W3C Ponders RAND Again

simonstl writes "Three unnamed W3C participants have suggested a new RAND policy that would let the W3C into the business of charging royalties for patent-encumbered specs. No consensus yet, but they sure seem to keep trying."

Comments

Charging for the Extensions?

[TibbonZero]

So they are going to charge royalties for extensions? Does that mean that I will have to pay to use .cgi on my site? Or am I thinking about the wrong thing?

Whatever it is, it seems to work against the standards that they are trying for

Interesting parallel...

[Noryungi]

To the Slashdot discussion that appeared on how webmasters were coding mostly for Microsoft Internet Explorer.

I mean, IE already does not respect the W3C standards as well as it should. And it is the dominant force on the net today.

If W3C was to charge money for some IP, what can prevent MS from saying: "Fsck off, we will set the standards, because we are the standard"?

Think about it. And be afraid. Be very afraid.

Don't get your panties in a wad just yet...

[onlyabill]

Unless I misunderstood, they are not saying that they would consider charging for IP. It appears to me that they were discussing how to handle standards proposals that included licensable, royalty work. That the normal(?) process is to break a specification into non-royalty works and royalty works in such a way that the non-royalty works would not depend on the royalty works and that the royalty works would be farmed out to another organization.

"When non-royalty-free IPR is discovered in a W3C Working Group, a Patent Advisory Group (PAG) might have a number of possible outcomes. The proposal suggests that the Working Group's specification could be split in two. "Core" work to be licensed royalty-free would be produced as a W3C Recommendation. "Extensions" that may require royalties could be done at W3C or by another standards organization."

The question appeared to be if that should continue. Of if W3C could work on the royalty works as well or if they could not go through the trouble of splitting the proposal and work on the entire thing.

"One person suggested Extensions work be dropped. Some said work at the other standards organization needs to be a cooperative effort with W3C. One person suggested a hybrid all produced at W3C. One asked if the Core/Extensions split makes Extensions work less important than Core. The consensus seemed to be that specs built on top of Core work are valuable and that standardizing them is important."

Re:Don't get your panties in a wad just yet...

[Ami Ganguli]

By posing it as an 'either/or' question (either work on the encumbered extension or let somebody else do it), you neatly avoid other alternatives:

• challenge the patent and try to get it overturned,
• develop an unencumbered alternative and standardize on that instead,

If the only existing solution (open or proprietary) for a given problem is patent encumbered then somebody needs to develop a non-encumbered alternative. Walking away from the issue and saying "there's a proprietary solution in this area, so we won't develop our own" is no solution at all. It pretty much defeats the purpose of having a standards body.

W3C backstabs their only significant supporters

[smiff]

We all know RAND is incompatible with open source. Without open source, there would be no W3C! If every open source browser stopped supporting new standards, there would be only one dominant browser. That one browser shows no intentions of adhearing to the W3C's standards. How can the W3C not see this?!? Without open source, they are DEAD!

Patents and prior art?

[Anonymous Coward]

This reminds me of Apple's patented alpha blending, and the time when we feared we had to burn our PNGs

Some granted patents are really bogus. If the W3C is working on some spec, and it stumbles upon someone's patent along the way, the patent should be scrutinized very closely. I would believe most could be disallowed on the grounds of prior art - instead of having W3C charging people for royalties. This could shake the whole IPR issue very nicely too.

z

No, this is not about W3C staff wanting RAND specs

[Ankh]

No, this is not about W3C staff wanting RAND specifications.

It's about what to do if we're working on a specification that the community (including the open source community) needs/wants, and we discover that some aspect of it is covered by software patents.

In that case, you might not be able to have an open source implementation, and W3C has to ask, (1) should the work be dropped altogether, or (2) is there a central core that can be implemented freely, avoiding the patent? If so, should the non-free part still be standardised, and under what terms?

By making an extension to a specification, an implementation can conform without that (possibly non-free) extension, but at the same time w3C can require W3C member organizations to agree to "non-discriminatory" terms, i.e. forcing them to agree to licence the patent to their competitors.

That may be better than having no specification at all, I don't know.

Note: I work at W3C, and am not involved in the specific work mentioned, nor in the patent policy group.

Stop Apologizing for these jerks

[FreeUser]

Don't get your panties in a wad just yet

Translation: "Be sufficiently cowed by my pre-emtive ad-homonim attack so as not to speak out on this important issue"

Unless I misunderstood, they are not saying that they would consider charging for IP.

This is deceptive semantics. It makes no difference whether it is the w3c charging a patent royalty, or Microsoft charging a patent royalty. If the royalty is requried to adhere to a w3c standard because the fools have incorporated patented material into the standard, then the damage is done regardless of who the web/browser authors have to pay the Vig to, and a once open and free (beer, speech) standard will have become much less open and very unfree, i.e. it will have become worse than worthless to the majority of small web publishers and free software authors, it will have become something to be actively avoided.

It appears to me that they were discussing how to handle standards proposals that included licensable, royalty work. That the normal(?) process is to break a specification into non-royalty works and royalty works in such a way that the non-royalty works would not depend on the royalty works and that the royalty works would be farmed out to another organization.

There is already a procedure in place: do not allow patent-encumbered works into open internet standards, period. As another poster pointed out, which part of No! No! No! didn't they understand the last time around? This is an attempt to change the procedure to allow something which, for very good reason, is disallowed now, i.e. the incorporation of patented procedures into w3c standards that will result in an open and free standard becoming encumbered, less open, and not free at all (in any sense).

In other words, this is yet another sordid attempt to sneak patented work into the standard through the back door, and it deserves every bit of derision and outrage it is getting.

Re:No, this is not about W3C staff wanting RAND sp

[Ankh]

Reading your other comments, I think all I can say usefully is that it's not a goal of the w3C team members (the staff) to publish specs that cannot be implemnted openly and freely. You (and others) may be reading more into the wording than was intended. But, it's hard to read and interpret email without the full context, and neither of us were at the meeting.

I'm not sure I have explained this very clearly - Digital Mage, feel free to email me if that will help (liam at w3 dot org). I have not seen an "official" W3C response, and neither can I give you one, but I can maybe help you understand the issues, and also show you how you can send effective feedback to W3C working groups.

Liam

[sorry, my personal web site is down right now, www.w3.org/People/Quin works though]

RAND cannot be standard

[HiThere]

Closed standards are not standards.
Proprietary standards are not standards.
Whether you want it or not, if you propose acceptance of RAND specifications, then you are proposing that the W3C cease to be a standards body.

The purpose of a standard is to define the way in which things should be done. If a patent prevents them from being done in some way, then that is not a possible standard way to do them. Something which had been a standard, if prevented from use by a patent, should be withdrawn, and the name made unuseable. (If names are trademarked, this means revoking all authorization to use the name. Otherwise it means revoking all authorization to use the name of the W3C in conjunction with the prior standard.)

Anything less is unacceptable. Patents are only allowable as long as they to not impede standard conforming usage.

"non-descriminatory" is a term created by companies that intend to descriminate. It is blatant camoflage, and should not be acceptable. Non-free specifications are not standards, and cannot become standards as long as they remain non-free. The concepts are in direct opposition. Proprietary is feasible, but there is a strong tendency to confuse proprietary with the right to be paid. So to be explicit here, if you must pay to use a specification, then it cannot qualify as a standard. Perhaps one would need to pay to be certified as standard compliant. That's a totally separate matter. The right to certify something as being W3C standard compliant must inevitably belong to the W3C, and perhaps it wouldn't be unreasonable for them to delegate that right to a third party. But charging for the use of a standard is incompatible with the concept of standards.