Access Advance pool doesn't really intend to solve duplicate-royalty problem: video codec implementers remain victims of unfair licensing scheme

This is a follow-up to 'Access Advance' patent pool can rename itself 'Suffer Setback' after Dusseldorf court ruling: flagrant FRAND abuse concerning video codec patents. In the previous post I discussed the logic of the Dusseldorf Regional Court's holding that certain Access Advance licensors owe Vestel damages for FRAND abuse, and cannot obtain injunctive relief against the Turkish TV manufacturer.

Weeks before I commented on this, Access Advance already sent an email to its existing and prospective partners, claiming that the outcome of the Dusseldorf cases was largely a win (as all of the asserted patents were deemed essential to the HEVC standard), and the denial of injunctive relief was disappointing but based on a narrow holding, which according to Access Advance wouldn't even have come down if a recent change to its Duplicate Royalty Policy had been made before--not after--trial. I disagree in at least three respects:

1. Access Advance can't explain away the significance of a denial of injunctive relief by a German court on unprecedented grounds. Dusseldorf is a video codec patent enforcement hotspot, and MPEG LA's licensors never faced any such issue.

2. While the court considered it dispositive that implementers who already had an MPEG LA license (such as Vestel) didn't enjoy legal certainty with respect to the avoidance or recovery of duplicate royalties when taking an HEVC Advance license, the court did express some skepticism concerning other aspects of the Access Advance business model--it just didn't have to analyze them to the full extent.

3. The duplicate royalty policy of the HEVC Advance pool is publicly available, and it doesn't withstand scrutiny. The problem is structural. The inadequacy of that duplicate royalty policy is crystal clear if one takes into consideration the way the Access Advance pool optimized its dealings with third parties (licensors as well as licensees) for just the needs of its four owners (General Electric, Philips, Dolby, and Mitsubishi). That conflict of interest is only going to exacerbate, as those four companies own a much smaller piece of the VVC patent cake than the one for HEVC.

It's high time that someone told the story of what's wrong with the Access Advance scheme. The more I've looked into this, the more I consider Access Advance a liability not only for HEVC but even more so with a view to the industry's transition to VVC (H.266).

"Unprecedented" would be an understatement for Access Advance's Dusseldorf disaster

In Germany, patent enforcement is all about injunctive relief. Damages don't matter unless you're asserting an expired patent. Access Advance's defeat is all the more shocking when compared to other video codec judgments in the same jurisdiction. I've been able to research 32 video codec enforcement campaigns (27 over MPEG-2, 5 over AVC) that started in Dusseldorf and were in each case brought by multiple MPEG LA licensors against partly large and sophisticated defendants (even including Huawei). 32 is a rounding error compared to more than 7,000 license agreements concluded by MPEG LA without litigation, but in absolute terms it's a large enough sample to serve as a point of reference here.

In at least six of those disputes, injunctions came down. Time and time again, MPEG LA's terms were considered FRAND, from the Dusseldorf Regional Court (where all those cases were filed) all the way up to the Federal Court of Justice. Is it just because MPEG LA's lawyers are better? Well, Krieger Mes's Axel Verhauwen and patent attorney Gottfried Schuell ("Schüll" in German) of Cohausz & Florack are extremely well-respected in the German patent litigation community. But I believe even they could not have defended Access Advance's duplicate royalty policy--a type of issue that never once came up in an MPEG LA case.

Access Advance's scheme to attract patent holders who are large-scale implementers

The duplicate-royalty issue facing those who previously took an MPEG LA (or direct) license cannot be fully understood without looking into how Access Advance entices a certain category of patent holders to join its pool as licensors. Not the only but by far the most important patent holder of this kind is Samsung, which holds far more HEVC patents than any other Access Advance member (arguably, Samsung alone accounts for about 40% of the pool)--but which is also an ultra-large-scale implementer with its phones, tablets, and TVs.

For a long time, during which the focus was almost exclusively on AVC/H.264, companies like that were actually happy to work with MPEG LA. They had an interest in encouraging widespread adoption of those standards, but due to their high-volume implementations of video codec standards they wanted to limit their spend on license fees.

So what was Access Advance able to offer an organization like Samsung--other than impeding the adoption of a standard and driving up licensing costs, none of which would normally be too appealing?

The story was that the likes of Samsung could still license all of MPEG LA's patents at the usual rate, but by joining HEVC Advance they also got the other patents in that (normally very expensive) pool at an extremely low cost in exchange for getting only a small payout from Access Advance relative to the actual size of their portfolios. If Samsung got its fair share of HEVC Advance's royalty distribution, it would get--like I said--about 40%. But it would drive up its royalty costs as an implementer, as Access Advance is like four times as expensive as MPEG LA. The oversimplified version is that Access Advance invited them to join the pool and the primary incentive was to save costs as an implementer, not to generate much of a license fee income as a patent holder--though the latter is what Access Advance's owners (GE, Dolby, Philips, Mitsubishi) very much care about.

Intuitively, one may ask: "What's wrong with that? Shouldn't patent holders who are also implementers be free to optimize their bottom line?"

Of course they can do that. The problem with the Advance scheme is just that its key success factor is not a more reasonable royalty rate--it's that others must be treated unfairly in order for Access Advance's owners to maximize their revenues and for certain patent holders who are large-scale implementers (first and foremost, Samsung) to optimize their bottom line.

And that takes us full circle back to the problem with duplicate royalties.

Access Advance's duplicate-royalty economics

Access Advance is not merely about double-dipping, such as if a patent holder licensed a supplier and had actually exhausted its patent rights, but tried to get paid again somewhere downstream. The problem is more fundamental than that:

• By the time HEVC Advance started, and especially by the time some other major patent holders defected from MPEG LA as a result of the type of deal I outlined in the previous section, wide swaths of the implementer landscape had already taken an MPEG LA license, and therefore remain licensed to some companies' patents even after those patent holders left MPEG LA. But Advance is all about "let's charge several times more." Its business model critically depends on extracting higher royalties from some who already have a license to many of those patents.

• It would seem logical that if a licensee can get a cheap license to a set of patents or an expensive one, the licensee will prefer the less expensive option. Alternatively, some might argue that the principled approach is that the first license you take is a done deal and must be respected, so any pool licenses you take subsequently must factor out the patents to which you are already licensed. Whether one agrees with the first approach (let the customer choose) or the second (earlier-concluded license shall remain in force), there is just no principled--FRAND--basis on which the second pool can tell licensees they should pay for another (and more expensive) license to patents they've already licensed, and should then seek a refund from the first pool when the second pool is way more expensive than the first. But that's what Advance has been trying for a while, and what it's still trying, though the Dusseldorf court has just complicated--if not thwarted--that scheme.

• Advance's economics just wouldn't work if they had to do the right thing, which is to reduce the royalty rate if a licensee is already licensed to many of the patents in the pool. That's because large parts of the portfolio come from patent holders like Samsung, who don't actually get much money out of the pool (because they joined with their interests as net implementers in mind). If, say, a licensee had to pay $100 million, 40% of which would be distributed to Samsung (minus management fees, which I'll talk about in a moment), and the licensee already had a license to Samsung's patents, then the royalty payment could be reduced to, say, $60 million. But here, while Samsung's patents would be used by the Advance pool in order to inflate the portfolio value when a court has to evaluate or determine a royalty rate, Samsung wouldn't get $40 million anyway, so there isn't really anything that Access Advance can deduct without its owners losing money on the deal.

• What makes all of this even more complicated is that Access Advance is known to charge very high management fees. In the aggregate of multiple types of fees and charges, Access Advance keeps roughly 40% of the royalty income, which is several times more than MPEG LA's cut according to what people in the industry say about it. The higher the management fee, the more of an incentive the pool administrator has to avoid any reductions (no matter how reasonable in light of prior licenses), and the harder it is to make an adjustment work economically for all parties involved.

For the reasons I've just explained, it is essential to the Advance scheme that implementers are forced to take a license at a far higher rate than the MPEG LA rate.

Access Advance's duplicate-royalty policy is fundamentally flawed and deficient

Having laid the foundation in the previous passages, let's now look at HEVC Advance's Duplicate Royalty Policy. That one is a joke.

The first glaring deficiency is that it talks only about patents that are "also included in the patent list of another patent licensing pool, or joint licensing program," without addressing the scenario in which someone took a direct (bilateral) license, which is not at all uncommon in this industry. Bilateral licenses can result from license agreements or even from patent exhaustion. The Access Advance folks know that, and it must be attributed to bad faith that their Duplicate Royalty Policy fails to address that problem.

Even with respect to pools and joint licensing programs, that policy is merely reflective of the problem and does not provide a workable solution. In a duplicate-license sceario, Access Advance does nothing for licensees other than to "refer the request to the applicable Licensor(s)" (footnote 3), which falls far short of what the Dusseldorf court wanted, which was legal certainty for licensees. Unless a licensor and a licensee reach an agreement (in which case one doesn't need a policy anyway because they're on the same page), each licensor then gets to choose the lesser of two refund amounts:

• the net distribution the Licensor received from Access Advance, or

• the net distribution the Licensor received from another pool or joint licensing program.

Footnote 5 clarifies that net distribution "expressly exclude[s] all fees and other allocations/deductions made by Access Advance prior to apportionment and distribution of net royalty collections to all Licensors based on their respective patent portfolios." One of the problems is that Access Advance takes a huge cut, which the policy declares immune from a refund. With a much more reasonable pool management fee, even licensees might accept not getting that part refunded, but in the case of Access Advance, it's a major economic issue.

Furthermore, Access Advance is so intransparent and asymmetrical that you can't just look at the percentage of the patents you've licensed and you know what refund you'll get: a Samsung may contribute lots of patents but not actually get much (if any) money out of the pool.

Another problem with "net distribution" is that if Samsung makes products and obtains a cheaper license for them by contributing its patents to the pool, why should--to use the current Dusseldorf examples as examples--Vestel and Xiaomi (who compete with Samsung) indirectly pay for that by Samsung's true benefits from is membership in the Access Advance pool not being taken into consideration in the refund context? It's so unfair and unreasonable.

There will be some licensors who get a lot of money out of Access Advance (especially its founders). They're obviously going to elect the more profitable option for them: they'll refer implementers to MPEG LA for a refund. Someone like Samsung, however, could argue--based on Access Advance's policy--that they don't really get much money out of that pool, so there's pretty much nothing they can offer as a refund.

Access Advance should not become the fox in charge of the VVC hen house

The root cause of all of the problems outlined in here (and I don't even claim that this is an exhaustive list of issues) is that Access Advance serves the interests of its four founders, then does asymmetrical deals with patent holders who are major implementers, and in the end tries to let the Vestels and Xiaomis of the world pay the bill, which in turn impedes the widespread adoption of the standard.

The exorbitant pool management fees I mentioned in the previous section are just an example of a construct that is just meant to make GE, Philips, Dolby, and Mitsubishi more money. The implications of asymmetrical terms (with some getting a lot of money out of Access Advance, while others are just interested in reducing their own licensing costs) for refunds are another example. And now that the industry is transitioning to VVC, there's another problem to bear in mind: the collective share in HEVC patents belonging to Access Advance's founders is far greater than their positions in VVC. IPlytics held an "Unpacking VVC SEPs and standards contribution data" webinar last summer. IPlytics explained how difficult it is to analyze who owns how much of VVC, but they gave it a try, and one can see that some companies who held no AVC or HEVC patents are now pretty big significant contributors to VVC (TikTok operator Bytedance, for instance), while GE, Philips and Mitsubishi don't seem to play a major role in VVC. Dolby is still somewhat significant, but not a leading VVC contributor either.

For Access Advance's founders, HEVC is therefore more lucrative than VVC. That's why it makes economic sense for Access Advance to offer a combined VVC-HEVC license with a certain discount: its founders own a larger share of a combined VVC-HEVC pool than a VVC-only pool.

With respect to VVC, Access Advance is an early entrant and may be able to avoid the problem that licensees got a less expensive license through another pool before Access Advance came to them and wanted them to pay even more for a license they already had. But the root cause of the Dusseldorf disaster is deep and structural. It's that Access Advance games the SEP licensing system to the benefit of a small group of companies. It's not a transparent pool administrator who treats everyone at arm's length regardless of whether a patent holder is a shareholder in the pool firm or not. Instead, Access Advance is a scheme that makes its founders money, makes NPEs money, saves patent holders who are major implementers money by giving them terms that are fundamentally different from the terms offered to founders and other net licensors, and then has to prey on other implementers in order to make money.

This is not the right way forward for the industry.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

TCL takes MPEG LA's AVC/H.264 license just before Dusseldorf trials in patent infringement litigation brought by NEC, Panasonic, IP Bridge

Yesterday Ericsson won a key appellate decision against HTC, and today there is good news for some other standard-essential patent (SEP) holders. Today, patent pool firm MPEG LA announced that "TCL Electronics Holdings Limited ('TCL') has become a Licensee to MPEG LA's AVC Patent Portfolio License" and that "all legal disputes related to patent enforcement actions brought by patent holders in MPEG LA’s AVC License against TCL have been resolved." In complaints filed with the Dusseldorf Regional Court and announced by MPEG LA on July 27, 2020, TCL was accused of infringing patents declared essential to the AVC/H.264 (MPEG-4 Part 10) digital video coding standard used in mobile devices, TVs, and other products.

I've been able to find out that the plaintiffs in those German patent infringement cases were NEC, Panasonic, and IP Bridge. The Dusseldorf court would have held trials pretty soon, but TCL folded.

MPEG LA has a track record of victories in Dusseldorf. MPEG LA itself cannot sue, as it does not own those patents, but its contributors do. Very often, if not always, MPEG LA's contributors are represented in court by Krieger Mes's Axel Verhauwen.

TCL appears to have read the writing on the wall. Just this summer, it suffered two defeats in the Dusseldorf appeals court (Oberlandesgericht Düsseldorf, or Dusseldorf Higher Regional Court) in cases involving a Via Licensing pool: as a result of being deemed an unwilling licensee, TCL was enjoined. Those cases involve a different pool, but chances are that TCL realized it had to start settling at least some of its video codec patent cases.

In July, TCL also settled one of the longest-running cellular SEP disputes by taking a license from Ericsson.

TCL gets sued left, right, and center, which also includes Mannheim, where LG obtained an injunction in March. They have to choose their battles wisely, and it apparently wasn't prudent anymore to decline to take a license from MPEG LA.

Finally, it's worth noting that MPEG LA, in the capacity of an amicus curiae, is supporting the Avanci cellular SEP pool firm against automotive supplier Continental's dubious antitrust claims. The United States Court of Appeals for the Fifth Circuit vacated a hearing scheduled for Monday due to Hurricane Ida. A new hearing date is yet to be determined.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Enforcement of German injunctions forces Huawei to take MPEG LA's AVC Patent Portfolio License

Qualcomm's enforcement of a likely invalid and most likely not infringed patent against Apple in Germany is a Pyrrhic victory that generates limited incremental chip sales but exacerbates its antitrust problems (instead of forcing Apple into a global settlement). But in another case, two German patent injunctions have brought about the desired result: MPEG LA, a patent pool company, "announced today that Huawei Technologies Co., Ltd., Huawei Device Co., Ltd and Huawei Device (Shenzhen) Co., Ltd ('Huawei') have become Licensees to MPEG LA's AVC Patent Portfolio License ('AVC License'). As a result of this agreement, all legal disputes related to patent enforcement actions brought by patent holders in MPEG LA's AVC License against Huawei have been resolved."

In November, MPEG LA announced that the Dusseldorf Regional Court ("Landgericht Düsseldorf" in Geman) ruled that Huawei and ZTE infringed patents of two contributors to its pool. In late December, MPEG LA announced that the Dusseldorf Higher Regional Court ("Oberlandesgericht Düsseldorf") denied Huawei's and ZTE's motions to stay the enforcement of injunctions over EP1773067 in Huawei's case and EP1750451 in ZTE's case. Both patents were filed for by Panasonic. But MPEG LA may have litigated over other patents as well (which would explain why they claimed that patents belonging to two pool contributors were enforced).

Presumably, ZTE will also settle in the near term. German injunctions can give a patent holder significant leverage, provided that they can't just be worked around.

MPEG LA appears to have a perfect litigation track record in Germany: whenever they sued someone there, MPEG LA prevailed on one or more patents. An MPEG LA press release states that "[a] team led by Axel Verhauwen of Krieger Mes & Graf v. der Groeben and Gottfried Schüll of Cohausz & Florack represented the plaintiffs." To the best of my knowledge, Mr. Verhauwen has represented MPEG LA in either all or at least the vast majority of its German lawsuits over the years.

It's unclear why Huawei and ZTE didn't just take a license, given that the cost of defending against a few patent infringement cases typically exceeds that of the annual royalty cap. It could be (and this is pure speculation) that those companies take issue with some non-monetary terms, such as a grant-back obligation.

In early December, 34 organizations signed an open letter calling for the application of the proportionality principle to patent infringement remedies in Europe. The list includes industry organizations such as CCIA and ip2innovate, major technology companies such as Samsung, Microsoft, SAP, Intel, HP, and Cisco, and automotive companies such as Volkswagen, Daimler, BMW, and Honda.

What makes Germany a particular hotbed for patent infringement lawsuits is the combination of speed, the enforcement gap (because it takes longer to invalidate a patent that should never have been granted than to obtain an infringement ruling) and, above all, access to injunctive relief as a legal remedy (as opposed to an equitable remedy involving an eBay-like analysis). In many cases, litigation watchers like me just have to look at the patent claims-in-suit to figure out that a plaintiff like BlackBerry (currently suing Facebook and its WhatsApp and Instagram subsidiaries) is simply playing the lottery with patents of dubious validity (software patents...), just because the grand prize is a Germany-wide injunction.

While I do agree with the above-mentioned open letter, the situation is more likely to get worse than to get better anytime soon.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Motorola Mobility takes MPEG-2 license from MPEG LA, settles disputes with Philips, Mitsubishi, Thomson

On Tuesday, patent pool firm MPEG LA announced that "Motorola Mobility LLC [a Google subsidiary for now] became licensed for its MPEG-2 products under the MPEG-2 Patent Portfolio License offered by MPEG LA" and that certain contributors to the MPEG-2 patent pool "agreed to dismiss patent infringement actions brought against Motorola in the United States District Court for the Southern District of Florida". In August I reported on those lawsuits filed by Philips, Mitsubishi and Thomson.

Law360 already reported in January on the stipulated dismissal of two of those lawsuits, a fact that indicated a settlement.

If Motorola Mobility had taken, in addition to an MPEG-2 license, an AVC/H.264 license, Google would no longer have been able to argue (on grounds that never convinced me) that its own MPEG LA license agreement didn't entitle Microsoft to a license grant-back on the same per-patent terms. I wouldn't be surprised to see Motorola Mobility take an AVC/H.264 license from MPEG LA as well -- or Lenovo may already have one. Anyway, Microsoft is not a contributor to the MPEG-2 pool.

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+.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Philips writes to ITC to defend MPEG LA AVC/H.264 reciprocal-licensing obligation

The grant-back (reciprocal licensing) obligation that licensees of MPEG LA's AVC/H.264 video codec patent pool must respect already came up in connection with Microsoft's enforcement of Google's (Motorola's) FRAND licensing obligations. Google tried to weasel out of that obligation, and Microsoft elected not to sue Google (the parent company) so as not to delay the proceedings against Motorola Mobility (the wholly-owned subsidiary), especially since the court-determined FRAND royalty rate was pretty close to the per-patent equivalent of the MPEG LA pool rate anyway. A Microsoft v. Motorola breach-of-FRAND-pledge jury trial is being held in Seattle as we speak, and the jury will probably try to render a verdict before the weekend.

It's interesting that Philips, a company known especially in European Union politics as a vocal supporter of strong intellectual property protection and known around the world as a top-notch licensor of standard-essential patents (SEPs), is also concerned about certain companies' attempts to shirk their back-licensing obligations and their related requests for injunctive relief, including but not limited to ITC exclusion orders (U.S. import bans). Philips submitted a public interest statement to the ITC in the investigation of an LSI/Agere complaint against Funai and Realtek. That ITC investigation previously drew a lot of attention in standardization circles when a senior district judge in the Northern District of California granted a Realtek motion for a preliminary injunction barring LSI/Agere from enforcing an ITC import ban should it win one at all. Based on the Administrative Law Judge's initial determination, there's no finding of a violation, but such preliminary rulings can be reversed as a result of a Commission review, which is why the parties to the dispute and third-party stakeholders must comment on public interest considerations now.

Given Philips's pro-IP, pro-enforcement tradition, I thought that it was going to make an argument in favor of SEP-based exclusion orders, only to find out that the Dutch electronics giant focused almost exclusively on the MPEG LA grant-back issue. Here's the filing (this post continues below the document):

13-08-30 Philips Submission to ITC on Public Interest Re. Patent Pools by Florian Mueller

The H.264 declared-essential patent Philips exclusively focuses on (apart from a footnote relating to a WiFi patent) was asserted only against Funai, not Realtek. No infringement was identified, but in case this changes during the review, Philips urges the ITC to "deny any relief for the '663 Patent as contrary to the public interest, or in the alternative, direct the ALJ to reopen the factual record to address [the] important patent pool licensing issues [Philips raises in its submission]".

There are certain differences between the Microsoft-Google disagreement on the grant-back obligation and the LSI-Funai dispute, even though both have their basis in the same grant-back clause in the MPEG LA AVC/H.264 patent pool license agreement.

One difference is not central to Philips's argument: LSI allegedly "was a licensor of the H.264 pool" until its "voluntary withdrawal in 2007". Motorola was involved in negotiations on the formation of the pool, but never actually became a licensor. Philips emphasizes that its submission "focuses on LSI's obligations as a pool licensee", which is also Google's status, but not without raising "the additional issue of whether LSI has complied with its more general FRAND obligation by attempting to license [Funai] on discriminatory terms compared to similarly-situated patent pool licenses". Philips, for example, is still licensed to LSI's H.264 patents because it was granted a license (by LSI, through the pool) before LSI's 2007 withdrawal from the pool. This question of discrimination is a very interesting one, but Philips doesn't elaborate on it. We may see it come up in some other context at some point (I'm not aware of anything in the pipeline, but this looks like the kind of issue that might resurface).

Another difference is that Google came up with theories that didn't really convince me as to why the patents held and enforced by its Motorola Mobility subsidiary somehow wouldn't fall within the scope of the grant-back clause. That is not an issue in the LSI case, at least not one that Philips mentions.

The bottommost sentence on page 2 of Philips's submission makes an important public-interest argument specific to the grant-back obligation:

"By ensuring that no licensee can take advantage of the pool license while asserting its own essential patents against other licensees, these commitments protect the incentives of licensors and licensees to join the H.264 pool, and in turn encourage the adoption of the H.264 standard."

I concur with Philips on this one. Hold-up through the pursuit of sales and import bans is totally unacceptable when a reciprocal-licensing obligation exists. The ITC should not allow anyone to perform an end-run around a grant-back obligation. The ITC has set a rather low bar (not infinitely low, but very low) for referring patents to arbitration. It would be inconsistent to make it easy to shirk a reciprocal-licensing obligation by handing someone an import ban over a patent that is reasonably likely to be subject to a grant-back clause.

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+.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Next patent infringement trial over Google's WebM/VP8 to be held on June 14 in Germany

On Friday I said that the VP8 video codec (part of the WebM project) still isn't off the hook with respect to patent infringement issues despite Google's recent license agreement with MPEG LA. Claims that VP8 is now free from per-unit or per-implementer license fees are grossly exaggerated. There are simply too many video technology patents out there, and the backers of WebM/VP8 are primarily companies whose own patent portfolios are too weak to resolve patent infringement issues through royalty-free cross-licensing. Simply put, the backers of VP8 are largely the losers of the patent game, while companies with strong patent positions prefer H.264.

A first Nokia v. HTC WebM/VP8 patent infringement trial was held in Mannheim, Germany on Friday (March 8, 2013). And that is only the beginning.

I have meanwhile found out that Nokia, which is known to have a strong position in video technologies and not a contributor to the MPEG LA AVC/H.264 pool (though it owns 36 patent families declared essential to that standard), is suing HTC over at least one other video codec patent allegedly infringed by Google's VP8.

The other litigation I found out about is scheduled to go to trial on June 14, 2013. The patent-in-suit in that action is EP1186177 on a "method and associated device for filtering digital video images". This is its claim 1:

"A method for reducing visual artefacts in a frame of a digital video signal, which is coded by blocks and then decoded, a block type being defined according to the prediction encoding method for a block selected from a predetermined set of coding types, the method comprising performing an adaptive blockboundary filtering operation on a block boundary formed between a first decoded image block on a first side of the block boundary and a second decoded image block on a second side of the block boundary, characterized in that the first decoded image block [has] been encoded using a first type of prediction encoding method and the second decoded image block [has] been encoded using a second type of prediction encoding method, wherein at least one parameter of the filtering operation is determined based on the types of the first and second prediction encoding methods, and the first and second type of prediction encoding methods are selected from a group of prediction encoding methods comprising at least: intra coding, copy coding, motion-compensated prediction code, and not-coded coding."

I plan to attend the trial and will then report on how the panel of judges over which Judge Andreas Voss ("Voß" in German) presides views Nokia's infringement allegations, and how how HTC and Google try to defend themselves.

Nokia does not have any FRAND licensing obligation with respect to VP8 (only with respect to AVC/H.264). Theoretically it could refuse to license patents that read on VP8 and seek sales and import bans against infringing products, while I'm not aware of any major patent holder who holds IP that reads on H.264 but didn't make a FRAND commitment. There is one such patent holder that made a FRAND commitment and doesn't really honor it (Google/Motorola), but defendants in H.264 patent infringement cases can at least demand a license on FRAND terms, while no implementer of WebM/VP8 can hold Nokia to a related FRAND pledge (because such a pledge doesn't exist). While it's less likely that Nokia would actually shut down implementations of VP8 if one of its patent families turns out to be essential to VP8, the fact that there's no FRAND commitment in place means that there's virtually no limit for how high a royalty Nokia could demand.

The Android ecosystem is taking Nokia's patent infringement claims quite seriously. One Android device maker sued by Nokia, ViewSonic, is already close to a settlement under which it will pay royalties (a fact that was confirmed in open court on Friday).

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+.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Microsoft, Google's Motorola complete briefing on implications of MPEG LA H.264 license agreement

More than three months after a FRAND trial was held in a Microsoft-Google contract case in the Western District of Washington, the rate-setting decision has not been made yet, due to the court's diligence on the one hand and Google's persistent efforts to evade its grant-back obligations under a license agreement with MPEG LA on the other hand. But it appears that briefing is complete now and Judge James Robart, the federal judge presiding over this litigation, now has all the facts in the file to determine a FRAND rate (a range and a specific number) for Motorola's standard-essential patents (SEPs). The rate-setting decision will pave the way for another trial: a breach-of-contract jury trial, at which the question is going to be whether Motorola's initial 2.25% royalty demand was blatantly unreasonable.

In order to discuss in greater detail the implications of a grant-back clause in the MPEG LA license Microsoft says entitles it to a reciprocal license from Google and its subsidiaries at the pool rate, which is a tiny fraction of the original royalty demand that corresponded to $4 billion in a conservative estimate, the court held a post-trial hearing on January 28 and requested three rounds of additional briefing:

• In time for the January 28 hearing the parties were requested to present "extrinsic evidence" relevant to the interpretation of the MPEG LA-Google license agreement (which has to be construed under New York law). Microsoft submitted a declaration by MPEG LA chief executive Larry Horn in support of its position on grant-back. Google's Motorola, which only had a declaration by its own lawyers on its side, could have figured that MPEG LA was going to be key here but didn't request the right to depose an MPEG LA official until a point at which this would have delayed the whole process. Therefore, the court denied a related motion.

  Microsoft additionally submitted a declaration that shows the addition of the word "Affiliates" to the grant-back clause.

• On February 14 the court requested further briefing on what Google's Motorola describes as an executed and enforceable H.264 license agreement in Germany while Microsoft argues that Google actually made a counteroffer which Microsoft rejects.

• On February 21 the court gave the parties until yesterday (Friday, March 1) to submit briefing on the implications of the mentioning of "Affiliates" in Section 3.1.7 on the royalty obligations of "Enterprise Licensees" in connection with a royalty cap for "Motorola's obligation to license its [SEPs] as an Affiliate of Google". The court asked the parties to comment, in particular, on its interpretation "that Licensee [i.e., Google] and its Affiliates [i.e., Motorola and its former subsidiary General Instrument] fall under the royalty cap prescribed therein only if the Affiliates are themselves licensee[s] under the AVC Patent Portfolio and are identified by the Licensee in writing to [MPEG LA]".

Now that the parties have briefed the court I'm going to explain their positions. Google has been arguing for some time that it doesn't have to honor the grant-back obligation (or at least not to the extent that the MPEG LA royalty caps would proportionately apply to the license fees Microsoft is going to pay) because it selected in 2012 (unlike in the past) an "Enterprise License", which (as I'll explain further below) is an option relating to some AVC/H.264 license offerings and potentially saving a group of affiliated companies money. This isn't Google's only argument against reciprocity, but it might be the last one the court needs resolved before the rate-setting decision comes down.

Google is trying to benefit from the fact that MPEG LA decided to write up a license agreement covering multiple fields of use for H.264 in one document. The confusion that Google is trying to create here could have been avoided in the first place if there had been separate (even if largely overlapping) license agreements for different types of use, but I can understand that MPEG LA, which has well over a thousand licensees for the AVC/H.264 pool alone (which isn't its only patent pool), preferred an all-in-one contract over two or more largely-duplicative separate agreements -- also because this was a way to ensure that a company wouldn't take a license covering one field of use but elect to infringe in other fields of use.

I'll show you the documents and then discuss the parts I consider relevant. You can click here to skip the documents and proceed to the discussion.

Documents

It was previously known that Google signed the same license agreement with MPEG LA as more than 1,100 other licensees. The actual document bearing a Google executive's signature -- signed in 2005 and renewed for another five-year term in 2011 -- was attached to MPEG LA chief executive Larry Horn's declaration:

Google's MPEG LA License of 2005 by

This is Google's (Motorola's) March 1, 2013 brief:

13-03-01 Motorola Letter Re. MPEG LA-Google License by

And here's Microsoft's March 1, 2013 brief:

13-03-01 Microsoft Letter Re. MPEG LA-Google License by

Discussion of relevant parts

There are two basic categories of MPEG LA H.264 licenses:

1. licenses that you need if you make technology products capable of encoding (for a consumer's own use or for transmission to a consumer) or decoding (by any consumer) video files in the H.264 format, and

2. licenses covering the distribution (especially, but not only, Internet broadcasting) of video files encoded in the H.264 format.

Either category has subcategories:

1. For the technology product licenses a distinction is made between the sale of such products ("AVC Products") to consumers and OEM distribution (for example, PCs running Windows, which comes with H.264 functionality). There's no difference here in terms of what the end user gets. It's just a commercial thing. As a result, Microsoft needs two licenses because it sells Windows directly as well as through OEM channels.

2. The distribution/broadcasting licenses include Title-by-Title AVC Video (note that AVC and H.264 are synonyms for purposes of this license agreement), Subscription AVC Video, Free Television AVC Video, and Internet Broadcast AVC Video Use. The names already indicate the scope of those licenses, but in connection with the grant-back obligation we can just lump all distribution/broadcasting licenses together.

The license agreement defines these fields of use, as well as numerous other terms, in Section 1, "Definitions" (pages 3-8 based on the page numbering of the agreement; pages 4-9 of the PDF document). The license grant clauses incorporate those definitions by reference. Those clauses are sections 2.1-2.7 (pages 8-9 based on the page numbering of the agreement; pages 9-10 of the PDF document). The order of those license grants does not reflect the two categories I discussed above. The tech product licenses are granted in sections 2.1 ("AVC Products") and 2.6 ("OEM Licensee"); the distribution/broadcasting licenses are granted in sections 2.2-2.5; and section 2.7 then offers licensees the option to choose for each calendar year a bundle called "Enterprise License", which represents the combination of any or all distribution/broadcasting licenses chosen (but, to be perfectly clear, does not involve any tech product licenses) and comes with the option to apply the annual royalty cap to a group of affiliated companies.

The royalties sections (3.1.1-3.1.7; pages 10-14 based on the page numbering of the agreement; pages 11-15 of the PDF document) have the same order as the license grant sections (2.1-2.7).

The Enterprise License clauses are 2.7 for the license grant and 3.1.7 for the royalties. This is how the Enterprise License royalties clause begins:

"Enterprise Licensees. Pursuant to Article 2.7 and notwithstanding anything to the contrary in Article 2.9 hereof, and in lieu of the royalties specified in Articles 3.1.2, 3.1.3, 3.1.4 and 3.1.5, a Licensee and its Affiliates which are licensees under the AVC Patent Portfolio License and are identified in writing to the Licensing Administrator by Licensee shall pay no more than the following total amounts in each Calendar Year for all such licenses for the combined Sales of Licensee and its Affiliates during such year:"

I'll now make it easier to understand by substituting section numbers with descriptions and defined terms with actual party names:

"Enterprise Licensees. Pursuant to [the Enterprise License grant section] and notwithstanding anything to the contrary in [a section related to the Enterprise License that disallows Google to sublicense its affiliates, but entitles its affiliates to licenses from MPEG LA if they request them], and in lieu of the royalties specified in [the distribution/broadcasting royalties sections], [Google] and its Affiliates which are licensees under the AVC Patent Portfolio License and are identified in writing to [MPEG LA] by [Google] shall pay no more than the following total amounts in each Calendar Year for all such licenses for the combined Sales of [Google] and its Affiliates during such year:"

The respect in which the Enterprise License terms are different from the tech product and individual distribution/broadcasting licenses is that the royalties section requires that a licensee identify in writing its affiliates which are also licensees, so as to ensure that the royalty cap is applied to the collective license fees paid by the whole group, as opposed to applying the cap to each company from the group.

Google says it exercised this potentially money-saving option in 2012 but didn't identify Motorola Mobility as an affiliate to benefit from the group-wide Enterprise License royalty cap, and on this basis argues that Motorola Mobility consequently doesn't have to grant Microsoft a license to its own H.264 SEPs, or at least doesn't have to do so on terms that include the royalty cap it doesn't benefit from.

The grant-back obligation in Section 8.3 clearly includes Affiliates (whether notified or not) and is meant to be "commensurate to the scope of the licenses which Licensee has selected [under the agreement]". Google now contends that the "scope" includes whether it includes a particular subsidiary (Motorola Mobility in this case) in its notifications relating to the Enterprise License ("selects to have its Affiliates licensed"). But Microsoft says that "the Agreement provides no ability for Google to make any such selection. It has some choice: it can use any or all of the different types of licenses the agreement covers (for example, it can elect to implement H.264 in products sold to consumers and/or OEMs). But it can't perform an end run around the grant-back obligation because the optional identification (with a view to the Enterprise License royalty cap) only relates to "its Affiliates which are licensees under the AVC Patent Portfolio License". Motorola doesn't have a license, so it wasn't and isn't eligible for notification. If it had taken or now took a license, then it would have its own grant-back obligation under an MPEG LA-Motorola agreement.

In light of that, Microsoft points out that it wouldn't make sense for the grant-back clause to include a licensee's "Affiliates" if this related to companies that have a grant-back obligation under their own agreements with MPEG LA.

Microsoft also points out that "Google has taken three separate and distinct types of license offered by MPEG LA": the AVC Products license, the OEM Licensee license, and the Enterprise License (a bundle of all distribution/broadcasting licenses). YouTube needs a broadcasting license, while Google's distribution of the Android and Chrome operating systems (directly and, especially, through OEM partners) falls under the AVC Products and OEM Licensee terms. Microsoft argues that "the Enterprise License and its royalty terms are irrelevant to the issues [in this FRAND rate-setting case]" because "Microsoft is not a video content provider and therefore does not need, and has not taken, an Enterprise License from MPEG LA". Therefore, "Microsoft is not seeking any grant-back rights for such a license in this litigation". It just wants a grant-back with respect to the tech product licenses. Simply put, what Google pays for the distribution of Android and Chrome OS (Google in this context including its subsidiaries such as Motorola Mobility without any need for notification) is what Microsoft is willing to pay, proportionally based on the number of H.264 SEPs, for Google's (Motorola's) H.264 SEPs in connection with Windows and the XBox. This is unrelated to video streaming.

While the H.264 FRAND ball is now in the court's court, Microsoft and Google (the parent company as well as its Motorola Mobility subsidiary) are going to meet at the Munich I Regional Court on Thursday for a (non-standard-essential) patent trial relating to Google Maps.

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+.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Google's WebM (VP8) allegedly infringes the rights of at least 12 patent holders

Google's attempts to promote "royalty-free" open source technologies just can't succeed in a world in which software is patentable -- a circumstance that Google increasingly realizes and complains about. No one can safely claim anymore at this stage that Android is a "free" mobile operating system without making a fool of himself, given that approximately 50 patent infringement lawsuits surround Android, an initial determination by an ITC judge just found Android to infringe two Apple patents (with many more still being asserted in other lawsuits), and ever more Android device makers recognize a need to take royalty-bearing licenses from Microsoft and other patent holders. Now Google's WebM codec project is apparently bound for a similar free-in-name-only fate as Android.

As a result, WebM seems unfit for adoption as part of a W3C standard, given the W3C's strict policy that its standards must be either patent-free or at least royalty-free.

In February I reported on MPEG LA's call for submissions of patents deemed essential to the VP8 video codec, a key element of Google's WebM initiative. I had already expressed doubts about Google's claims of WebM/VP8 being unencumbered by third-party patents shortly after WebM was announced more than a year ago. The commercial issue here is that Google's claims of WebM being "royalty-free" would be reduced to absurdity the moment that any patent holder rightfully starts to collect royalties on it.

I just became aware of a new streamingmedia.com interview with MPEG LA. MPEG LA serves as a one-stop shop for licenses to AVC/H.264 and other multimedia codecs; streamingmedia.com is the website of Streaming Media magazine. In that interview, MPEG LA stated affirmatively that there have been submissions relating to the February call, and disclosed, at a high level, a preliminary result of the vetting process that commenced subsequently to the submissions period:

Thus far, 12 parties have been found to have patents essential to the VP8 standard.

12 parties -- that's really a high number, and it could even increase in the future.

For now, MPEG LA doesn't want to name those companies. Chances are that there is an overlap between those 12 companies and the ones that contributed to MPEG LA's AVC/H.264 pool. I sent MPEG LA an email to inquire about this, but the only answer I received was that "confidentiality precludes [MPEG LA] from disclosing the identity of the owners".

Whatever the names of those companies may be, it's obvious that they wouldn't have submitted patents to MPEG LA if they weren't interested -- at least in principle and always subject to agreement on the particular terms -- in collecting royalties on WebM. While the Moving Picture Expert Group (MPEG) is a standardization body that also has plans for a (truly) royalty-free codec, MPEG LA is independent from MPEG and in the licensing business. Even MPEG LA offers freebies. For example, it doesn't charge for the use of AVC/H.264 for free Internet video. But that's fundamentally different from declaring a codec royalty-free without any field-of-use restrictions.

The WebM Community Cross-License intiative can't solve WebM's patent problem

I'm sure that none of those 12 companies is a member of the Google-led WebM Community Cross-License initiative. The companies behind the WebM CCL are Google partners who have committed not to assert their patents (should they have any that read on WebM) against that codec. The significance of that initiative was overestimated by some people. It's just a non-aggression pact. Those companies didn't commit to launch retaliatory strikes against patent holders who may bring assertions against WebM. Also, there's a notable absence: Motorola is a top three Android device maker and should be an obvious partner for Google but apparently reserves the right to sue WebM adopters. A Motorola subsidiary named General Instrument Corporation is suing other companies, such as Microsoft and TiVo, over various codec-related patents, including (but not necessarily limited to) U.S. Patent No. 5,949,948, 6,356,708, 7,310,374, 7,310,375, 7,310,376, and 7,529,465. If Google can't even get all of its Android partners on board, that shows how incomplete the WebM cross-licensing group is.

