Wednesday, January 19, 2011

Google is patently too weak to protect Android

About a week ago, IFI CLAIMS Patent Services published its new ranking of the 50 companies awarded the most US patents in 2010. Microsoft is still the number three patentee with 3,094 new patents, Apple is a rising star with 563 new patents (+94%, a greater gain over the previous year than anyone else on the list), but Google isn't even on that list.

I ran a couple of queries on the USPTO's patent database. Last year, Google was granted 282 US patents and -- at the time of posting this -- owns 576 in total.

While Google has ramped up its patenting activity in recent years, the gap in portfolio strength between the Android developer and its mobile operating system competitors actually appears to be widening.

Quite often I talk to people who vastly "misoverestimate" :-) the power of Google's portfolio. There's a popular belief that all major high tech companies own tons of patents, and many consider Google fairly innovative. But its patent portfolio is dwarved by those of its competitors. Microsoft's portfolio, for instance, is 25 to 30 times larger (and, as I'll explain further below, much more diversified).

When I point out those facts, many people are just as surprised as I was when I learned a long time ago that Australia has almost the size of the "lower 48" states of the United States but only about 5% the number of inhabitants. Counterintuitive facts are nevertheless facts.

Android is a "suit magnet"

The number one patent-related trend in 2010 was, beyond all doubt, the flood of infringement suits that swept over the smartphone industry and that I expect to further intensify this year. Prior to 2010, there were a few cases, most notably NTP vs. RIM, which had the US BlackBerry service on the verge of a shutdown and resulted in a $612.5 million settlement, and Qualcomm's fights with Nokia and Broadcom, which also raised antitrust questions. But what's going on now is far bigger, and potentially far more devastating.

The operating system that is the target of more infringement action than any other is Google's Android. When one of those many suits was filed, I already quoted Maureen O'Gara, who noted that "[t]he freebie operating system is proving to be quite a little suit magnet." That's a funny metaphor, and by now it would be an understatement.

Here's a list of Android-related disputes that started in 2010 (in chronological order):

  1. Apple vs. HTC (see this overview of "Apple vs. Android")

  2. Oracle vs. Google (five months later, Google still hasn't been able to countersue Oracle)

  3. Interval Licensing vs. Google and others (analysis of amended complaint)

  4. Microsoft vs. Motorola (overview of state of affairs at end of 2010)

  5. Apple vs. Motorola (see this overview of "Apple vs. Android")

  6. Gemalto vs. Google, Samsung, Motorola and HTC (analysis)

  7. Vertical Computer Systems vs. Samsung and LG (analysis)

  8. Helferich Patent Licensing vs. Huawei (the mid-November complaint listed 24 licensees including Apple, Microsoft and HTC, but Chinese device maker Huawei apparently refused to pay)

  9. Multimedia Patent Trust (Alcatel-Lucent) vs. LG and others (analysis)

  10. Hybrid Audio vs. HTC, Dell and others (analysis)

  11. Hopewell Culture & Design vs. Motorola, Samsung, HTC, LG and others (analysis)

  12. Sony vs. LG (complaint; this is the first such suit of one Android adopter against another)

That's a dozen already, and I have no doubt there will be more. Ten of the disputes listed above started just last quarter. Some of those suits also target Apple (which has a clash of titans going on with Nokia), but fewer than Android although it would be economically more attractive; a small number of cases involve BlackBerry or other platforms. I'm not aware of any suit accusing a Windows Phone 7 device at this stage, although Microsoft is frequently sought out by patent holders.

Google can't solve Android's problems through cross-licensing

There is indeed a connection between the rampant enforcement of patents against Android and Google's weak patent portfolio. Only a few of the litigants are non-practicing entities while most of the plaintiffs, especially the ones asserting large numbers of relatively strong patents, sell products of their own. Those who actually have products on the market would have to think (at least) twice before attacking Android if they might need access to some of Google's patents.

Cross-licenses are the way most patent disputes between large companies are resolved. If there is parity in terms of how much each party needs the other company's patents, the deal may be done without money changing hands. In most cases, however, one company will have the upper hand and make a payment to compensate for the difference in portfolio value. Still, such payments tend to be much lower than the cost incurred by a "have-not" who needs a license from a powerhouse. In a price-sensitive, highly competitive market such as smartphones, the cost of patent licensing is eminently important.

No matter how influential Google may be on the World Wide Web, its patents apparently didn't deter Oracle from suing. I'm sure Oracle took a close look at Google's portfolio and determined that there was no risk of a serious counterstrike, or of any at all. If you look at my visualizations of other smartphone patent disputes involving major players on both sides (Apple vs. Nokia, Apple vs. Android, Microsoft vs. Motorola), you can see how they escalate and involve ever larger numbers of patents. By contrast, Oracle still doesn't face any infringement allegations.

If Google could countersue, it might already have a favorable settlement with Oracle in its hands. Since it can't, it will either have to fend off all seven patents asserted by Oracle (plus any others that Oracle could assert in a second suit), in each case by taking the patent down or proving that there's no infringement, or it will have to come up with some theory that it was entitled to a license of some sort. Otherwise, Oracle will prevail and the vast majority of Android applications would presumably have to be rewritten. So chances are this will cost Google (and possibly the Android ecosystem at large) dearly.

Even the 576 patents Google owns are a significant number and set it apart from many others who have even less. But when you're competing in a highly litigious environment in which a diversity of patents is required to build a solid product, that's too little. In Google's case, it's not just weakness in numbers. There's also a lack of diversification. I've looked at a random sample of Google patents, and most of them relate either directly to search or to closely related technologies such as location-based services.

Such a narrowly focused portfolio just can't frighten players like Apple, Microsoft or Oracle, all of whom innovate in a variety of fields of technology. I'm not saying that Google isn't innovative. It's just not innovative in the way that gets rewarded by the patent system with a sizable and diversified portfolio.

Google leaves its partners in the lurch

It's quite possible that even some device makers who adopted Android overrated Google's ability to defend Android against patent threats. Now they see that many patent holders seek royalty payments from makers of Android-based devices, and roughly a dozen have already gone to court with infringement allegations directed at Android.

Google probably doesn't have any contractual obligations. It puts out Android on open source terms. If things work out well, Google reaps most of the rewards. If things go wrong, others bear the brunt of the patent litigation (only three of the twelve suits I listed name Google among the defendants).

Obviously, those device makers knew all along that Google could benefit from Android, but they felt they could still benefit from selling the hardware. Patent issues may turn Android-based devices into an unprofitable business at some point, regardless of consumer demand, and at that point it will be hard for anyone other than Google to make any money with Android.

I also think that Sony vs. LG may only be the first of a number of suits in which one maker of an Android-based device tries to get rid of a competitor. I don't want to name names but I could see some Android device makers trying the same kind of cannibalization.

The Android patent situation would definitely be different -- fundamentally different -- if Google had a stronger portfolio and could show some authority. At a rate of a couple hundred new patents per year, and without much diversification, Google won't become a major force in this game anytime soon.

The patent weakness I described is also going to be a serious problem for Google's WebM codec. That's a slightly different but related subject that I've covered separately and plan to address again.

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: