Patent Wars: The W3C Strikes Back

October 17, 2001

Kendall Grant Clark

In last week's XML Deviant column I summarized some of the important events and policy positions surrounding the W3C's Patent Policy Working Group's (PPWG) proposal, as well as the XML and free software community's reaction to the RAND provisions.

This week I will consider how the W3C is responding to the massive show of opposition to RAND -- at last count the W3C had received more than 2,000 public comments -- as well as provide an update of the XML development community's ongoing conversation about patents, intellectual property, rights, and the nature and history of the Web.

The W3C's Response

There are some reasons to think that the system, at least at this relatively early date, is working. The W3C has certainly been taking notice of the public opposition expressed to RAND policies. And, whether cynical or not, the PPWG has taken a significant and welcome step by altering its institutional makeup. Until very recently, the PPWG was composed almost exclusively of corporate vendor representatives. Now, as a result of public opposition, the W3C has, in its own words, "opened its patent policy process for continuing public dialog" (W3C Patent Policy Comment mailing list). That means, in part, inviting Eben Moglen, general counsel of the Free Software Foundation, and Bruce Perens, HP's open source point person, to join the PPWG as "invited experts". In fact, the PPWG has been meeting this week, and both Perens and Moglen have already joined those meetings in some degree.

The PPWG has further committed to producing a second public Last Call version of its "W3C Patent Policy Framework" policy document. This is also welcome news, since it assures the various interested communities that the participation and input of Moglen and Perens, as well as the substantive comments and criticisms contained in more than 2,000 public comments, will have some opportunity to matter to the final form of the W3C's patent policy.

While it may be too soon to celebrate the demise of a RAND patent policy, it seems that the basic soundness of the system -- which means, in this instance, the W3C's formal commitment to some degree of public responsiveness -- has been reaffirmed. As with all its recommendations, the W3C asked for public comment about a proposed policy, and the public responded massively, even if very near the last minute. So far the W3C has taken the appropriate minimal steps to take account of the public's response. The inclusion of Perens and Moglen does not insure any particular outcome, but it should reassure the free software and open source communities, if not others, that their interests and concerns will be represented in the W3C's ongoing debate about its patent policy.

Patents, Innovation, and Openness

As I wrote last week, one of the clear points of consensus in the XML development community's debate about RAND was the threat RAND posed to free software and the future of the Web. Given the widespread XML community perception of that threat, I thought it would be worthwhile to get some idea of how members of the W3C's technical staff viewed the importance of the kinds of (relatively) unfettered technical collaboration that open source fosters. "You will find, in general," Tim Berners-Lee, W3C Director, said, "very strong support for open source among the W3C staff. People tend to work at W3C because (among other things) they like the freedom of constraint of a particular corporate employer."

"The success of the Web," Dan Brickley said, "came in large part from the open source and free software community; open standards and open source are stronger when they work together. They naturally complement each other." Brickley works on RDF and Semantic Web infrastructure for the W3C, which is some of the most cutting-edge Web work anyone is doing these days. He added that the recent developments with regard to patent policy "also, albeit in a roundabout manner, got a lot of people thinking about the relationship between the free, open source software and the need for open standards on the Web. We heard from thousands of people who value W3C's specifications".

Dan Connolly, who works at the W3C with the Semantic Web and was for a long time the XML Activity Lead, had a similarly positive response about the relation between W3C and open source development. "I frankly can hardly afford," he said, "to pay attention to anything but open source development". He offered two reasons: first, "I can't really use anything else in my own work, because if I did, I couldn't guarantee that all the W3C membership could benefit from that work". Second, "the marketing hype around stuff that's not open source is so thick that I can't afford the time it takes to get the actual technical details". These are important points to remember during patent policy debates. In this case, at least, there's some indication that open source fuels innovation by removing some of the friction that would otherwise exist.

According to Connolly, patent encumbrances make development more difficult. For example, Connolly "wanted to invite some developers to come and chat with us, but there were patent issues that just weren't worth the time and hassle to deal with". "Patents are certainly an impediment to 'technical innovation' in the sense of new ubiquitous technology," he said, "they may have their place in a technology that's deployable by one particular company alone".

Both Dan Brickley and Dan Connolly expressed satisfaction that Bruce Perens and Eben Moglen are willing and able to join the PPWG as invited experts.

Berners-Lee said that the most innovative periods of the Web's history were composed of open source development. "The Web's critical growth period involved only what we now call 'open source' software," Berners-Lee said. "In fact one of the challenges early on was to get industry interested," he added, "because there was a community which would not take it seriously unless they could get software from a company." Berners-Lee made a distinction, however, between parts of the Web that are relatively stable and parts that are still undergoing rapid, innovative development. "The Semantic Web is again similar: academic and independent enthusiasm, open source sofware, and industry is reluctant to commit until they understand it more."

Berners-Lee's views about the impediment to innovation that patents impose are well-known. In his Weaving the Web, he writes, "I mention patents in passing, but in fact they are a great stumbling block for Web development. Developers are stalling their efforts in a given direction when they hear rumors that some company may have a patent that may involve the technology" (Weaving the Web, Chapter 13).

XML Community Reaction Continues

So far members of the XML-DEV mailing list have taken what one might call a hopeful, wait-and-see approach to the W3C's ongoing patent policy activity. There seems to be a muted appreciation for the inclusion of Moglen and Perens, and an underlying recognition that they alone are unlikely to turn the tide against RAND.

Simon St. Laurent said, "It's a start, and I'll be curious to see how it goes. This seems to be the first serious threat to the 'software developing' public's perception of the W3C's mantle of legitimacy, so I suspect there's a lot more at stake here in the long run than just patent licensing."

Don Park's evaluationof the W3C's response was even more mixed. "I applaud the PPWG's decision to open up its activities by inviting open source experts and opening a home page for PPWG." But, Park added,

I am disappointed with the lack of information on [Moglen and Perens] ability to communicate with people they represent, the public. Under normal W3C policy, invited experts are not allowed to share WG's confidential information. Given that all information generated during WG activity are assumed to be confidential unless specifically marked as public information, I have to assume that the invited experts will be out of communication with the public as soon as they join the WG unless they are willing to face lawsuits.

...I am also disappointed with PPWG's decision to not open up access to internal PPWG mailing lists. Please open them up.

Park thus raised the possibility that, having been invited to join the PPWG's deliberations, Moglen and Perens will be hampered in their ability to talk freely about their work. Such an impediment would seem to hamper Moglen more than Perens who, despite his credibility in the open source community, is an employee of HP.

Ann Navarro, presently a W3C invited expert, suggested that it's unclear whether or to what extent Moglen and Perens will represent the open source and free software communities.

The W3C has a single set of rules for invited experts, which (to the best of my knowledge) have not been altered for this case. Those rules do not include any provision for "representing the public". If you wanted to take a particularly pessimistic view, an invited experts offer up their brain for draining, with very little flowing in the opposite direction.

Talk of whether Moglen and Perens represent some non-W3C contituency, like "the public", and which "public" they might be able to represent led quickly to talk about the W3C itself, its mission and its constituencies.

Len Bullard added that,

[p]erhaps for the first time, the software developing public is coming to understand that the W3C's "mantle of legitimacy" does not depend on their consent. That is why it is a consortium. This does not mean that independent developers are powerless; it means their power is much more limited than they were previously led or willing to believe.

Mike Champion suggested, however, that the W3C provides a place to balance or mediate opposing forces and interests.

The public needs the vendors to implement the Web technologies, and the vendors need to keep the public from getting angry enough at them to invest the time to hassle with the alternatives. The W3C is the arena for this balancing act; opening up the patent policy WG strikes me as an acknowledgement that the open source folks are collectively an alternative "vendor" that the W3C has to accomodate.

It is possible that the end-user, web surfing, Amazon-buying public doesn't care about this issue one way or the other, or might not care if a few dollars of their OS or app server or DBMS license goes to license patents ... they just want the bloody mess to work...

I basically agree with the anti-RAND position ... but I'm concerned that it WILL NOT get a fair hearing in the W3C unless you all address the issue on the same terms that the W3C addresses it ... or scare hell out of them with a credible alternative. The former is a lot easier to do than the latter!

The view of the open source community as another, albeit different kind of, vendor has its virtues. It does raise a particular tension that runs through most of the patent debate so far. There are open source developers like Red Hat, for example, which create and use open source and free software, but in order to realize a financial profit for its investors. However, there are other parts of the open source and free software community that develop software -- some of which is very high quality, some of which is not -- not in order to realize a financial profit, or at least not directly. While there are entitities that use, for example, the Apache web server and Perl and Python and Squeak programming languages, indirectly for profit-making activities, there are just as many that do not, and, equally crucial, the primary development effort of each of these technologies is non-commercial.

In other words, open source and free software advocates have spent a lot of time diminishing the importance of the distinction between commercial and non-commercial open source development. And that's often been important since the distinction is often irrelevant. But in the patent policy debates, the distinction may well be crucial.

Simon St. Laurent reflected part this tension clearly.

Describing open source projects as "underfunded" seems to reflect a serious lack of understanding about the nature of open source and some of the basic reasons that open source and patented processes aren't compatible.

Believe it or not, not everything is about cash.

This leads directly to another fault line running the entire length of the patent policy debate, namely, how should we understand and accommodate a situation in which non-commercial and corporate developers find themselves competing for market share? In fact, the question is even more difficult to pose since non-commercial projects, and perhaps Apache is the best example, may have commanding market share, which is neither intended nor leads directly to financial profit.

Two members of the XML-DEV mailing list offered opposing views on how patents and intellectual property rights intersect this fault line between corporate and non-commercial software development.

Jim Ancona offered the non-commercial perspective.

So I'm working on an open source or hobby project. BigCo sends me a letter claiming I'm violating their patent. I think their patent isn't worth the paper it's written on. I can:

a. Take out a second mortgage on my house to hire a patent attorney to fight them. If I win, I get to continue working (uncompensated) on my project, or...

b. I can say, "Sorry, BigCo, I won't do it again. Please don't sue me!" If I'm lucky they won't. If I'm unlucky, see a. above.

Which would you choose?

Joshua Allen represented the corporate view.

My impression is that hobbyists and small companies really don't have to worry about large companies litigating against them, while large companies tend to worry a lot about smaller competitors or opportunistic hobbyists litigating. If a hobbyist has even a reasonable chance of success, it is not difficult to find a lawyer who will litigate in return for a cut of any potential (likely out-of-court) settlement. On the other hand, large companies that litigate patents against smaller competitors often get counter-sued for anticompetitive behavior and lose impressively. Litigating against much smaller competitors, let alone hobbyists, is a very high-risk and low-reward behavior for a large company.

Allen went so far as to speculate that this issue, the clash of corporate and non-commercial interests with regard to patents, isn't likely to occur in the PPWG's deliberations.

Another XML-DEV member suggested that this tension is representative of two different approaches to technological funding and development.

If I, as a tax payer, and consumer, prefer to support my internet habit by having tax-funded universities donate IP to the common wealth and supporting companies that support unencumbered technologies, rather than by paying tolls to monopolistic corporations, then that's my right. Unless you can prove that this business model won't work ... I'll keep fighting that corner. Companies and states alike are here to serve humans, not vice versa.

What's Next?

From the most cursory examination of the multifaceted debate about the W3C's patent policy, at least one conclusion is inescapable: patents and intellectual property rights cut across some of the most substantive and fundamental public policy issues imaginable, including the nature of markets, the arrangement and allocation of public resources, and the rights of individuals to creatively pursue the realization of their interests, abilities, and projects. In short, this is an immensely difficult issue.

Given its difficulty, we can also conclude that, despite having invited two very experienced and capable experts, Eben Moglen and Bruce Perens, the W3C's PPWG needs, whether its members realize it or not, the constructive, substantive feedback, criticism, and comment of the XML development community, as well as of other interested observers and participants.

There yet remains time for you to make your voice heard on this significant and complex issue.