The process

Following a process similar to that of standard-setting organizations (though MPEG LA is a licensing body for patents related to standards previously set by others), MPEG LA evaluates the essentiality of such submissions, meaning that a group of patent experts has to make a determination on whether a patent declared essential really reads on a standard. Without such a vetting process, companies could declare patents essential only for the purpose of participating in a royalty-sharing agreement with those whose patents truly are essential to the relevant standard. In the Streaming Media interview, MPEG LA stresses that "[i]ndependent evaluations of patent essentiality are key to MPEG LA's pool licensing programs and have stood up well throughout its history".

While some have claimed all along that MPEG LA has a vested interest in WebM being deemed non-free (since that result takes a major differentiator away from AVC/H.264's competitor), it's important to consider that MPEG LA's credibility depends upon (among other factors) the quality of its essentiality assessment process. If the experts who perform this analysis overshoot and declare too many patents essential to a standard, MPEG LA's patent contributors will be concerned that their share in any royalty-sharing agreements related to a pool gets diluted. And if MPEG LA failed to recognize the essentiality of patents, licensees would be less likely to believe that MPEG LA is really a one-stop shop. MPEG LA doesn't give guarantees that it's a one-stop shop, but it certainly strives to form pools that minimize the risk of licensees facing infringement assertions related to patents that are not part of the given pool.

In the Streaming Media interview, MPEG LA said that there was a meeting "with VP8 essential patent holders in late June to facilitate a discussion among them whether and on what terms they may want to create a VP8 patent pool license", and that these efforts are continuing.

MPEG LA now has a clearer idea as to which patents would have to form part of the pool. Such a pool "typically remains open for the inclusion of additional essential patents", but there's now a group of a dozen patent holders who may or may not reach an agreement on the formation of a pool. To form a pool, they will have to determine what royalty rates MPEG LA would charge for the pool and how it would redistribute income from the pool to its contributors.

MPEG LA is a licensing -- not litigation -- entity

MPEG LA always stresses (in that interview and on previous occasions) that MPEG LA itself does not file patent infringement lawsuits on behalf of contributors to its pools. I guess that if a company uses a standard and refuses to take a license, MPEG LA will inform its contributors (maybe proactively, maybe just on request). But contributors don't appear to have an obligation to do so if an alleged infringer is unwilling to pay.

Obviously, if none of the actual patent holders ever sued an infringer, licensees might prefer to "roll the dice on litigation". If the 12 patent holders identified by MPEG LA so far (or possibly even more further down the road) join the pool, and if there are deep-pocketed organizations behind them that can afford to bring infringement actions, it's very likely that most of the alleged infringers would give serious consideration to a license deal.

Given that MPEG LA itself doesn't sue, it would be incorrect to claim that MPEG LA created this pool in order to litigate against Google or its partners. MPEG LA can legitimately claim that it offers a way for infringers to more efficiently address a problem that someone else (in this case, Google) may have created. It's not even that MPEG LA was needed to make the holders of those patents aware of the fact that their patents read on WebM. Those patent holders participate in a call for submissions, so it was the patent holders' own assumption that WebM infringes their rights.

It's important to understand that MPEG LA is just an aggregator. Even if MPEG LA didn't exist, those patents would still be around, and an adopter of WebM seeking to obtain a license to all of the declared-essential patents would have to talk to a dozen different companies.

A few months ago, the Wall Street Journal reported that the United States Department of Justice is looking into MPEG LA's activities. In that article, MPEG LA's CEO is quoted as describing his entity as "a convenience store" for licensing patents.

MPEG LA neither made software patentable nor filed the applications that resulted in the patents that were apparently submitted in response to its call for submissions. I believe a world without software patents would be a better place, and in such a world, MPEG LA wouldn't exist either. But in a world in which software is (unfortunately) patentable, MPEG LA can help to address the problem in one particular field.

I wouldn't even hold it against MPEG LA that Google itself didn't ask for the creation of a VP8 pool. The possible creation of an MPEG LA VP8 patent pool wouldn't be on the agenda if there weren't patents that read on VP8 according to their holders as well as MPEG LA's essentiality evaluators. This would also be a non-issue if Google had truly cleared all of the relevant rights instead of creating what could in the worst case become another lawsuit magnet like Android, potentially exposing a variety of companies and also non-profits like the Mozilla Foundation to the risk of litigation. I'm against the root cause of the overall problem, but MPEG LA is a partial solution in one field.

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.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Share|

Royalty-free MPEG video coding standard ups the ante for Google's WebM/VP8

Amid all the debate about whether Google's WebM video codec VP8 is truly patent-unencumbered and a valid alternative to the market-leading video codec (AVC/H.264), a very interesting project nearly went unnoticed: the development of a royalty-free MPEG video coding standard.

At a recent meeting of the Motion Picture Experts Group (MPEG), the decision was taken to issue a call for proposals (which will formally happen in March 2011) on video compression technology "that is expected to lead to a standard falling under ISO/IEC 'Type-1' licensing, i.e. intended to be 'royalty free'." The announcement also states the following:

"In recognition of the growing importance that the Internet plays in the generation and consumption of video content, MPEG intends to develop a new video compression standard in line with the expected usage models of the Internet. The new standard is intended to achieve substantially better compression performance than that offered by MPEG-2 and possibly comparable to that offered by the AVC Baseline Profile."

It's important to note the difference between MPEG, the standardization body, and MPEG LA, the licensing agency. MPEG defines standards, and those can be royalty-free or royalty-bearing, depending on what the owners of the essential patents agree upon in each case. MPEG LA manages patent pools related to royalty-bearing standards, serving right holders and licensees as a one-stop solution by way of aggregation.

Rob Glidden, a former Sun executive and digital TV expert, made me aware of MPEG's royalty-free codec plans via Twitter (thanks for that!). I found several interesting posts on his blog, such as this recent one arguing for a royalty-free video codec standard. Just like me, Rob appears to be somewhat skeptical of claims that WebM/VP8 is truly patent-free.

Three formidable hurdles for WebM/VP8

Google and its WebM allies (primarily Mozilla, Opera and the Free Software Foundation) wanted to pose a direct challenge to AVC/H.264, the incumbent and ubiquitous video codec. Their strategy was to seek adoption of WebM/VP8 as an Internet standard, claiming that it was the only format of its kind to be available on royalty-free, open-source-like terms. They thought they were already challenging the champion in the final match, but not so fast: MPEG LA may soon start collect royalties from VP8 adopters, and if MPEG's own royalty-free codec initiative succeeds, VP8 will have to beat it -- or be forgotten.

Using that tournament analogy again, VP8 is only in the quarter finals. MPEG LA's formation of a VP8 patent pool and MPEG's call for proposals on a royalty-free video codec are such important developments that the results they produce will have to be awaited before standardization on VP8 will even be seriously considered by the W3C or IETF. There is no rush because everyone can demonstrably watch Internet videos for the time being even though HTML 5's video tag is currently codec-agnostic.

I have drawn up a simple graphic that shows the three hurdles VP8 is facing:

If MPEG LA's patent pool for VP8 disproves claims that Google's codec is truly free of charge, it loses its intended unique characteristic and is out of the game. At that stage, even the Free Software Foundation may feel forced to withdraw its support. It wouldn't be possible for the FSF to advocate a royalty-bearing format.

Should VP8 overcome that hurdle, it would play a semifinal match against MPEG's royalty-free standard. I'll explain further below why this one is also going to be a tough challenge.

Only if VP8 survives that round as well, it will get to challenge AVC/H.264. To dethrone the reasonably-priced market leader, VP8 would have to deliver high quality, not just royalty-free availability.

The odds are long against WebM/VP8

Surmounting three significant hurdles in a row -- since failure at any of them means "game over" -- is really difficult. Even if one assumed a 60% chance in each case, the aggregate likelihood of success would just be slightly above 20%. In my opinion, VP8's chances are less than that. Let me explain my view on each of the three challenges.

1. MPEG LA patent pool for VP8

MPEG LA requests submissions of patents essential to VP8 until March 18, 2011. That's less than six weeks from the announcement. I presume MPEG LA already knows of a variety of relevant patents, and has discussed this with its membership. The first time MPEG LA CEO Larry Horn mentioned the possibility of such a pool was in an interview with AllThingsD in May 2010.

After the submissions deadline, MPEG LA's patent evaluators will make a determination as to which patents are truly essential to VP8. Obviously, whatever they conclude isn't the same as a judicial declaration of infringement. But MPEG LA is trusted by industry, and the moment it specifies the patents it believes read on VP8, Google and its allies will either have to invalidate those patents (which is unlikely to work for a reasonably long list) or prove that there isn't an infringement.

There could also be patents which VP8 infringes but which aren't submitted to MPEG LA in that process. I didn't mention that before, just to keep things simple at the time. When assessing the likelihood of VP8 truly being patent-unencumbered, this additional risk must, however, be taken into account.

2. Royalty-free MPEG video coding standard

It's plausible that MPEG can pull off a royalty-free coding standard. It will be able to draw from three types of patents for that purpose:

• patents that have already expired

• patents approaching expiration (which limits their commercial value)

• patents whose owners are willing to forgo short-term revenue opportunities in favor of longer-term objectives, or who consider it a prudent strategy to support a royalty-free standard for web purposes and monetize related (but other) patents in other fields of use such as digital cameras and digital television

Codec development was already reasonably advanced in the early to mid 1990s, and US patents of that kind have a maximum term of validity of 20 years. Therefore, the first two groups of patents may already be sufficient to produce a codec at a level with, or even significantly superior over, VP8.

In order to compete successfully with a royalty-free MPEG standard, VP8 would have to be clearly better just to be considered. That's because MPEG has the backing of many vendors and offers an open and inclusive process. Under MPEG's stewardship, such a standard can continue to be developed (as new patents expire, approach expiration or are made available by their owners for other reasons) in a balanced, consensus-based way, while WebM is dependent upon Google, a company with specific strategic interests.

The argument that the availability of a WebM reference implementation on open source terms ensures vendor independence isn't sufficent. Theoretically, it's true that anyone could take it in a different direction. Practically, such forking would result in fragmentation, and it's doubtful that anyone would ever put significant resources behind such an effort.

3. Incumbent AVC/H.264 standard

Should VP8 really overcome the challenges previously described, it would still have to garner broadbased industry support. It wouldn't be in the interest of standardization bodies like the W3C and the IETF to formally declare a "standard" that fails to be adopted widely. Such failures can undermine the credibility and diminish the influence of those organizations.

AVC/H.264 isn't royalty-free, but its terms appear to be acceptable to a majority of industry players. In order to take any sizable part this incumbent's market share away, VP8 would have to be at least technically comparable. Degradations wouldn't be acceptable to consumers. However, my own impression and that of most of the opinions I read is that there's still a noticeable gap.

One of the challenges for VP8 is the limited number of hardware products supporting it. For hardware companies it's particularly important to be careful about patent-related risks. A software company may be able to just change its code. Hardware companies would be hit hard by injunctions or ITC import bans. VP8 faces a chicken-and-egg problem with the hardware industry, and the two new initiatives I described before will also have an effect in that area.

Google and its WebM allies won't give up. They are determined to turn this into a new edition of the "videotape format war" of the late 1970s and early 1980s. Looking at the drawing that shows the three hurdles and considering the facts, I have my doubts that they will get very far. However, even if the odds appear to be long against VP8, it may have a positive effect on the willingness of patent holders to support MPEG's royalty-free codec standard.

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:

MPEG LA puts Google's WebM video format VP8 under patent scrutiny

A call for submissions issued by MPEG LA, the leading licensing entity for multimedia codecs, may provide a definitive negative answer to Google's claims that the VP8 video codec, which is part of its WebM set of codecs, is free from patent encumbrances.

When Google recently announced that its Chrome browser would drop the ubiquitous H.264 video format in connection with the HTML 5 'video' tag, many experts doubted that such a decision would really tip the scales in favor of Google's WebM/VP8 technology. For example, book author and ZDNet blogger Ed Bott asked whether Google was avoiding a trap or walking into one, and quoted a patent attorney who warned that "[a] codec is like a mechanical device with hundreds of parts. Any one or more could be the subject of a patent."

Google and its allies (particularly Mozilla and Opera) claimed that WebM was royalty-free and, therefore, uniquely suited for use in Free and Open Source Software.

When WebM was announced last spring, I commented somewhat skeptically on the original announcement and on a subsequent license change, and I took a closer look at MPEG LA's license terms for AVC/H.264, the market-leading standard standard supported by an impressive list of major vendors.

"Patent-free" is easier said than done

In my opinion, claiming that a relatively late entrant can be "patent-unencumbered" in a field that's been a patent thicket for many years is just about as realistic as saying that you can walk safely through a minefield. It may be possible, theoretically speaking, but who would actually take the risk without a need?

Of course, anyone can make claims of that kind until there's evidence to the contrary. Some will believe such claims and some others won't even seriously think so but will, for strategic reasons, deny the threat in their public statements as long as they can.

Let me make this very clear: since I'm against software patents, I'd be happy to see a patent-free video codec put competitive pressure on dozens of major patent holders. But I make a clear distinction between what I'd like to see happen and what I consider realistic.

Deadline for submissions of essential patents: March 18

We may now be only a few months away from a definitive and potentially negative answer. Yesterday, MPEG LA published a call for patents essential to the VP8 video codec.

Such a call is the beginning of a formal process that will, if a number of such patents are submitted, most likely result in the creation of a patent pool. MPEG LA's patent pools aren't free of charge. They may allow free-of-charge use of their standards but only for particular purposes. In other words, "free" will be subject to field-of-use restrictions, and everything else will be charged for.

MPEG LA asks for initial submissions to be made by March 18, 2011. The focus will be on actually issued patents, but patent applications that are still being examined may also be submitted if considered essential.

MPEG LA's patent evaluators (legal and technical experts) will analyze all of the submitted patents to determine their "essentiality". In any standardization process, an "essential" patent is understood to be required for an implementation of the standard in question, as opposed to patents that may be somewhat related to it but not absolutely necessary.

A standardization or licensing entity can't be sure that all essential patents were actually submitted. Some patent holders may not be interested in participating. But the goal is to identify as many essential patents as possible.

Commercial negotiations

After the essential patents for a proposed pool have been identified, there is always a negotiation between the contributing patent holders. They have to agree on the terms on which to make the entire pool available to licensees and on their internal revenue-sharing arrangements (which should be reflective of the value of the contributions made by the parties).

I don't know whether MPEG LA will publish the list of patents identified after technical evaluation of all submissions, or only after successful conclusion of commercial negotiations. The latter seems more likely to me. One way or the other, we may only be months away from seeing a list of patents found to read on WebM/VP8.

Should MPEG LA form a VP8-related pool, then they will approach WebM/VP8 adopters and ask for royalty payments. Since MPEG LA is an entity that means business when it comes to patent licensing, their demands for royalties won't just be ignored by commercial players interested in legal certainty. As a result, Google and its WebM allies would have to show that none of the patents identified are valid and infringed. For each patent, the proponents of WebM could either seek invalidation or try to prove that there isn't an infringement issue. Depending on the size and quality of the pool, that can be a formidable or practically insurmountable challenge. So much for free in that scenario...

When it comes to patents, the odds are long against Google

I want to be frank. Despite my dislike for software patents, if I had to bet money, I would bet it on MPEG LA, not on Google.

MPEG LA has a pretty strong track record in codec patent licensing. One may or may not like their business model, but they clearly know what they're doing. By contrast, Google's flagship open source project, Android, has turned out to be a patent suit magnet.

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:

MPEG LA's AVC/H.264 licensing terms: further analysis

There's still a lot of controversy over the video codec question, in connection with HTML 5 and generally. I have commented on this on several occasions, most recently on Google's announcement of the "free" WebM (VP8) codec (this posting) and the subsequent modification of the WebM license.

I have received some useful clarifications from MPEG LA concerning its AVC/H.264 licensing terms and would like to share that information and comment on it in this posting. Prior to doing so, let me explain my approach to the controversial subject.

Separating politics and economics

As a long-standing opponent of software patents, I'm all for patent-free formats but I want to understand from a pragmatic point of view what the risks and opportunities are for the average software developer (FOSS as well as non-FOSS) when adopting one video codec or another:

• For WebM/VP8, there's a vague assurance by Google that its own patents, which are licensed on a royalty-free basis, are all you need. But Google doesn't publish a detailed analysis, nor do the license terms include indemnification. So it's pretty much a "trust us" story. I've seen opinions that agree with Google's view, and others who disagree. If Google offered indemnification, that would change the situation, but they don't.


• For MPEG LA's AVC/H.264 codec, patent royalties are due unless one does something that falls under an exception (such as certain kinds of non-commercial use) or the number of units sold is below a certain threshold. The patent pool is huge and from many companies, mostly large ones. Since AVC/H.264 is already extremely widespread, third parties owning patents that read on it would likely have enforced them already, which makes it reasonably likely that MPEG LA controls all of the essential patents.

From a philosophical point of view, I wish there weren't any software patents. In that case, there wouldn't (have to) be an MPEG LA. But I try hard (and encourage everyone else to try equally hard) to separate my political/philosophical preference from practical/economic considerations under the circumstances we have to live with.

In that respect, the licensing terms for AVC/H.264 play a key role. If they are reasonable (under the circumstances), then one might accept them as a practical, totally unemotional choice. There are patent holders who don't make licenses to their patents available on any terms, or not on any reasonable terms. Those strategic patent holders with an exclusionary strategy are the biggest problem as I recently explained.

Further analysis suggests the AVC/H.264 royalty cap works the way it should

MPEG LA has a royalty cap so that companies selling high-volume products know beforehand the maximum amount of royalties they'll have to pay to MPEG LA in a given year. The current $5 million cap really isn't much for a big player possibly generating many billions of annual revenues with products that include an AVC/H.264 encoder and/or decoder.

More importantly, this makes it economically possible for entities like Mozilla and Opera to give away huge numbers of web browsers to end users on a free-of-charge basis (note that Mozilla, unlike Opera, offers truly free software -- free as in free speech, not just free beer).

Two weeks ago I attended an event at Google's Brussels office: a roundtable on open source business models. Its roughly 30 participants were from a diversity of companies as well as the European Commission. When the codec question came up, I mentioned the AVC/H.264 royalty cap, and someone at the event basically claimed that I shouldn't believe what MPEG LA's website says about it because MPEG LA could charge different fees for different types of products. According to the house rules of the event, I can quote what was said but without naming the person who made the statement. So I asked that person for permission to name the source when quoting that particular statement on my blog, but the person prefers not to be named and clarified that "for most companies, they probably don't see much more than one overall fee."

That was basically a retraction of the original claim. Nevertheless I wanted to know from MPEG LA, the licensing firm that manages the AVC/H.264 pool, whether the $5M per-customer per-year royalty cap has anything to do with product categories such as smartphones versus other devices.

Here's the answer I got from an MPEG LA spokesperson:

The royalty cap for AVC Product in our AVC License includes the combined sales of all of a Licensee’s AVC Product (encoders, decoders or combination device equals one unit) in that calendar year. Therefore, a Licensee would not pay more than $5 million in royalties for any combination or amount of AVC Product it sells in 2010, regardless of what type of device it is. It should be noted that caps and royalties are subject to possible increases to be determined for the renewal term 2011-2015.

I'd like to add that MPEG LA made a commitment to keep those increases within a specified limit (see the last item on this FAQ page).

That answer suggested the cap works the way it should. I still wanted to know more about the cap: Are there any customers who for whatever reason may have to pay more than $5M in a given year? And if so, why?

And this is what MPEG LA told me:

Caps apply to each of AVC Product, OEM AVC Product distributed through a PC operating system incorporated in the end PC product of another party, and AVC Video. While rare, there are instances of a Licensee and its affiliates being subject to caps for both AVC Product as well as AVC Video in a given year. There is also instance of a Licensee being subject to caps for both AVC Product and OEM AVC Product for PC operating system product.

To avoid a misunderstanding, that doesn't mean there's a cap for each product (in which case a company with 20 products could pay up to $100 million). The term "AVC Product" means a license type. MPEG LA has three types of licenses that are relevant here and I'm quoting from the license agreement, which was provided to me on a confidential basis with permission to quote these definitions:

• AVC Product(s): "any product or thing in whatever form which constitutes or contains one or more fully functioning AVC Decoder(s), AVC Encoder(s) or AVC Codec(s). AVC Product(s) shall not include OEM AVC Products."


• OEM AVC Product(s): "AVC Product(s) sold to an OEM Licensee Customer."


• AVC Video: "video encoded in compliance with the AVC Standard"

Let me put it in more colloquial terms: If you obtain an "AVC Product(s)" license, it covers the products you build that include an H.264 codec in some form. The "OEM AVC Product license" category is a special case for companies selling a PC operating system to OEM customers (computer manufacturers who then bundle the product). And "AVC Video" is a license for those offering H.264 video content. All licenses relate to the same patent pool but to different business models.

In terms of the cap, there can be cases where companies need more than one of those three licenses. Most companies won't, but those who do may pay, for an example, twice the cap. But that doesn't mean that their costs are totally unlimited. They simply have more than one license fee account and each account has a cap.

I asked a third question about the cap: Is it true that if companies infringe your patents and have to pay back-royalties, the cap doesn't apply but they may have to pay more?

MPEG LA answered:

That is not correct. The caps applicable to any given year in which a new Licensee owes Back Royalties (with applicable interest, if any) still applies. For example, a new Licensee owes AVC Product royalties for 2007 would be subject to the cap of $4,250,000 for that year.

All things considered, the royalty cap appears reliable to me and I don't have the impression that MPEG LA tries to mislead anyone about it.

Patent enforcement: are licensees protected against infringing competitors?

I recently talked to someone who said that MPEG LA goes after some but not necessarily all patent infringers, which means that if you're a company in a given market that is forced to pay royalties and you have competitors that are not, you may be at a strategic disadvantage with others undercutting your prices. Therefore, my fourth and final question to MPEG LA was: Is it true that if a company pays royalties to you and faces significant competition from an infringer of your patents in its core market(s), the legitimate licensee has no contractual basis to require MPEG LA to go after the infringer?

MPEG LA answered:

It is correct that there is no requirement in the AVC License for litigation to be brought against an unlicensed, infringing company or organization. We actively pursue unlicensed companies or organizations that may be using the AVC standard to offer them the license, but, any litigation against a non-Licensee would brought by patent rights holders, not by MPEG LA.

The first part confirms that a licensee doesn't have a guarantee that his infringing competitors will be pursued. The second part is a legally technical thing: MPEG LA collects royalties on behalf of the actual patent holders but can't assert those patents against infringers in court. That's why litigation would have to be brought by the actual patent holders. I understand, but nevertheless I think it would be good for MPEG LA's licensees to have a legal commitment from the patent holders that they will either collect royalties from all significant players in a given market or from none of them.

One might assume that MPEG LA's obvious desire to maximize its royalty income would practically guarantee that all significant companies infringing its patents would be pursued. But the worldwide market is huge and if you have a country that's only medium-sized, then MPEG LA may decide for efficiency reasons to collect royalties only from the market leader. That one would face a risk of litigation if refusing to pay, but if that licensee's competitors don't pay because MPEG LA doesn't go after them for the time being, then that's an awkward situation for the licensee.

So while I'm (for now) satisfied with MPEG LA's answers on the royalty cap, I would like them to give some real protection to licensees in the scenario I just described, for the sake of undistorted competition. Even though litigation would be the prerogative of the actual patent holders, the license agreement could alternatively waive the obligation to pay if a licensee faces significant competition from infringers who don't pay.

I will keep following the codec debate and continue to look at licensing terms and practices.

If you'd like to be updated on patent issues affecting free software and open source, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents.