Tuesday, July 26, 2011

Google's new anti-patent stance has four credibility issues -- but not the one many people think

The purpose of this present blog is and will continue to be analysis, not advocacy. I want to be right in terms of describing what's going on and predicting as accurately as possible what's going to come out of this, as opposed to pushing for reform. I dropped out of the fight against software patents in October 2006 and I've moved on definitively. But there are exceptional situations (such as this one) in which policy issues mix with the disputes and transactions I monitor.

Last evening, TechCrunch published an exclusive interview with Google's General Counsel Kent Walker, who condemned the patent system in by far the strongest terms any IT executive ever has to the best of my knowledge. According to TechCrunch, he "said that innovation itself is in danger", and here are a couple of verbatim quotes:

"A patent isn't innovation. It’s the right to block someone else from innovating."

"Patents are government-granted monopolies. We have them to reward innovation, but that's not happening here."

"An average patent examiner gets 15 to 20 hours per patent to see if it's valid. It can take years to go back and correct mistakes. It has become a kind of lottery."

At the highest level, "patents are not encouraging innovation," he says.

MG Siegler's very well-written TechCrunch article raised some good questions and connects those statements with the reality of what Google is doing as well as with possible reasons and motivations.

From my vantage point, those statements (even though I largely agree with them) have four rather serious credibility issues. Those relate to

  1. Google's history,

  2. a taste of "sour grapes",

  3. the fact that Google's respect for other people's and companies' intellectual property rights is doubtful, and

  4. Google's failure to support Android app developers against Lodsys and other trolls.

I will elaborate on those issues in this post. But before I get there, let me explain why I believe the most obvious credibility problem in many people's eyes -- Google's incessant efforts to buy patents -- is actually not an issue.

Non-issue: Google's efforts to buy patents

It may appear counterintuitive to say that Google's new anti-patent stance and its continued efforts to bolster its patent portfolio through acquisitions are unrelated. But that's the way I view it because I try to be pragmatic.

I'm already on the record saying that "Google must keep trying to buy strategic patent assets" in response to a Reuters question concerning Google's purported interest in patent holding company InterDigital. I believe it would make a lot more strategic sense for Apple to acquire that particular company, and I'd be surprised if Apple didn't outbid Google once again, but in general, Google must make an effort to address its strategic weakness in terms of patents. It should have been making such an effort for a much longer time, in fact -- long before the prices for such purchases skyrocketed.

When I wrote in January that "Google is patently too weak to protect Android", I was apparently the first one to raise this issue in the media, and I encountered disbelief. Now, about six months later, there's no more doubt since Google itself has admitted on several occasions a need to do something about it. What it does now could end up being too little too late for Android. But at least they must try.

I don't dispute anyone's right to try to perform well in a game under the existing set of rules while advocating change. If a particular soccer team makes use of the offside trap (meaning that in certain defensive situations they move up all players so that an attacking striker is in an "offside" position and not allowed to score), its leaders, members and fans still have every right to promote the abolition of the offside rule.

I do know that this perspective is not supported by many people in the free and open source software community, but even their organizations (such as the FSF and its various satellites) have no qualms about receiving most of their funding from pro-patent sponsors. That's really bad because those companies heavily influence the agenda. I also find it disturbing that the Open Source Initiative's favorite law firm files lawsuits on behalf of patent trolls. By comparison, I don't object at all to Google participating in patent auctions.

Credibility issue #1: Google is the biggest IT company ever built on a software patent

The first problem Google faces with its dismissal of software patents is that it's the biggest entrepreneurial success story owing to a software patent. Apple, Microsoft, IBM, Oracle, SAP -- none of them was built on a software patent. Those companies started before software was patented on a noticeable scale. They are now big in the patent business, but they don't owe their initial success to a software patent -- unlike Google. The Wall Street Journal recently mentioned that Google's founders received funding without even having a business plan in place:

You would have needed uncanny foresight or powerful pharmaceuticals to envision Google's success in 1999. Or maybe just money to burn. Kleiner Perkins and Sequoia Capital had something, because the two venture capital firms invested $12.5 million each, leading cynics in the Valley to define "Googling" as "getting funding without a business plan."

They may not have had a business plan, but the very basis on which Google's founders were able to start was U.S. Patent No. 6,285,999 on a "method for node ranking in a linked database", invented by Lawrence (Larry) Page and filed for on January 9, 1998. It was assigned to Stanford University, which later gave Google's founders an exclusive license. And it was an extremely broad one. Look at its claim 1:

A computer implemented method of scoring a plurality of linked documents, comprising:

obtaining a plurality of documents, at least some of the documents being linked documents, at least some of the documents being linking documents, and at least some of the documents being both linked documents and linking documents, each of the linked documents being pointed to by a link in one or more of the linking documents;
assigning a score to each of the linked documents based on scores of the one or more linking documents and processing the linked documents according to their scores.

Good luck trying to work around that one if you want to build a competitive search engine...

Of course, some may claim that perhaps Google might have succeeded even without patent protection. There's also no doubt that Google developed a whole lot of technology to manage large server farms that wasn't expressed in or by that particular patent. But Google as a company was built on the PageRank patent. The fact of the matter is that they had a software patent before they had a business plan, and there's at least a strong possibility that the patent played a key role in Google's ability to attract funding.

Interestingly, one of Google's investors and current board members, Kleiner Perkins partner John Doerr (widely considered Silicon Valley's and the world's most influential venture investor), recently raised on Twitter the question of what to do about software patents:

hmmm... how shd we fix patent system? ban or shorten sw patent life? see onforb.es/rpzdwZ and oreilly.com/news/patent_ar.... And LMK @johndoerr

But back when he invested in Google, it's quite possible that he viewed the patent as a key asset and might not have invested without it.

The broad PageRank patent may also have helped deter competitors from matching Google's quality especially in its critical early years. By now they're the dominant player in the search engine business and, as a beneficiary of strong network effects, have bigger problems with antitrust regulators than with potential imitators. Nevertheless, even a relatively recent SEC filing by Google states the following:

Intellectual Property

We rely on a combination of patent, trademark, copyright and trade secret laws in the U.S. and other jurisdictions as well as confidentiality procedures and contractual provisions to protect our proprietary technology and our brand. We also enter into confidentiality and invention assignment agreements with our employees and consultants and confidentiality agreements with other third parties, and we rigorously control access to proprietary technology.

The first version of the PageRank technology was created while Larry and Sergey attended Stanford University, which owns a patent to PageRank. The PageRank patent expires in 2017. We hold a perpetual license to this patent. In October 2003, we extended our exclusivity period to this patent through 2011, at which point our license will become non-exclusive.

Circumstances outside our control could pose a threat to our intellectual property rights. For example, effective intellectual property protection may not be available in every country in which our products and services are distributed. Also, the efforts we have taken to protect our proprietary rights may not be sufficient or effective. Any significant impairment of our intellectual property rights could harm our business or our ability to compete. [...]

The passages quoted above are not justifiable with the same arguments as Google's efforts to build a retaliatory arsenal as a deterrent for other patent holders. They stand in direct contradiction to the statements quoted by TechCrunch because those passages from an SEC filing underscore the value of patents to Google's business, based on such concepts as "exclusivity" and "proprietary technology". Also, note that Google's discussion of its intellectual property rights starts with patents. They could have started with copyright, trademarks and trade secrets, putting patents last. No, they put them first.

So are they against patents? It seems they like their own patents very much (including the absurd Google Doodle patent, which they spent ten years fighting for) and are only against other people's and companies' patents.

If you ever hear them denounce their PageRank patent as a youthful mistake or as an impediment to innovation, please let me know.

Credibility issue #2: a taste of sour grapes

Google would have had much more credibility as a critic of the patent system if it had spoken out at a time when it was either a winner or at least not a loser of the patent game.

Back in 2004, when I started the NoSoftwarePatents campaign in Europe (which I managed until the spring of 2005), Google was not willing to speak out against software patents. The CEO of one of the companies sponsoring my campaign contacted Eric Schmidt, then Google's CEO, and while Schmidt acknowledged receipt of the email, nothing ever came out of that contact. The same sponsor talked to Google co-founder Sergey Brin at a conference about software patents, and the answer he got was that they hold patents and consider them valid and valuable.

In June 2005, less than two weeks before the European Parliament's decisive vote on a proposal for software patent legislation, a fellow activist (Erik Josefsson) and I met in Brussels with a Google lobbyist and a Google programmer from the Netherlands. The lobbyist forwarded to various of her political contacts an activist group's proposed amendments to the bill. But to forward something is not the same as endorsing it. The European anti-software-patent movement didn't have any big company officially on its side. If Google had spoken out unequivocally at that point, it would have had major impact. They could have been honest and said that they did benefit from the PageRank patent but now that they've analyzed the issue more broadly, they must conclude that the impact of software patents is negative on the bottom line. Instead of doing that, Google's excuse was that the process was already at a late stage and they didn't want to join the debate at that point. But in a democracy there's nothing that would have had to prevent them from chiming in even on the day of the parliamentary vote.

The really disappointing part of that conversation was that Google's only real interest in connection with that European software patent bill was to ensure a so-called interoperability privilege. They only wanted to make sure they could continue to index PDF files without being deemed to infringe Adobe's patents. I told them that my view was the opposite one: software shouldn't be patentable in the first place, but if it is patentable, then there's no reason why Adobe shouldn't hold and enforce PDF-related patents.

In connection with the famous Bilski case, Google also stopped short of advocating abolition. Google basically told the U.S. Supreme Court that patents aren't bad but there are some bad patents out there. The alternative would have been for them to argue that the patent system is out of control and fails to "promote the progress of science and the useful arts", the purpose for which the U.S. Constitution allowed patents in the first place. Google's General Counsel now says that patents don't encourage innovation. Google's Bilski brief wasn't unequivocal. It was -- once again -- a lot more like "our patents are good but other people's patents are bad". But it was certainly a more patent-skeptical position than the one Google took in connection with that EU process six years ago.

The day after the Bilski decision I listed the top ten losers besides the "inventors" whose patent application was rejected. Number 7 on my list was Google's foray into new markets (Andoid, WebM). If I had to redo that top ten list today, it would certainly move up from #7 -- possibly all the way up to the top spot. But at that point, Google's problems with other companies' patents were already visible.

There really is a correlation between Google's problems with entrance barriers put up by others and its stance on patents. That's normal, but it does nothing to enhance Google's credibility. In particular, the statements Mr. Walker gave TechCrunch smack of sour grapes. Let's compare them to how he actually sought to justify Google's participation in the Nortel auction after Google won a pre-auction test bid to become the "stalking horse" bidder. Let me quote and comment on a few passages from the related blog post dated April 4, 2011:

The tech world has recently seen an explosion in patent litigation, often involving low-quality software patents, which threatens to stifle innovation.

He's right on the explosion of patent litigation, but the complaint about "low-quality software patents" was still pretty much like Google's Bilski brief: some software patents are bad because their quality is low, as opposed to claiming that all software patents must be abolished (arguing that most of them have a low quality so that the field as a whole doesn't lend itself to patenting).

Some of these lawsuits have been filed by people or companies that have never actually created anything

That isn't an argument against software patents. It's an argument against trolls, and a rather weak one, because if something is considered a legitimate invention deserving of patent protection, there's nothing wrong with a certain division of labor, with some focusing on inventions, others helping them with enforcement, and another group building products based on patent licenses.

Moreover, if they now buy patents, they also obtain protection without having created the technologies to which those patents relate.

Google is a relatively young company, and although we have a growing number of patents, many of our competitors have larger portfolios given their longer histories.

This distracts from the real issue. It's not that Google just has fewer patents because it's a young company. Patents have a term of validity of 20 years from the filing of an application, and Google is 13 years old. Google's founders were apparently also aware of the availability of patents even before they started their company. If this was just about Google's shorter history, then its run rate of new patent applications would have to be high, but it's still not at a very high level. Google simply doesn't do much of the kind of innovation that the patent system rewards. I like a number of Google services, very much in fact (including Google Plus, where you can find my profile at this address), but Google doesn't do much of the hardcore kind of R&D that companies like Apple and Microsoft spend more money on.

The last two sentences of that blog post are the most telling ones:

If successful, we hope this portfolio will not only create a disincentive for others to sue Google, but also help us, our partners and the open source community—which is integrally involved in projects like Android and Chrome—continue to innovate. In the absence of meaningful reform, we believe it's the best long-term solution for Google, our users and our partners.

Look at what he said: "the best long-term solution". He didn't say that it's a short-term fix under the circumstances. He had apparently given up on what he calls "meaningful reform", and instead wanted to buy a massive patent portfolio out of a bankruptcy estate (regardless of whether Google ever created any telecommunications equipment itself) -- again, as a "long-term solution". "Long-term". And now that this didn't work out, he takes a much broader anti-patent position.

Again, there's nothing wrong with trying to buy patents. But the position he took back in April would be much more consistent with the one he took yesterday if he had not presented that envisioned transaction as a "long-term solution".

Credibility issue #3: strong suspicions of willful infringement

When I was campaigning against software patents, the kinds of allies I was most uncomfortable about were those who were not only against software patents but had a broader anti-IP agenda. There were some in that movement who hoped that doing away with software patents would be the beginning of the end (or the end of the beginning, if you will) of a wider-ranging effort to weaken intellectual property rights. Not only did they have plans that would put me at loggerheads with them sooner or later (since I'm clearly in favor of copyright, and I'm not against all patents, though against many) but they also adversely affected the whole movement's ability to gather political support. A broad anti-IP agenda works only far left of the center. Center-left and (even more so) center-right politicians abhor it.

Google's track record in respecting intellectual property rights is problematic to say the least. Think of YouTube or Google's aggressive approach to its Google Books project.

In a very recent blog post I mentioned the latest statements by Judge William Alsup, the federal judge presiding over Oracle's lawsuit against Google:

Google may have simply been brazen, preferring to roll the dice on possible litigation rather than to pay a fair price.

The same judge previously raised the question of Google's suspected willful infringement of Oracle's Java-related intellectual property rights. I uploaded the related formal notice to Scribd.

That's a circumstance that casts serious doubt on Google's implicit assertion that Android's intellectual property problems are just due to the shortcomings of the patent system (combined with Android's success). It seems more likely to me that those problems are a combination of various factors, and some of those are Google-specific. That doesn't mean that there aren't also very general problems with software patents -- it's just that a company suspected by a judge of totally reckless, willful infringement is not in the best position to advocate intellectual property reform.

Credibility issue #4: Google's failure to speak out on the Lodsys mess

A patent troll named Lodsys has sent out infringement notices to numerous mobile app developers, including many Android app developers, and keeps sending out more all the time.

Lodsys is already suing a number of mobile app developers. Initially Lodsys sued seven -- mostly rather small -- app developers. That original lawsuit filed in late May already attacked one Android app (Labyrinth for Android) among mostly iOS-based apps. Lodsys just amended that complaint last week and also added Angry Birds for iOS as well as Angry Birds for Android to the list of accused products.

The fact that Lodsys asserted its rights against Android developers was raised in an Android developer discussion group post on May 26. Two months later, Google still hasn't spoken out. There was no reply even to the questions raised by Android app developers on Twitter, such as this message to Google's Android boss Andy Rubin, which was retweeted many times:

One of the affected developers wondered "why Google does not care about this":

While I believe Apple's motion for an intervention is not enough and app developers need blanket coverage, Apple has at least made some effort. It's not a basis on which one can recommend to app developers to ignore Lodsys's letters. Many iOS developers will probably end up paying for lack of a viable alternative. But Google doesn't say or do anything at all so far, and that's very disappointing.

Google's General Counsel complains about the problems the patent system causes to his company at this stage. It should be easy for Google's management to understand how much worse the situation is for all those little guys whom Lodsys and other trolls are shaking down.

I explained in a blog mid-June blog post that Apple's and Google's hands may be tied. It's possible that they are precluded from challenging the validity of Lodsys's patents under their own license agreement with Intellectual Ventures, a major patent holding organization that previously held Lodsys's patents and previously operated a small satellite with purportedly "the same exact address, down to the suite number", as Lodsys.

Google is an Intellectual Ventures licensee and investor. Its money enabled Intellectual Ventures to buy the patents that Lodsys is now asserting against Android app developers.

Google also finances RPX Corp., a publicly traded patent aggregator accused by antivirus software maker Kaspersky of extortion, racketeering and wire fraud. In other words, Google routinely provides money to operations whose patents create problems for other people, and in Lodsys's case, this is affecting Google's own developer base.

I believe that Google should take a clear position on patent issues affecting mobile app developers, and should explain to developers to what extent they can count on Google's help.

As long as Google doesn't do that, I can't help but view all of their criticism of the patent system as being motivated by their current problems. Looking at how their position has changed over time, and how self-contradictory it is in some ways, I guess Google's position may easily change again as soon as they've managed to buy a large patent portfolio in some kind of auction or other transaction.

Marco Arment, the creator of the wildely popular Instapaper app, recently wrote that he hopes Google will "get their ass kicked a bit by patent litigation so they're motivated to challenge the patent system more seriously than any of us ever could." That's not my approach -- but I can understand how an app developer watching this situation might take such a position.

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: