determination), work around the whole issue by incorporating NVIDIA graphics chips into its Macintosh computers and avoiding the distribution of Intel's allegedly infringing chips.

From an economic point of view, it would make a whole lot of sense for Apple to do this in the event the ITC orders an import ban, even if it resulted in a limited reduction of the gross margins of its Macintosh product line. There's a lot more for Apple to be gained by fending off Android than by settling with HTC only in order to maximize its flexibility in the Macintosh business.

For a realistic worst case, that's pretty manageable. But Apple may be able to avoid even that one.

Reexamination of S3's asserted patents

In addition to defending itself against S3's ITC complaint and bringing a declaratory judgment action against S3's asserted patents in a federal lawsuit, Apple requested reexaminations of those patents by the US Patent & Trademark Office.

The Litigating Apple blog believes HTC's strategy to use S3 as leverage against Apple "may have some holes in it" and points out the following fact:

The very day (July 1) the ITC issued its ruling that Apple infringed claims from the First Patent Group, the Patent Office issued actions in Apple's favor, finding that the relevant patent claims from the First Patent Group were not patentable. As such, all of the patent claims remaining in the ITC ruling against Apple now have been rejected by the Patent Office. Again, this all happened on the same day. As a result, the ITC has not yet had an opportunity to consider these new Patent Office rejections of the First Patent Group. What does this mean? It means Apple is likely eager to get these recent developments in front of the ITC for consideration.

If the term "First Patent Group" confused you, that's because the USPTO processed Apple's reexamination requests concerning S3's four asserted patents in two groups of two patents each.

I agree with Matt Macari that timing is important. Those first Office actions by the USPTO arrived by a hair's breadth too late for Apple to be able to influence the Administrative Law Judge's decision on that basis. Also, I could imagine a hypothetical scenario in which S3 obtains an import ban, Apple works around it by buying NVIDIA chips, and at some point the USPTO invalidates those patents, in which case the import ban would no longer be in force and effect, so Apple would regain its flexibility to purchase other graphics chips.

However, I tend to be cautious about first Office actions. As I wrote in connection with Oracle's litigation against Google, those aren't final. There are examples of patents asserted in high-profile lawsuits that also appeared to be in bad shape based on a first Office action but were ultimately upheld. That could happen here as well.

Also, the ITC and the federal courts don't necessarily take first Office actions into account. Patent holders have successfully prevented such non-final actions from being presented in jury trials. In this ITC investigation (where there is no such thing as a jury), Apple previously tried to convince the Administrative Law Judge to admit first Office actions as new evidence. The ALJ found those two S3 patents invalid, and maybe he was going to do so anyway, with or without those first Office actions. But Apple's request to admit those findings as evidence encountered resistance. S3 Graphics and the ITC staff (the Office of Unfair Import Investigations, which acts as a third party trying to defend the public interest) opposed Apple's motion. For instance, the OUII wrote:

"Contrary to Respondent's contention that the office actions are relevant and instructive, the Staff submits that at best the PTO reexamination office actions are informative of the proceedings in the PTO, but are not relevant or instructive on the issues of invalidity in this investigation. As an initial matter, the PTO office actions are not final office actions. Thus, the rejections are not a final assessment and determination of the patentability of the claims by the PTO examiner. Moreover, even if the office actions were “final” for purposes of the reexamination, they would still be non-final in that they would be appealable to the Board of Patent Appeals and Interferences and then to the Federal Circuit."

"The Staff submits that contrary to [Apple]'s contention, the introduction of the office actions in the reexamination proceedings is prejudicial to [S3 Graphics]. As explained above, the PTO office actions are not final office actions. The reexamination proceeding allows for several opportunities for the patentee to respond to the office action to rebut the PTO examiner's rejections. [...] Introduction of this issue at this late stage is unbalanced and prejudicial and should not be permitted."

That was from a June 23 pleading by the OUII, which is a third party. S3's lawyers opposed Apple's motion in even stronger terms:

"The reexamination file histories stand in a premature state, with the potential for any number of additional responses, arguments, amendments, office actions, and even appeals still on the horizon. Accordingly, based on the timing of Apple's requests for reexamination the PTO's processing of those requests, the record before the PTO at this stage is preliminary, entirely one-sided, and wholly unbalanced."

I guess Apple will try to get mileage out of the first Office actions that call into question the validity of the S3 patents deemed infringed in the initial determination. Apple will ask the ALJ to consider them. S3 will surely oppose this, and the OUII will probably raise similar concerns as in June. But even if the ALJ didn't formally admit those first Office actions as new evidence, he would see them (if he hasn't already) and they may affect his position on the validity of those patents anyway.

Just like HTC reaffirmed its intent to appeal the initial determination that found it to infringe two Apple patents, so is Apple still trying to defeat S3's compaint. We're going to see some more fighting, and if HTC wants a near-term settlement, it will probably have to remove certain features from its products, at least in the U.S. market -- and it might additionally have to pay royalties for a license to only some but not all relevant Apple patents.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.

Share with other professionals via LinkedIn: