On December 13th, Daniel B. Ravicher penned an article titled "Will Patent Office Eliminate Vringo's Right to Future Royalties From Google?" that was subsequently published on Seeking Alpha. The central argument of the piece is that significant risk exists to the future royalties Vringo stands to collect from Google for patent infringement: "Google's attack on I/P Engine's patents at the PTO could also terminate any Vringo entitlement to future royalties."
The piece contained a number of opinions, running the gamut from insinuation, to exaggeration, and all the way up to and including patently false statements. I have considered the issue of USPTO re-examination in the past, albeit with a vastly different assessment of the risks involved than Mr. Ravicher's. This article will respond to Mr. Ravicher's position on the potential impact United States Patent and Trademark Office actions may have on the legal grappling between Vringo (VRNG) and Google (GOOG).
The immense subjectivity that clouds the Vringo case has created a speculative market, and it demands an analysis of the arguments made, as well as the reasons that individuals make those arguments. As Mr. Ravicher correctly notes, his writing represents opinion, not financial or legal advice. Mr. Ravicher has been vocal in his opposition to software patents. In the 2008 Google TechTalk "Protecting Freedom in the Patent System," he takes the position that patents on software stifle scientific development: "If patents aren't increasing the rate of innovation in an industry, they should get their nose out of it. That's an extreme position, I know, but it's one I hold onto." As far as non-practicing entities like Vringo: "They're doing what's best for them, even though it's not good for society for people to go around chilling conduct and chilling research." And Mr. Ravicher has given lectures to Google employees on the subject of patents at Google TechTalks twice, in 2008 and 2012. He is the Legal Director of the Software Freedom Law Center, a non-profit that Google supports. A Google employee makes this relationship explicit at another Google TechTalk when he states, "We financially support Eben whenever we can and his work with the Free Software Foundation as well as the Software Freedom Law Center" (Eben Moglen and Mr. Ravicher are both Directors of the Software Freedom Law Center). In addition to his position at the Software Freedom Law Center, Mr. Ravicher is a Litigation Counsel of the Software Freedom Conservancy, an organization to which Google has "donated substantially." Mr. Ravicher occupies positions in organizations that have a material financial relationship with Google. Although no direct financial engagement exists between Google and Mr. Ravicher, it would be unsound to accept Mr. Ravicher's opinions without knowing his background. One can now dissect the opinions he voices in his article chronologically (emphasis is added to the points that will be argued).
I. Google's Re-Examination Requests
In its litigation against Google, Vringo asserted two patents, U.S. Patents Nos. 6,314,420 and 6,775,664 (the '420 and '664 patents, respectively). To be uber technical, I/P Engine asserted only a handful of the claims within each of the two patents, specifically claims 10, 14, 15, 25, 27 and 28 from the '420 patent and claims 1, 5, 6, 21, 22, 26, 28 and 38 from the '664 patent. The jury found in its verdict that each of the 6 claims of the '420 patent and the 8 claims of the '664 patent were infringed by Google and its co-defendants (customers who use Google's systems). The jury also found each of these claims were not invalid for anticipation. The judge will decide whether the claims are invalid for obviousness, although the jury did find several preliminary factual questions for the judge to consider on that issue in Vringo's favor.
The underlined portion of the paragraph is factually incorrect. A ruling has already been handed down by the court, finding the patents nonobvious. Mr. Ravicher voices this misunderstanding of what has already occurred in the case no fewer than 3 other times in the article.
In the meantime, Google has asked the PTO to reexamine each of the asserted claims in the two patents. First, back in March, Google filed a request for reexamination of the 6 claims of the '420 patent. In its request, Google set forth 11 different reasons why the various claims were invalid. In July the PTO granted Google's request, noting that all of the prior art submitted by Google was new to the PTO (i.e. had not been considered during initial examination of the '420 patent) and that there did not appear to be a specific reason given by the original examiner for allowance of the six claims. This was quite an admission by the PTO, to say that the Examiner who issued these claims did so without a specific reason. I guess it was close enough for government work.
Here, Mr Ravicher makes a factual case, but mischaracterizes the agency's granting of Google's ex parte request for re-examination as a significant and indicative event. Requests like Google's are granted a staggering 92% of the time, according to the USPTO. In fact, Mr. Ravicher is aware of this statistic - he cites it in his 2012 TechTalk. Given this, Vringo's position that Google's request for re-examination is "a standard and typical tactic used by defendants in patent litigation cases" illustrates the fact that this was not "quite an admission by the PTO" at all, as Mr. Ravicher mistakenly calls it.
In September the PTO issued an Office Action rejecting all 6 of the claims under review and adopting each of the 11 different arguments Google made as to why the claims are invalid. The Office Action is called "Non-Final" by law, as patentees like I/P Engine must be given an opportunity to respond to any rejections made by the PTO before the PTO can make those rejections "Final." However, that does not mean the rejections are not serious. The PTO could have decided in its Office Action to uphold some or all of the claims and/or it could have only adopted some of the arguments Google made. It did not do so. It rejected every claim and adopted every Google argument.
Vringo responded to the PTO's Office Action in late November, and now the matter is back before the PTO to decide whether to maintain some or all of its rejections. I find it very unlikely that the PTO will do a 100% about face and abandon all of the 11 rejections it made of the '420 patent's claims. Instead, I expect the PTO to stand by at least some of its rejections and issue a Final Office Action in the near future indicating that it intends to cancel all of the 6 challenged claims.
The author again states the facts correctly, but draws an exaggerated conclusion. Five critical reasons discredit this conclusion.
1. Empirically, of the 9090 re-examinations conducted from 1981-2012, 22% of patents are entirely upheld, 67% have some claims modified, and only 11% are entirely cancelled. What Mr. Ravicher expects will happen has only happened 1 in 10 times historically.
2. The feedback on the validity of the Lang patents to this date runs against Mr. Ravicher's opinion. Of all patents, only 1-2% are litigated. Mr. Ravicher, in his 2008 TechTalk Powerpoint, calls these the "most valuable" of the patent pool. of those 1-2%, only 10% make it to trial, and an even smaller percentage of the patents are subject to favorable jury verdict. Of that smaller subset, 30% of them are found to be invalid, and 40% are found to be obvious. The jury and Judge Raymond Jackson found none of these problems in regards to the Lang patents. They are patents of high caliber, making up less than 0.1% of all patents that the USPTO grants.
3. Non-final really does mean non-final. Mr. Ravicher has personal experience with how often the USPTO reverses its non-final rulings. In his 2008 TechTalk, he mentions his organization's victory in invalidating Microsoft's FAT patent, and even shows a Powerpoint slide of the 2004 CNET article, "Microsoft Fat Patent Falls Flat" as a testament to the success of the re-examination. Fast forward 2 years to 2006 and the USPTO reversed itself and upheld the patent. Non-final office actions have minimal predictive power in the direction of final rulings, and historical data on the current situation lies squarely in favor of Vringo.
4. Vringo's response to the re-examination is strong. Mr. Ravicher takes every opportunity to summarize Google's arguments, yet mentions Vringo's responses as blips that heed none of his attention.
5. The term "near future" is mistakenly used. In the anecdote above it took more than 2 years for the USPTO to reach a final decision. On average, the entire process from filing takes 25.4 months, and can last up to 10 years. Google requested re-examination in March of this year. Even if every single card falls exactly as Mr. Ravicher wants it to, Vringo has the right to royalties until the USPTO issues a final ruling on both the '420 and '664 patents.
In addition to the reexamination of the '620 patent that is completely going in Google's favor, Google just a few weeks ago filed a request for reexamination of the 8 claims of the '664 patent. In that request, Google made 8 separate arguments why the claims were invalid. The PTO is now considering whether to grant Google's reexamination request, which usually takes them about two months to do. I fully expect the PTO will grant Google's request to reexamine all 8 asserted claims of the '664 patent and then subsequently issue an Office Action rejecting those claims for at least some, if not all, of the reasons given by Google in its request.
Here, Mr. Ravicher extrapolates his previous conclusion on the '420 patent to the '664 patent. Yet the USPTO has not even granted a re-examination. It has not even taken initial Office Action. And it certainly has not made a final ruling. The assertion is specious at best. Most importantly, the jury found that Google infringed on every single claim of both patents. The jury did not attribute damages and the ongoing royalty to particular claims or one patent, but rather to the infringement. Mr. Ravicher himself makes this argument in his article on VirnetX (VHC). He states, "In reality for VirnetX to 'lose' means that every single patent claim it is asserting must be found to either be invalid, unenforceable or not infringed. So long as one claim from one of VirnetX's seven currently asserted patents is found to be valid, enforceable and infringed by defendants' products, VirnetX wins, and is entitled to seek both monetary damages for past infringement and an injunction to prohibit future infringement." Vringo secured a verdict on 2 patents and 13 individual claims. Interpret Mr. Ravicher's point regarding VirnetX's situation as you wish.
II. PTO Can Cancel Patents Even if Jury/Judge Find Patents Valid
Although I predicted the jury would find the Vringo patents valid, I believe the PTO will likely cancel (i.e. invalidate) all of I/P Engine's asserted patent claims from both patents now under review, or at least require they be changed to a narrower form before confirming them. This may seem as though I've changed my opinion on the issue of whether the patents are valid or not, but I haven't. The apparent difference in my opinion results from two important differences between reexaminations at the PTO and validity challenges in court.
First, patent claims under review at the PTO are given their broadest reasonable construction, whereas in Court patent claims are given the construction that the Judge thinks is most correct, whether that be broad or narrow. In fact, District Courts in some circumstances aresupposed to interpret patents narrowly so as to preserve their validity. A broader patent claim is more easily invalidated, as it is more likely to encroach upon the prior art, that would thus invalidate it. So, the fact that patents under reexamination are interpreted broadly means they're more likely to be found invalid. The Court's claim construction in the Vringo v Google case did not adopt the broadest reasonable interpretation of the claims, and therefore it was narrower than how those same claims will now be reviewed by the PTO. This means it is more likely the PTO will find those claims invalid.
The distinction Mr. Ravicher draws between the broader interpretation of the USPTO and the narrower scope of the judicial process is well made. Yet he somehow reaches for the insinuation that this difference means a higher probability that the USPTO will find the claims entirely invalid. The imminently logical conclusion of his point is simply that the broader nature of the claims will be found invalid, and the patent claims will be modified into a more narrow construction. This happens more than two-thirds of the time historically. If this is indeed the case, it does not effect the outcome in court, because as Mr. Ravicher notes, the court already interpreted the patents in a narrow view and the jury found that this narrow view was both valid and infringed upon by Google.
The second hugely important difference between reexamination of patents at the PTO and validity challenges in court is that the burden of proof to show invalidity is much lower at the PTO, where only a preponderance of the evidence has to show that the patent claim is invalid, whereas in court the evidence must be clear and convincing that the claims are invalid. This difference in burden of proof has already caused situations where the exact same patent was upheld in court but later found invalid through reexamination at the PTO. This result may seem strange, but it is completely in accordance with the law. Just this October, the Court of Appeals for the Federal Circuit, which is the exclusive Court of Appeals for all patent cases, said in the case of In re Baxter International: "The majority here concludes-rightly in my view-that a prior court decision in which a party has failed to prove a patent invalid does not bar the Patent and Trademark Office from subsequently reexamining that same patent. And, it concludes that, despite a final court judgment reaching a contrary conclusion as between the patent holder and one alleged infringer, the PTO is free to conclude that the patent is, indeed, invalid. That proposition is an unremarkable one." So, just because the jury upheld the I/P Engine patents as valid under anticipation, and even assuming the judge does not reverse that decision or find the patents invalid under obviousness, that does not mean the PTO will necessarily do the same. The PTO can -- and I expect will -- find I/P Engine's asserted patent claims invalid and thus either cancel them or require I/P engine to change them. If and when the PTO cancels the claims, they can no longer be infringed and Google would therefore not be required to pay any more royalties from that point forward. However, to avoid any confusion, a canceling or changing of the patents by the PTO would not require a refund of any monies paid by Google to Vringo prior to that time.
Again, Mr. Ravicher is correct in his summary of the distinction in burden of proof between the trial and USPTO re-examination. However, he pieces together facts into what amounts to be a very light web.
1. Even though the standard of proof is lower, the empirical statistics regarding USPTO re-examination subsume the argument. Even with a lower burden of proof, only around 1 in 10 patents are entirely invalidated.
2. Ex parte re-examination gives the advantage to the patent holder. What Mr. Ravicher does not mention is that Google could request re-examination for either patent only through the ex parte channel because the Lang patents were filed before 1999. Yet the ex parte procedure gives a marked advantage to the patent holder. As David O'Dell and David McCombs of Haynes and Boone LLP write, "as for ex parte reexamination, it is often regarded as insufficient because after a reexamination is ordered, the third party's participation is limited to one statutory reply prior to the examination process, which may only be filed if the patent own files a pre-examination optional statement." Google no longer has any further input into the re-examination process of either patent. Vringo does, including Lang's rebuttal that was submitted in late November as well as potential appeals in the future. Vringo controls the future direction of the re-examination as well as its pace. And only a final ruling on both patents that entirely invalidates every single one of the thirteen infringed claims would give Google an exemption from future royalties. Mr. Ravicher found this out the hard way in his loss against Microsoft in the USPTO re-examination process. As he lamented, "Microsoft has won a debate where they were the only party allowed to speak, in that the patent re-examination process bars the public from rebutting arguments made by Microsoft."
3. Even if the USPTO finds the Lang patents invalid, the court's judgment of past damages and ongoing royalties will most probably stand under the principle of res judicata. Mr. Ravicher provides an extremely shallow summary of the Baxter International case which he cites as support for his opinion. Although he is correct that the court found that the USPTO can invalidate patent claims that a court has already ruled on, the judges noted that "these conclusions do not mean, however, that when the PTO does act in the context of a reexamination proceeding, its conclusions can alter the binding effect of a prior judgment in a judicial proceeding" (In re Baxter International, Inc.) In the case, the patent office itself noted that "if a federal court awards relief to a patent holder against an infringer, a subsequent reexamination decision that the patent is invalid does not disturb the judgment of the court or alter its binding effect on the parties." In the remote instance that the USPTO invalidates every claim of both patents by 2014 or 2015, Vringo's past damages award and ongoing royalties to that point will be safe. In addition, the circumstances of the trial in question are vastly different - the jury had actually ruled that the patent claims were invalid and obvious, a verdict which was subsequently overturned by the Judge. Finally, the appeals court's ruling in Baxter did not resolve the issue permanently. The opinion contained dissent, and different judges on the Appeals Court may rule differently in Vringo's much stronger case.
After a fundamental review of Mr. Ravicher's position on the USPTO's re-examination of the '420 and '664 patents, it is evident that only a small part of the story is told. Yet the disagreement on the subject is also a reason that both sides exist in the marketplace. And Mr. Ravicher's willingness to take a contrarian position should be commended. Vringo remains a speculative investment rife with known and unknown risk. It is of the utmost importance for every investor to analyze the risks that surround the company from as many perspectives as possible.
Disclosure: I am long VRNG.