Tuesday, October 19, 2010

The European Interoperability Framework (EIF): FOSS-compatible FRAND licensing could be the solution

ZDNet UK couldn't have chosen a more appropriate context than its "Communication Breakdown" blog to report on the state of the European Interoperability Framework debate. Looking at that article, other media reports and some Twitter messages, there are strong indications that this process has hit a snag, or even an impasse.

What I mostly see in the public debate is a blame game: some free software advocates and anti-patent activists teaming up with open hypocrites attacking the Business Software Alliance (BSA) over its support of fair, reasonable and non-discriminatory (FRAND) patent licenses.

I have my own views and I don't fully support the demands of either side so far. On both sides I see a number of companies and entities (such as the BSA) against which I had to fight when I fought against a proposed EU software patent law. Meanwhile, those on the "open standards" side who also opposed that bill (FSFE, FFII etc.) now seem to care more about their alliances with companies like IBM and Oracle in the "open standards" context, so they turn a blind eye to those organizations' wrongdoings in connection with software patents and interoperability. Worse than that, they claim that FRAND-based standards are inherently incompatible with free and open source software, which is incorrect.

The EU wants a new EIF and I think it would be desirable to work things out, but in order to get there, the debate has to move on from overgeneralizations (such as the one I just mentioned about FRAND and FOSS) to a more specific description of the issues and of possible solutions. As long as some say that only "royalty-free", or more generally, "restriction-free" patent licenses allow "open standards", this won't work, I'm afraid.

Today I had a Twitter conversation with two FOSS lawyers -- Carlo Piana (FSFE Counsel) and Andrew Katz (FSFE Fellow) -- and a FOSS-specialized journalist -- Glyn Moody -- over the EIF, and the discussion demonstrated both the obstacles and a possible path to a solution.

In that Twittersation, three things became absolutely clear to me:

  1. Only certain -- but not all -- FRAND patent license terms are incompatible with FOSS licenses (the unpragmatic and irrelevant GPLv3 aside).

  2. It's wrong to focus on "royalties" as the sole knock-out criterion because royalty payments can be made to work for FOSS, and field-of-use and other restrictions are also part of the equation.

  3. The process could still be concluded with an outcome that would be satisfactory to all stakeholders -- public administrations (for whom the EIF is supposed to be a useful set of procurement guidelines), patent holders and open source -- if the "royalty-free standards" camp could describe exactly which FRAND license terms don't work for FOSS and propose realistic, targeted solutions that address those concerns within a FRAND framework.

    That would, of course, also require the BSA and other representatives of patent holders to agree to a FOSS-compatible FRAND solution. I don't know if the BSA ever claimed that all FRAND works for FOSS, but at any rate it should recognize that item #1 needs to be addressed.

I'm in the middle between the two camps because I'm pro-FOSS and against the patentability of software, but I also know that FRAND is a good concept in principle. Those who dismiss it (such as by claiming that it's not as simple as "royalty-free") appear to underrate it dramatically. Sometimes you don't know what you've got till it's gone. I've experienced situations in which a FRAND commitment would have prevented problems from arising, in IT (such as this EU antitrust case, which wouldn't be necessary if IBM promised to make whatever required intellectual property available on FRAND terms) and elsewhere (commercial exploitation and governance of professional sports).

In the following I would like to provide some more specific thoughts on the foregoing. Again, I'm not in the position to speak on anyone else's behalf: I just outline my own independent thinking. I don't have a complete compromise proposal at hand, but I have suggestions for how to get there:

Don't insist on "royalty-free"

Much of the EIFv2 debate has so far been centered around the question of whether royalty-free access to the relevant patents is an indispensable requirement for a standard to be considered open. That claim has been made by FSFE, ECIS, OpenForum Europe, and others. It runs counter to how the ICT sector has defined open standards for a long time. The main argument of the aforementioned lobby groups is that they say only royalty-free standards work for open source.

I wonder why anyone in the EU takes that claim seriously in any way. There's overwhelming evidence to the contrary. In another blog post I have already mentioned several examples of GPLv2-based patent license deals involving royalty payments. The first one of those was done and announced back in 2006. So if patent royalties don't work for GPLv2'd software at all, a host of companies would be in breach of the GPL now for distributing Linux. The Free Software Foundation has the Linux copyrights assigned and it is the creator and guardian of the GPL. It wouldn't refrain from enforcing the GPLv2 for such a protracted period of time. Therefore, it's clear that such deals are possible under GPLv2. They would even have been possible under the early drafts of GPLv3 as Richard Stallman admitted.

Red Hat is probably the most dishonest one of the proponents of that "royalty-free" dogma. It entered into at least one -- more likely two and possibly even more than two -- patent licensing deals under which it paid royalties to patent holders, still distributes the related software under GPLv2.

A European Red Hat lobbyist has repeatedly made the claim that his company doesn't implement certain patented standards such as MP3 because it "can't". If Red Hat can pay royalties to other patent holders, I can't see why it can't do a license deal with MPEG LA. In fact, a European competitor of Red Hat, Canonical, ships its Linux distribution called Ubuntu with MP3 and several other proprietary formats. It also became the first Linux company to license MPEG LA's AVS/H.264 video codec.

With baseless red herrings (red hats, in particular) of this kind, any political process can get derailed...

Don't narrow the issue down to royalties -- address all patent-related restrictions

I explained in the foregoing that royalties can actually be made to work for FOSS. It doesn't make sense to narrow the debate to an aspect that's actually a non-issue.

There are serious people who point out that the question of patents in connection with standards is a broader one. If those say "RF", they mean totally "restriction-free" access to patents, not just "royalty-free". Glyn Moody always refers to RF as restriction-free and I've seen him correct others on Twitter when they used the narrower term. Simon Phipps, a board member of the Open Source Initiative and formerly chief open source executive at Sun Microsystems, just tweeted the following sentence, which I consider accurate:

Explanations which hinge on royalties distract from the real issue, which is both fiscal & non-fiscal restrictions.

There are other terms than royalties that patent holders can impose. Sometimes those non-monetary ones are even more important. For instance, the Java patent license is royalty-free, but it's very restrictive. Oracle is suing Google over Java patents despite that license being available on a royalty-free basis to those complying with its various terms and conditions.

The only explanation I have for people narrowing the debate and some of the proposed language to "royalty-free" is that they hope to get some language into the final document that they will then try to interpret as "restriction-free" even though they only demand "royalty-free" at this stage. Those kinds of tactics are common in politics, but those pursuing them shouldn't be surprised that they meet stiff resistance: patent holders might as well stop paying renewal fees to the patent office if they're required to waive the entirety of their rights. It's not that the EIF would require them to do so, but I can understand if they're concerned about anything that someone might try to interpret that way later.

In the "royalty-free" context, I wonder how the FSFE can pursue those tactics even though it usually always emphasizes the importance of the "four freedoms" (some of which are unrelated to royalties) and the meaning of "free as in speech", not just "free as in beer". Richard Stallman himself would always stress this holistic approach. His European affiliate organization, which is just a lobby group but didn't play any role in the creation of the movement, is unfortunately less faithful to his principles.

Specify all restrictions and requirements that might be considered FRAND but don't work for FOSS

In my Twittersation with the three European FOSS advocates I mentioned, it became clear pretty quickly that they have concerns about certain terms of FRAND licenses that they say don't work for FOSS.

One example that was mentioned is that one can't impose patent royalties on downstream users. So if an open source developer publishes a piece of software under a FOSS license, the developer can't keep track of every copy of the software that gets distributed (since anyone downloading or in some other way receiving the software can pass it on to others, and so forth). That creates a problem for per-unit patent royalties.

But that doesn't mean that no royalties can be paid at all. Per-unit royalties are a question of responsibility, accountability, and collection/refinancing. For example, MPEG LA has an annual royalty cap for its AVS/H.264 video codec license. A company paying that fixed amount never has to worry about the number of downloads. Chances are that Red Hat also negotiated fixed amounts with the patent holders from which it has obtained licenses. And there must have been ways -- maybe other ways -- in which this problem was solved for all the other companies in the industry who pay patent royalties on GPLv2-based software.

Another question is field-of-use restrictions. It's understandable that patent holders will grant licenses on a restricted basis (I actually mentioned Oracle's Java license before). The MPEG LA license is also restricted, but that doesn't mean that software can't be published on open source terms. It's just that if anyone utilizes that software in fields of use for which the patents haven't been licensed, the patent holders may approach the non-compliant user. Maybe they'll work this out with an additional payment (that's what would happen in MPEG LA's case). It's just important to make sure that someone publishing software under a FOSS license won't be held responsible for actions of others that are beyond control. I'm sure this has also already been resolved in some patent license agreements involving open source (since there are so many of them in place).

Two quotes from Andrew Katz in today's Twittersation show that FOSS doesn't have a problem with all FRAND terms -- only with some. Firstly he said:

my personal view is that RAND is against all freedom in spirit, and against some licences in word (e.g. GPL)

RAND is a shorter version for FRAND; in EU competition law the preferred term is FRAND, but if someone in the US says RAND, this means the same. Andrew correctly made the distinction between FOSS philosophy and the legal meaning of FOSS licenses. I also distinguish all the time between my dislike for software patents and the fact that solutions must be found to deal with them.

When I pointed Andrew to the fact that Richard Stallman admitted even the early drafts of GPLv3 wouldn't have blocked the patent license deal between Microsoft and Novell, he said something very important:

it depends on what you mean by Rand licences. It is possible to draft a compatible licence. Most aren't

That's the whole point I'm trying to make here: FRAND can be made to work with FOSS, including the GPLv2. A FOSS-specialized lawyer and FSFE Fellow -- who also blogs about open source law for ComputerworldUK -- just confirmed it.

So the best next step is for the FSFE and others to identify which kinds of FRAND terms don't work, and to make constructive proposals for criteria that make FRAND licenses compatible with FOSS. By "constructive" I obviously don't mean "royalty-free", as I explained. But without a doubt, FRAND license agreements that prohibit the use of patents in FOSS, explicitly or as a consequence, wouldn't work. I don't think the EIF would necessarily go into too much detail, but it could provide some guidance for FOSS-compatible FRAND licenses, which would be a pragmatic solution and obvious compromise.

Don't claim discrimination against FOSS when closed-source proprietary software is also affected

In connection with all of this, it's also important not to claim "discrimination" where there isn't any.

A few weeks ago Glyn Moody wrote that FRAND licensing isn't non-discriminatory because, as he said, it discriminates against FOSS. This was also part of today's Twittersation. After the point on per-unit royalties was made, I asked him why he then (instead of continuing to oppose FRAND as a whole) doesn't just demand that there must be a possibility for "FRAND lump sums". Glyn replied:

because that would allow companies to demand huge lump sums that were impossible to pay, citing millions of downstream users

He immediately added:

it would also be impossible for projects that had no company behind them, leading to a two-class system

This is a perfect example of calling something "discriminatory" that actually isn't. Glyn is right that licensing patents is much more difficult for smaller companies, let alone projects without companies behind them, than for larger ones. But that isn't open-source-specific. A small company writing closed-source proprietary software and distributing it free of charge (as part of a "free" or "freemium" business strategy) would face the same issue. Discrimination as a term only applies if a group is uniquely disadvantaged. The advantages of large, deep-pocket companies over smaller, sparsely-funded ones, or of any kind of company over non-company software development projects, are manifold. Patent licensing is just one area in which large organizations benefit from size, and again, you don't have to be open source to face those challenges. It's just that you're small, or that you're not a company.

Don't betray the EU's FOSS-friendly attitude

EU politicians have a concern for FOSS, and that's a good thing. What I'm afraid of is that some may try to capitalize on the EU's love of open source for the purpose of a short-term tactical advantage in a political process and ultimately hurt the cause.

By the foregoing, I don't mean Glyn Moody; he's a journalist, not a lobbyist, not an attorney. I mean the ones who have now been trying for a long time to claim that FRAND is the antithesis of FOSS. Some of them have been doing this even though in their daily legal work they know very well how to solve those problems all the time -- and they know that almost all patent license deals involve royalties, including deals involving open source.

I'm more than sympathetic to their philosophy, but I don't think software freedom is an end that justifies any means. If we don't like something, we have to say we don't like it -- not that it isn't legally possible even though it is.

From my personal point of view, the best outcome would be a compromise where the FOSS world accepts a FOSS-compatible denomination of FRAND and where major patent holders recognize that FRAND must be made to work for FOSS to solve such problems as accounting for per-unit royalties.

Like I said further above: sometimes you don't know what you've got till it's gone.

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.