- From: Jeffrey Zeldman <jeffrey@zeldman.com>
- Date: Tue, 9 Oct 2001 17:38:56 -0400
- To: www-patentpolicy-comment@w3.org
- Cc: Web Standards Steering <wspsc@webstandards.org>, list@webdesign-l.com
A clear W3C patent policy is probably overdue - and a good one could ensure interoperability and protect web standards while avoiding nuisance litigation such as Intermind's recent attempt to undermine P3P. But the working draft published at http://www.w3.org/TR/2001/WD-patent-policy-20010816/ invites more problems than it solves. The policy will require significant revision if W3C hopes to rally the support of the web development community. The draft as written is is maddeningly vague and thus more than vaguely alarming. For instance, lack of clarity in the language has led thousands of developers to fear that they'll have to "pay royalties to use HTML." Developers are not to blame for these fears; the draft policy, as written, invites them. Some have been so baffled by the policy's language and so fearful of its implications that they are threatening to fork the web by creating "new standards" within new, ad hoc standards bodies. I'm sure you'll agree how destructive that would be to the interoperable, standards-based web that W3C has worked so hard to promote and develop. I believe there is a message in the general outcry from a community that up until now has absolutely supported W3C efforts. Remove the rhetoric, and what you hear is many intelligent voices expressing concern for good reason. I also believe that many of the community's concerns could be resolved if W3C presents a substantially revised document that corrects the weaknesses of the current working draft. Here are a few areas that strike me as needing work (there are surely others): Section 2.2 says "it is especially important that the Recommendations covering lower-layer infrastructure be implementable on an RF basis," but fails to specify what is meant by "lower-layer infrastructure" and what specific remedies W3C would persue if Members abused W3C by attempting to charge royalties for these (so far unspecified) elements of lower-layer infrastructure. To address these weaknesses, W3C first needs to spell out which standards (both existing and under development) are and will remain royalty-free. Such terms as "core," "essential," and "non-essential" must be explicitly defined, and each definition must *list the specific standards it encompasses.* Failure to explicitly state which standards are protected from patents and/or royalties will lead many in the development community to assume the worst, and could encourage corporate attorneys to engage in predatory behavior. Even if corporate attorneys do NOT engage in predatory behavior, fear that they MIGHT do so could short-circuit standards currently under development and paralyze Open Source projects that benefit us all. Clear, specific language could alleviate all those concerns. Likewise, if RAND licensing is truly intended to apply only to "inessential," web standards, we need to know precisely what those "inessential" standards are. Next, W3C needs to define explicit penalties for breaches of good faith, lest errant Members quash emerging standards after months or years of collective work, merely by failing to disclose an existing patent (since there is no explicit punishment for such behavior). Per the WaSP opinion (http://webstandards.org/opinion.html) previously sent to this list (http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1226.html) : "As the policy stands, [no W3C remedies are specified], and we are to take it on faith that not only will Members disclose patents and applications they may not have had a hand in submitting, but if they do not, the Working Group will be disbanded and the technology abandoned in its current form. "This unacceptably weak link in the policy essentially allows any Member to sabotage any W3C technology simply by filing for a patent application." That WaSP opinion also proposes EXPLICIT remedies for breaches of good faith, which I repeat here for your convenience: "We suggest that any Member found in violation of the disclosure clauses of the policy be barred from participation in or communication with the other Members of the various W3C Working Groups for a period of no less than two years. This should extend not only to the individual representative, but to all representatives of any Member determined to be in bad faith, whether currently involved in Working Groups or not. "Such violators should also be required to abandon all claims to RAND licensing for the technology in question, and instead make the technology royalty-free as a demonstration of their commitment to open standards and fair and open cooperation." These are merely recommendations, and the committee may come up with better ones. The point is that the policy must provide strong, clear language including explicit consequences for bad behavior; otherwise the policy will be powerless to prevent such behavior. W3C has taken the first steps toward developing a patent policy. Now it needs to take write an *explicit* policy that avoids any loophole capable of being exploited by corporate patent attorneys, defines what is meant by *essential* web standards, and guarantees that all such essential web standards will be either unencumbered by patents, or if they are patented, will remain royalty-free in perpetuity, per the promise of your own Member Agreement (http://www.w3.org/Consortium/Agreement/Full), which grants Members a "non-exclusive, royalty-free, irrevocable, right and license" to use and contribute to web standards. Addressing these weaknesses now will enable W3C to prevent grief instead of inviting it, and will help ensure that innovation, interoperability, and fairness continue to be the goals of W3C and the watchwords of the web. Thank you for your time. Jeffrey Zeldman z e l d m a n now in pulp for your pleasure: TAKING YOUR TALENT TO THE WEB http://www.zeldman.com/talent/ daily: http://www.zeldman.com/ weekly: http://www.alistapart.com/ as needed: http://www.webstandards.org/
Received on Tuesday, 9 October 2001 17:40:27 UTC