Sunday, August 14, 2011

Motorola doesn't have a license to kill Android but can cause problems

[Update] This post was published the day before it turned out that Google is trying to acquire Motorola Mobility. [/Update]

Before I get to the actual subject (Motorola's potential collection of royalties from other Android device makers), please bear with me for a few paragraphs on my modus operandi. If you're not interested in that, just click here to skip it and go straight to the Motorola story.

FOSS Patents' modus operandi

Most of you probably became aware of this blog because I frequently issue rapid-response comments on breaking news. For instance, my previous post commented on Google's reexamination requests against two Lodsys patents within about an hour of Wired's scoop on that development. I will continue to chime in at the outset of many news cycles because immediate reactions are very useful in the fast-paced age of online and social media. However, there's a whole lot going on now with patents in the mobile industry (and more media interest in it than ever before). In certain cases I will afford myself the luxury of waiting a couple of days before I take a position. I became aware of Motorola's statement on Android and patent royalties on Thursday but wanted to think about it thoroughly on a quiet day. I have a similar kind of post planned for tomorrow (Monday) on a different subject.

There's a third kind of article, of which I've also done a few: those posts outline my thinking on some broader strategic issues that aren't related to any particular news. Examples are my post back in January on Google's weak patent portfolio (a fact Google itself has meanwhile recognized), and more recent ones on Apple's focus on product differentiation as opposed to patent licensing revenues and the BRIC countries as a possible safe haven for Android.

I don't monetize this blog per se because I'm in the consulting (not the click-through) business. There's actually a paywall in place that most of you never see: if professionals want to discuss these issues in greater depth (and in a form that is specifically tailored to their research needs), I provide holistic analysis in the area where technical, business, intellectual property and sometimes political and regulatory considerations intersect. Most of my clients are in the financial services industry, and most of them are international organizations who direct questions to me from (mostly) the U.S., (increasingly) Asia, and (rarely, though that's where I'm physically based) Europe.

Such projects range from one-time assignments (for which I work with three agencies listed in my profile in the righthand column) to longer-term engagements. Financial investors are primarily interested in issues affecting long-term market shares of different companies, quarterly results (imagine the potential impact of a settlement between Oracle and Google, or a verdict in such a case), and they frequently evaluate the landscape with a view to particular transactions (mergers and acquisitions).

There are major synergies between my consulting business and my blogging. In both fields, I rely on the analysis of court filings as my primary research method and depend on a reputation for rational, facts-based positions and reasonably accurate predictions (to the extent possible). That's why I put my political positions (such as on software patents) and any possible loyalty considerations aside when assessing the likelihood of different outcomes of certain processes or developments. It wouldn't work any other way. And when I write comprehensive blog posts on strategy issues such as this Motorola-related topic, my blogging is particularly close to my work as an analyst.

Recap of recent patent-related statements by Motorola Mobility, one of its shareholders, and an analyst

UnwiredView was apparently the first website to quote a patent-related passage of a speech delivered by Motorola Mobility chairman and CEO Dr. Sanjay Jha at Oppenheimer's 14th Annual Technology & Communications Conference. The speech can still be viewed on the web. This is the part on patents:

"I would bring up IP as very important for differentiation (among Android vendors). We have a very large IP portfolio, and I think in the long term, as things settle down, you will see a meaningful difference in positions of many different Android players. Both in terms of avoidance of royalties as well as potentially being able to collect royalties. And that will make a big difference to people who have very strong IP positions."

Two weeks earlier, Dr. Jha had already said the following in a Motorola Mobility earnings call (second quarter of fiscal year 2011; the full transcript was published by SeekingAlpha):

"I also want to comment briefly on the topic of intellectual property. As most of you know, we own one of the strongest and most respected patent portfolios in the industry. We have over 17,000 patents granted and over 7,000 patents pending with particular strength in 2G and 3G essential, non-essential patents important to the delivery of competitive products in the marketplace, video particularly compression, decompression and security technologies and finally, a leading position in 4G LTE essential.

With new entrants to the mobile space resulting from the convergence of mobility, media, computing and the internet, our patent portfolio is increasingly important. We regularly review the company’s strategies opportunities and assets including IP with the goal of creating and enhancing value."

Another week earlier, investor Carl Icahn had suggested that Motorola explore strategic alternatives for its patent portfolio, such as a sale of all or most of that portfolio in light of currently very high valuations.

Even Icahn wasn't the first to think about a patent-oriented strategy change at Motorola. Back in April, reported on a research note from Global Equities' Trip Chowdry, who said that "selectively attacking with patents other Android phone OEM's is a better strategy [than trying to differentiate Motorola Mobility's products through software enhancements of Android]". Techdirt's related headline was the following: "Motorola's Best Play Is To Become A Patent Troll & Destroy Android Ecosystem With Patent Lawsuits"

Chowdry presented a chart of the patent holdings of different companies in the Android ecosystem and grouped them into three categories in terms of possible approaches by Motorola Mobility: "Deterrence" (meaning that Motorola would be protected against major patent holders Samsung and LG by the concept of mutually assured destruction), "Selective Engagement", and "FULL ATTACK". You can find that chart in the aforementioned article.

At around the same time, I already noticed that Motorola was reserving the right to sue other Android companies. When Google announced the "WebM Community Cross License" initiative in April, Motorola Mobility and its affiliated entities were notably absent as I pointed out in a discussion forum. The WebM CCL is a non-aggression pact among a group of companies with respect to Google's WebM codec offering. By joining, Motorola would have waived its right to sue other members of that alliance. Therefore, the fact that Motorola stayed away already said something -- though few people noticed back then.

Strategic conflict between aggressive patent enforcement and Motorala Mobility's operating business as an Android device maker

The statements made by Dr. Jha, Icahn and Chowdry suggest different ways to leverage Motorola Mobility's patent portfolio against competitors:

  • Chowdry's proposal would resolve any conflicts of interest between the two businesses by placing the emphasis on patent licensing as the more lucrative opportunity, but I believe Motorola Mobility will try for some more time to succeed in the handset market. That, in turn, will limit Motorola Mobility's flexibility. Only as a non-practicing entity would it be able to sue anyone else without any risk of retaliation.

  • Icahn, like Chowdry, believes that "the patent portfolio may be the company's most valuable asset". His idea of separating Motorola Mobility's patents (through a sale or spin-off) from its operating business was vague and general. In particular, three key elements were missing:

    • He didn't address (at least not according to the reports I saw) how many and which kinds of patents the operating entity would have to retain in order to be able to defend itself against competitors. A totally defenseless operating entity wouldn't survive in this environment. Compared to such an approach, Chowdry's proposal was actually makes more specific: he distinguished between different kinds of companies against which to assert patents based on their retaliatory power.

    • Another important limitation of the practicality of Icahn's proposal is the fact that Motorola Mobility already has a number of cross-license agreements in place and made FRAND commitments to standard-setting organizations (I'll get to that part further below).

    • He didn't say how Motorola Mobility could sell its patent holdings in the middle of wide-ranging patent disputes with Apple and Microsoft.

    Icahn probably just wanted to be provocative and hadn't thought this through. From his perspective as a shareholder, any impulse that results in more focus on patent monetization is probably useful. This is about pressure, not about a particular, convincing plan.

  • Dr. Jha's approach is a balanced one of keeping all options open. That's what he has to do since he's at the helm and needs to reach consensus. He can't just propose a radical departure from Motorola Mobility's business model while there are still opportunities left in the rapidly-evolving handset market. Dr. Jha's statement also makes more sense than Icahn's unclear ideas.

    In some ways, Dr. Jha's approach is closer to Chowdry's than Icahn's, but there's one very important difference: Dr. Jha appears to be fully committed to Motorola's operating business as a handset maker and unlikely to sacrifice that business in favor of a patent-centric approach. Much to the contrary, Dr. Jha proposes to leverage those patents in order to kill two birds with one stone:

    • generate additional revenue (at a gross margin of 100% since licensing doesn't come with any per-unit costs)

    • improve Motorola Mobility's competitive position especially vis-à-vis other Android device makers (while still seeking to differentiate through software as well)

    Dr. Jha says between the lines that Android device makers without a strong patent portfolio aren't going to have a viable business in the long run. Patent licensing costs will adversely affect their competitiveness. (And Dr. Jha didn't even mention the fact that it's not only about license fees but also product features when a patent holder like Apple seeks to optimize product differentiation.)

    What Dr. Jha suggests appears to be consistent with what other major players in the industry (such as IBM and Microsoft) do in terms of outbound patent licensing. He has certainly chosen a great market climate in which to highlight the strategic implications of Motorola Mobility's patent holdings for its overall business, and there's no doubt that he will look more closely than ever at patent-related opportunities, but he doesn't propose a fundamental strategy shift in response to the current patent-buying frenzy. Icahn was presumably hoping for something more impactful in the short term, such as an auction.

Let's look at the potential impact of Motorola's patents on Android, and at limitations that may already exist or could be imposed by Google.

The Android community is shocked

The headline of TechCrunch's article on Dr. Jha's patent statement started with this conclusion: "Android Patent Trouble Worsens" But the article ends with the assumption that Motorola would ultimately hurt itself (as a member of the Android ecosystem) by hurting Android as a whole, which is similar to the take of xda-developers' YouTube anchorman Azrie Noch.

Unfortunately, I don't believe that Motorola will look at the broader interest of Android as it decides on how to leverage its patents for its own purposes. Motorola isn't going to shut Android down, and it actually can't (for reasons I'll discuss below). Motorola would just try to create a relative cost advantage by taxing the competition. Motorola's direct competitors aren't Apple or other non-Android device makers -- it's intracompetition among Android device makers that worries Motorola the most at this stage. Google allows only some software differentiation (Motorola itself experienced the limits of that when Google bullied it out of shipping Skyhook's software on its devices). For the most part, it comes down to hardware differentiation (a field in which Samsung clearly has the lead, which is why I've already bought two generations of the Galaxy phone) and, incresasingly, price competition.

I'm not saying that this wouldn't have negative implications for Android. Obviously, Google's whole idea was to have a maximum degree of hardware and price competition, which has worked for a long time for "Wintel" (Windows + Intel) PCs. Patent royalties are useful to rent-seeking companies like Motorola but not in Android's best interest. Still, Motorola would be only one more patent holder taxing Android devices. This would make a difference, but only a gradual one (not a fundamental one) as compared to the status quo.

Motorola wouldn't be the only Android device maker to use patents against others in the ecosystem. Samsung or LG might consider the same course of action, to give only two more examples.

Let's take a logical approach to the interpretation of what Dr. Jha said.

Interpretation of Dr. Jha's statement with a view to Android

Just to recall the key part, he was talking about "a meaningful difference in positions of many different Android players" due to IP portfolios, "both in terms of avoidance of royalties as well as potentially being able to collect royalties".

The avoidance part is easily explained: it can only relate to other practicing entities since no one benefits from a strong patent portfolio when dealing with trolls. Non-practicing entities don't have products of their own against which one can countersue. Always keep in mind that a patent is the right to preclude someone else from doing something; it's not a guarantee that one is allowed to do what the patent describes (since one's implementation may very well infringe other people's patents). So Dr. Jha means that Motorola will have a cost advantage if, for example, it can settle its dispute with Apple on more favorable business terms than someone like HTC could. Indeed, there's a noticeable difference between how Motorola and HTC are duking it out with Apple: in December I published a chart that showed the state of affairs in those two disputes. Apple asserted a long list of patents against Android device makers. HTC brought underwhelming counterclaims, while Motorola attacks Apple quite hard. Even if one includes S3 Graphics' lawsuits against Apple, which I did in this more recent chart that shows Apple's dispute with HTC and its soon-to-be subsidiary S3, HTC's counterclaims and countersuits against Apple still aren't nearly as strong as Motorola's.

However, it's possible that HTC actually has a better patent licensing deal in place with Microsoft than the one Motorola will ultimately strike. Things are not going well for Motorola at the ITC, and Microsoft probably holds far more patents that read on Motorola's products than the other way round (here's a battlemap). HTC agreed to pay Microsoft, and was never sued.

But in general, it's true that Motorola is in a far better position than HTC -- and most other Android device makers -- to negotiate favorable cross-license deals with strategic patent holders.

The part on "potentially being able to collect royalties" doesn't state in a 100% explicit form that this relates to other Android device makers. But it's more than obvious to me that that's the way it was meant to a large extent.

Motorola is already embroiled in litigation with Apple and Microsoft. I think Motorola is going to end up having to pay Microsoft, just like I predicted that Apple was ultimately going to have to pay Nokia. Between Apple and Motorola, the only way that Motorola can "collect royalties" is if they enter into a cross-license deal under which Apple gets whatever it wants from Motorola while defending the exclusivity of certain features, especially those that make the iPhone/iPad user experience unique. In that case, Apple might gladly accept to pay a limited amount of money to Motorola, but this wouldn't be a great outcome for Motorola's competitiveness.

Considering that

  • litigation is already ongoing with respect to Apple's and Microsoft's platforms,

  • Symbian is being abandoned,

  • the BlackBerry is in rapid decline,

  • webOS has yet to take off, and

  • Mozilla's "Boot to Gecko" hasn't even been launched yet,

it's virtually impossible to see where Motorola Mobility could monetize its patents if not within the Android ecosystem.

And for the reasons I explained before, that's exactly where taxing the competition comes with a dual benefit: not only incremental revenues for Motorola but also higher prices for some of its most direct competitors' products.

In light of those obvious strategic reasons and the specific Android-related context in which Dr. Jha talked about royalties, I believe the correct interpretation is that he meant this generally (beyond Android) but had a particular focus on Android in mind.

I guess the instincts of the Android community were right: Motorola Mobility could soon become the first black sheep of the Android family.

Internecine war: Sony v. LG

Prior to current speculation about what Motorola might do, Sony's ITC complaint against LG (filed in December as part of a wider dispute that was settled a few days ago) came dangerously close to an Android v. Android litigation.

I took a close look at Sony's complaint. Sony asserted a variety of patents, at least some of which may very well read on Android -- actually, I doubt that none of them reads on Android. Sony brought the usual type of accusation, saying that it deemed those patents infringed "by LG phones, including, for example [followed by a list of certain non-Android devices]". Sony routinely added that it "reserve[d] the right to assert [any given] patent against additional and future models of LG phones as discovery progresses". Due to the settlement, this investigation never got to that point, but it could have.

If the ITC had ordered LG not to import products infringing certain patents, LG might have had (depending on the details of such a decision) a problem with respect to its Android phones even though those weren't named in the original complaint.

Strength of Motorola's portfolio

What matters in a patent dispute is not how many patents you have in total, but how many you have that your opponent needs, how likely they are to be upheld if their validity is contested in court or in reexamination, how likely they are to be infringed, and how hard it is to work around them. And don't forget possible encumbrances that limit their use against rivals.

It's obvious that Motorola Mobility's chairman wouldn't downplay the strength of his company's portfolio. Nor would a shareholder like Icahn, who may genuinely overestimate it. And Trip Chowdry (the analyst who suggested assertions against other Android companies) presented a simplistic chart that only showed how many patents each of those companies held, though many of the patents in question aren't related to mobile devices. For example, Samsung, LG and Motorola hold patents in rather unrelated fields of technology, and Google's patents in that chart were mostly search engine and closely related patents as opposed to Android-related patents.

Dr. Jha highlighted some areas in which Motorola Mobility believes to hold strong patents, and those were plausible. Many of them are patents on industry standards, which are (except under the very special circumstances of the Nortel transaction, which would be too much of a digression to discuss here) usually encumbered by existing FRAND licensing commitments (if you're unfamiliar with that concept, please read this fairly recent blog post, which discusses FRAND in the particular context of Apple v. Samsung but contains a lot of information that also applies to other patent holders). Indeed, both Apple and Microsoft have raised a FRAND defense against some of Motorola's assertions against them. So will others in future disputes, I guess.

Motorola also holds many unencumbered patents, and some of them are certainly in areas that are of concern to Android device makers.

I believe Motorola isn't as strong as most people may think based on the sheer numbers of patents it holds, but it's definitely in a much stronger position than most of the 39 companies that sell officially-licensed Android devices and likely able to impose a significant patent tax on many of those, possibly on everyone except Samsung, LG and SonyEricsson.

License terms can restrict patent enforcement

Now that I've addressed the obvious motivation on Motorola's part to gain advantages at the expense of other Android companies and talked at a high level about the strength of its portfolio, there's one very important question remaining: do they have a "license to kill" or are they restricted by existing license agreements?

The short version is that there are probably some restrictions but only of a limited scale, and even those limited restrictions could be avoided by Motorola if it dropped Android in favor of other operating systems, either a homegrown one (a possible plan about which there have been some rumors) or a platform like webOS or Windows Phone (which Motorola appears to be open to).

Note that device makers can theoretically use some (not all) of Android's program code without entering into an OEM license agreement with Google. But in that case they're not allowed to use the little green robot trademark, and they don't get some key components such as the Android Market and some Google-specific clients (Google Mail etc.).

Unfortunately, there's no way to know what exactly Google's agreement with Motorola stipulates with respect to patent enforcement against other Android companies. Google keeps those agreements secret and protected their confidentiality vigorously in the Oracle lawsuit.

Theoretically, anything is possible from no restrictions at all to a complete non-aggression arrangement. But the first extreme -- no restrictions -- is far more likely than the latter. It wouldn't be reasonably acceptable to Android device makers if Google required them to check in all of their patents at the door. For example, Samsung wouldn't make Android-based devices if everyone else could just build their patent-protected screens.

Concerning assertions against the operating software of those devices, things could be tricky with some difficult line-drawing to do between various layers:

  • Linux drivers:

    Proprietary drivers for Linux are a controversial topic, but to the extent they exist, they can also give rise to patent infringement assertions. No idea whether Google took any precautions in its Android OEM license.

  • Linux kernel:

    GPLv2, the license under which the Linux kernel is distributed, has some patent-related language in it but its limited in scope, and can be circumvented in various ways. If Motorola holds patents that read on Linux, it can most likely enforce them against other Android device makers unless Google's Android OEM license bars licensed device makers from doing so.

  • Apache-licensed Android software on top of Linux:

    Most of the software developed by the Android team is published under the Apache Software License. That license makes an aggressor lose his own license to a program if he asserts patents against it. That's a pretty good clause that GPLv2 doesn't have (GPLv3 tried to imitate it). As long as Motorola wants to sell Android-based devices, that clause might restrict its ability to enforce. The problem is that Motorola and other OEMs have their own license agreement with Google in place. That license agreement may or may not replicate or offset the patent defense clause in the Apache license. It's speculative but I wouldn't be surprised if it turned out that Google doesn't have any patent provisions in the Android OEM license apart from a disclaimer of liability in its own favor.

  • WebM codec:

    Google's WebM codec comes with its own license, which in turn comes with an "Additional IP Rights Grant". If Motorola sued WebM adopters while distributing WebM itself, Google would have the right to assert its own WebM-related patents against Motorola. Motorola might simply opt out of WebM, depending on whether or not Google allows Android OEMs to drop it (to my knowledge it's a mandatory component, but I don't know the details of Google's OEM license).

  • Closed-source Google clients:

    I don't know what Google's OEM license says about patent assertions against Google's closed-source clients such as Android Market, Google Mail, Google Maps etc. If Google restricts enforcement in any way, it will presumably protect its own clients in its OEM license agreements.

  • Vendor-specific Android enhancements
    (Motorola Blur, HTC Sense etc.)

    Those are highly unlikely to be protected by Google's OEM license agreement. Some OEMs invest significant amounts of money in such enhancements and wouldn't be comfortable doing so without strong intellectual property protection. Since Google is too afraid of some of Apple's patents to provide some functionality that matters a lot to consumers (especially some that are key to the user experience), those Android enhancements are critical. As a result of all of that, Motorola can probably bring infringement claims against enhancements such as HTC Sense (and HTC could conversely sue Motorola over its Blur software, too).

As you can see, it's complicated. Should Motorola have to sue other Android device makers who refuse to pay the license fees it demands, we may actually get clarification on some of this as the defendants in such cases might raise license-related defenses. At least we would see against which layers of the Android stack Motorola brings patent infringement claims. For now, there are many question marks, but I wanted to provide this overview so everyone can get a feeling for the issues that must be considered when trying to assess the potential impact of Motorola Mobility's intent to be more proactive on the Android patent front.

Should Motorola drop Android in favor of other platforms, then the Android ecosystem will have to get really scared. In that case, Motorola would likely be more aggressive and demanding, and not restricted in any ways.

Some believe that Motorola Mobility may threaten with patent enforcement against Android in order to pressure Google to buy it. I don't know, but it's possible that this is one of the intended effects. But even if Google doesn't buy Motorola, it should at least ensure that its license terms will prevent internecine patent wars between Android device makers to the greatest extent possible.

[Update] This post was published the day before it turned out that Google is trying to acquire Motorola Mobility. [/Update]

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 and Google+.

Share with other professionals via LinkedIn: