At the center of the patent storm

If you want to adhere to the latest official protocol for building a Web application, it could cost you.

Welcome to the latest controversy roiling the World Wide Web Consortium, the standards body responsible for shepherding Web technologies like XML and HTML.

This particular controversy began brewing nearly a year ago when the W3C first contemplated a change that would let its working groups incorporate technologies that already had intellectual property claims--and royalties--attached to them. The W3C subsequently backed away from that stance in the face of strong opposition and reaffirmed its policy of only recommending royalty-free technologies.

But the issue is far from over. A task force is still drafting a proposed exception to the royalty-free policy. Danny Weitzner, who chairs the working group charged with deciding the question, incurred the wrath of anti-royalty working group members by granting the task force an extension on its deadline. The full working group will evaluate the task force's proposal at the end of this month.

Weitzner is no stranger to Internet policy controversy. For a decade before joining the W3C as a full-time staff member, in 1998, the New York native worked in Washington, D.C., first as the Electronic Freedom Foundation's first Washington-based lawyer, and then for the Center for Democracy and Technology, which he co-founded and on whose board he still sits. In 1997, he was instrumental in bringing one of two lawsuits that helped bring the Communications Decency Act (CDA) to defeat at the hands of the Supreme Court.

At the W3C, the 38-year-old Weitzner heads up the consortium's Technology & Society Domain, which includes the W3C's Semantic Web activities--a project that has been likened to a resurrection of artificial intelligence. In addition to being co-principal investigator for the Semantic Web, he is chair of the Platform for Privacy Preferences (P3P) Coordination Group and chair of the Patent Policy Working Group.

Weitzner spoke to CNET News.com about the fate of the W3C's royalty-free policy and potential royalty exception, and about the role of the Semantic Web.

Q: Many--if not most--standards groups have some kind of provision for using technologies that are encumbered by some sort of license. The IETF (Internet Engineering Task Force), for example, has approved the use of licensed technologies for years. So why is it so controversial for the W3C to be considering an exception to its royalty-free policy?
A: Our conclusion that we need a royalty-free policy is about what we think is significant about the Web and the sorts of applications that implement out standards, and also the time and the patent environment that we're currently in. You could argue that the W3C came to this issue at a particularly difficult time because there has been an explosion of Web-related patents. Explaining why the IETF isn't in that environment is like trying to prove a negative. We've certainly never felt that the patent policy that's right for the Web is necessarily right for the rest of the world. When we were examining our options, people said we should just have a RAND policy like all the other standard groups, and we said we are not like the rest of the world and we should have a policy that reflects that.

We've certainly never felt that the patent policy that's right for the Web is necessarily right for the rest of the world.

To what extent are the W3C's RAND proponents concerned that the consortium is setting an example they don't like that other standards bodies might follow?
I have observed that some of the opposition to the royalty-free approach at the W3C has been less about royalty-free Web standards than about the possibility that that approach might be brought to other standards organizations. But my view is that each standards organization has to make its own choice based on what its needs are.

Can you say more about what, specifically, is different about the Web that makes it less suitable for accepting so-called encumbered technologies?
One of the things that's significant about the W3C environment is that you have a very broad, diverse group of participants and an even broader group of implementers than a lot of the other Internet standards do. For instance, MPEG standards are implemented by consumer-electronics and computer hardware companies. Those participants are in a better position to negotiate on a level playing field with patent holders.

What's the core motivation, then, for promoting a royalty-free policy?
The critical concern that has led us to push so hard for a royalty-free policy is that for all the different Web software implementers it would be terribly hard to negotiate with the parent holders. They don't have their own patent portfolios or intellectual property lawyers. How are they going to do it? That doesn't mean they can go and steal the IP, but we want to prevent putting them into a situation where they have to negotiate on such a lopsided playing field. We want to create standards that can be implemented without infringing on patents. In some cases, patent holders will still see it in their advantage to disseminate their technology for free.

Sounds like a lot of lawyers will get involved.
The goal is to avoid these very difficult, time-consuming licensing negotiations that require that implementers have lawyers and patent portfolios that they can trade with. We think there are very important pieces of the Web that can be developed by people who don't have those resources.

Like the open-source groups?
The open-source community has played a really important role at the W3C because, clearly, royalty-bearing standards create a fundamental problem for open-source software. But the need for royalty-free standards would exist even if there were no open-source solutions. What has been so successful about the Web is that its technology can be implemented everywhere by large and small developers, who don't have to worry about these licensing negotiations. The need for royalty-free standards goes beyond just the needs of the open-source community.

We reported that the fate of the W3C's RAND policy could hinge on your working group's Paris meeting in July. Did it?
There were two important points about that meeting. It was the last opportunity to raise new issues or reopen existing issues. It was the last time people could come and say this whole policy is a very bad idea, or there's something wrong in this little section. Most of the issues are now closed, but the question is whether we have a RAND exception process or not. What does the W3C do if we find out that we've been working on a standard and there's a technology with no royalty-free alternative? Can we ever implement a RAND technology?

How is this question different from the one your working group originally set out to answer?
Our group started with the premise that what we have is a royalty-free policy, and there were certainly people who said, "Let's try RAND." So first we asked, "Should we go to a RAND policy?" As a group, we have moved past that question. So number one, there was this affirmation of where we stand with the royalty-free policy. The issue that we spent a lot of time on is whether to have this RAND exception or not. I had put this issue on the agenda for the July meeting. What sort of RAND exception would we have if we had one? I had hoped we would be able to make a final decision on that question. But the actual details of that were not sufficiently worked out so that we could make a final decision.

What will it take to figure that out?
Some time ago, in January 2002, we created a small task force of the working group with the task of figuring out what a RAND exception would be. The group had a lot of time to work, but when it came down to July, I didn't feel that the proposal was ready for a final decision, or vote by

If Tim (Berners-Lee) were going to impose his own view as the policy, he would have done that two and a half years ago and saved us all the trouble.

the patent policy working group as a whole. So what I did was to defer the question until our next face-to-face, September 30 and October 1 in New York. What we call the RAND task force has until then to develop a final proposal. There was considerable objection from a number of the members of the group that I gave this group extra time, because they have had ample time to develop their proposal. But this is such a central question that we had to know at the end of the process whether we should have a RAND exception policy or not.

What happens to your work once you decide about the exception?
When a working group has developed a proposed recommendation, that proposal goes to the advisory committee, which consists of one representative of every single of the consortium's 490-odd member organizations. The director (Web inventor Tim Berners-Lee) looks at the comments and decides whether the thing should become a recommendation or not.

To what degree will Tim Berners-Lee's personal views on this decide the outcome?
Tim obviously has personal views on this question. In his role as director, he realizes that for lots of our members this is a core question about how the consortium operates. I think he's looking for a consensus position that he can support and that will be good for the Web. If Tim were going to impose his own view as the policy, he would have done that two and a half years ago and saved us all the trouble. He's watched this group work, looked at our product, and I think he'll respect the process we're going through. He'll make a decision. And I don't exactly envy him, because the role of the director is to assess consensus and identify a way to move forward.

As head of the W3C's Technology & Society Domain, you are responsible for the work being done on the Semantic Web. What's the connection between the Semantic Web and the society domain?
The Semantic Web is about making information on the Web more accessible to people. It's about being able to build machine-readable content to find and sort information better. From a public policy standpoint, the Semantic Web technologies are crucial because they help people understand the social and legal contexts they're operating in.

Can you give an example?
P3P, the Platform for Privacy Preferences, is a language, a set of semantics for describing privacy policies and individual privacy preferences. From a policy perspective, you look at the Web and say, "This is an incredible piece of technology for the way that people access and publish information, but it has really created significant policy problems arising in a privacy sense from the difference in understanding the social contexts you're operating in." In the real world, when you begin to interact with some organization or individual, it's pretty rare that they're able to sniff out information about you.

Unless they're Equifax?
But even Equifax--they don't sniff it out; you hand it over. Maybe you handed it over for a different purpose, but you handed it over. Let's say you walk into a library. If anyone's keeping track of what books you're pulling of the shelf, chances are you know that because they're standing right behind you. If you're at an online library, you have no knowledge of that. That's fundamentally a technology failure as well as a social policy problem. We've built this technology that allows this kind of information gathering to go on, but we've given users no tools to know about it. That's what P3P is about.

So P3P is part of the Semantic Web?
P3P is a technology that makes information more accessible to people by making it available in a more structured and machine readable way so that your browser can read it and make it useful to you. And that's why the Semantic Web activity is in Technology and Society. It's important for lots of reasons beyond privacy and security, but it's important in a particular way for public policy reasons.

How did your work fighting the CDA shape your thinking on crafting Internet policy?
That was a case where technology played a huge role, both in that the regulation of Web technology was the center of that case, and it was also one of the first indications that you could use technological tools to help solve social problems.

In what way, exactly?
One of the things that the Supreme Court found in striking down the CDA--which said that you couldn't publish indecent or patently offensive material on the Web--was that technology could be developed in a way to give individuals control over what information they or their kids had access to. So from that, we know that we don't have to regulate every single interaction between individuals and what's published online. We can in some degree rely on technology to do that.

Censorship is constitutionally accepted in some cases when it looks like only censorship can protect people from information that they might not want to be exposed to. What the Supreme Court found was that Web technology could help people make those choices themselves. That was a case where the characteristics of technology were crucial in sorting out the public policy question.