A belated storm of protest has greeted a move by the World Wide Web Consortium to bless fee-bearing patents as official web standards.

The proposal would allow patents, such as the notorious GIF image format, to become web standards, thus giving the patent owners the right to exploit them commercially.

Since so much information is stored and transmitted using web standards, the implications are far-reaching:-

"A patent-encumbered web threatens the very freedom of intellectual debate, allowing only large companies and big media houses to present information in certain ways," observed Linux kernel lead Alan Cox in a mailing mirrored at LinuxToday, which alerted readers this weekend. "Imagine where the web would be now if only large companies were able to use image files"

Obeying such pay-to-play standards will become a preclusion of remaining in the W3C club, if the policy is ratified:-

"If you are a W3C Member, you will be expected to comply with the Policy as a condition of continued membership," notes the W3C's FAQ on the subject.

This policy has enraged free software advocates and web developers, who liken the policy to laying down rules of behaviour for an occupation army, before the war has even started, let alone been lost.

What is RAND?
The W3C usually deals with technical details such as document and interoperability formats, but this is a "good behaviour" framework that's emerged from the W3C's Patent Policy Working Group (PPWG), and deals with how to distribute royalties in the event of a patent-holder seeking compensation.

Traditionally patents have been enshrined as standards on a RF or "royalty free" basis, but the PPWG proposes what it calls RAND, or "reasonable, non-discriminatory" license terms for IP where the licensee is wants compensation, and deals with collecting and distributing royalty income. A RAND license would not allow the patent holder to discriminate between licensees or add additional conditions to specific licensee. In other words, every will get screwed, but they'll get screwed equally.

However, this policy of "non discrimination" is essentially discriminatory, note critics: the non-commercial Debian Linux team would have to pay the same amount as Microsoft to incorporate a RAND license… and whatever that maybe in terms of dollars and cents, it's more than it is now, which is zero.

The plot thickens, however.

The W3C acknowledges there isn't anything on the table right now that could necessitate a RAND license. But it predicts there will be, and wants to deal with that eventuality.

It's also drawn a curious distinction between "lower" and "higher" level patents, as Adam Warner observes in the critique that brought the issue to the community's attention this weekend:-

"For example it might be considered that a moving picture format is sufficiently high level for RAND licensing to be appropriate. But if that moving picture format becomes an integrated baseline technology in future products then the chance of a future fee being associated with that technology could be devastating."

Out of a clear blue sky?
There's much introspection and anger at how such a wide-ranging proposal could sneak under the radar. The proposal was published on August 16, with an invitation for comments, and as veteran license watcher Karsten Self notes here the timeline makes for a sorry read:-

- August 10, 2001: discussion list posted.

- August 16, 2001: draft published.

- August 22-28: 5 spam mailings posted to list.

- August 26: First topical post to discussion by Hartmut Pilch, well kown to gnu.misc.discuss, asks what impacts of RAND for free software might be. There is no response to this post.

- September 1 - 20: 5 additional spam postings to the list are made.

- September 24: First topical post is made to list for the month, by a W3C staff member.

- Sept 27-28: Three posts are made to list detailing concerns regarding RAND and free software. No response is made.

- September 29-30: 747 posts are made to the list. Wups...I guess the word got out ...

I'm afraid what we've got here is a failure to communicate," he writes.

Self takes himself, many of his peers and the press - including us - to task for missing this. And it's a fair point.

As is how the W3C has been so effectively captured by the patent lobby.

For now, however, focus is on derailing RAND, as the W3C is slated to make this official policy in February 2002.

The omens aren't good: in a response to LinuxToday at the weekend, W3C's Janet Daly wrote:-

"As of now, many comments sent to W3C's Patent Policy Comment list simply say, "Don't!" By responding in this manner, writers give the Working Group nothing on which to build a constructive response or to consider."

More positively, we understand that the coalition-building process that takes the web standards out of the hands of any one group, particularly one as susceptible to capture as the W3C, has already started. ®

Related Link

The W3C's RAND FAQ