Weblogs: Atom

W3C invitation for Atom Working Group Proposal

Thursday, May 13, 2004

Via Danny Ayers' Raw Blog Atom Invited to the W3C, the W3C have replied to the Atom IETF proposal by inviting Atom to propose a W3C Working Group. This is mainly because Atom is a web-focused specification, so it fits in well with the W3C's recommendations.

W3C sees Atom as an important Web application and one that supports the W3C mission to lead the Web to its full potential. The Atom group has done wonderful work achieving a consensus in the community by defining requirements and initiating the standardization process.

Praise for Atom

There is praise from the W3C over the Atom process so far. They are impressed with the number of implementations, and more importantly an established testing effort. Also worthy of praise was the intention to make Atom a standardised specification.

If Blogger's adoption of Atom wasn't a big enough sign of the community support of Atom, a vote of confidence from the W3C leaves no doubt as to Atom's worth.

Issues / Direction

The W3C have outlined issues / requirements they think is important: XML-Accessibility, Internationalisation, Device Independance, XHTML cooperation. Their position on RDF is that they are agnostic to whether Atom is XML or RDF.

Annotea

The one interesting synergy or potential conflict I see is W3C's Annotation Project Annotea. The plus point is that Annotea's API is REST-based, and the W3C host the SOAP specification - this in a way confirms that the W3C can be agnostic. Annotea, however, uses RDF - well, it is a demonstration application of Semantic Web Development.

Annotea is used to create page-independant annotations of web pages. It also allows these annotations to be shared across Annotea users. Quite a large number of blogs are focused on commenting on other peoples pages / blogs, so there is an area of contention. The Annotea API is similar to Ken McLeod's Particle Wave idea.

My current opinion

I'm excited about the W3C's reply, although I don't have the experience or knowledge to decide either way whether the W3C's proposal is good for Atom or not. The actual people behind the proposal are all actively involved in the Semantic Web activity. This does raise the concern of re-doing the discussion over RDF all over again.

Of course, I don't mind which standards organisation the Atom group decides to go with - they have my support either way. My heart says working with the W3 would be cool, but my head says a good adoptable standard needs a lot more than cool to be successful. I await with interest what other Atom participants think.

Concern about individual ad-hoc developers

My initial concern is the membership is largely members or invite only - which may stifle a few people. I'm not really a syntax developer, I'd rather take a specification and code stuff using it, try out things in code. I'd obviously like to continue participating in that manner in the current ad-hoc basis I've done so far. I probably don't need to be invited - as long as I can follow along and participate in the mailing list when the reason neccessitates.

Eric Miller is the current W3 spokesman on this invitation, and his reassuring reply is:

I think we meant to say we're not going to take this out of the hands of the people who are working on it: Atom has matured thanks to the people who have so far contributed to it, and we want those people to continue to contribute to it in the standards process.

Concern of the community versus business consortium representation

My next concern is the status of Mark Pilgrim and Sam Ruby. At the onset of this project Sam Ruby had a sign-off from his manager to steer the Atom process - and which he's done a magnificent job so far. Mark Pilgrim joined IBM a few months ago as an accessibility architect - his endeavours with Atom were unrelated to his new job with IBM. Both are key members of the Atom project - they both need to be able to work with others in this endeavour, preferably with equal voting rights. If the W3C Working group is the preferred process, something needs to be done to ensure that both can continue uninhibited. (I guess the ideal is for Mark Pilgrim to be a part of this project as an individual - independant of IBM). On the same note, a number of Google employees are working on Atom too - I'd want this to continue as a community effort, not a one company one member process.

The voices of Atom contributors (updated)

Danny Ayers is quite positive about the proposal.

Atom deserves to be treated as a first-class citizen, up there alongside the "big" specs. The IETF is a good organization, but there is a danger of Atom becoming obscured alongside 10,001 other RFCs. I believe the W3C would provide more than enough visibility to guarantee widespread adoption.

Joe Gregorio is positive with some reservations:

While I think its great that the two organizations are both interested in Atom I do have my reservations about the W3C which mostly mirror that of Mark Nottingham. My biggest concerns that anybody be able to join and participate which isn't how the W3C currently operates, and that RDF not be foisted upon Atom.

Sam Ruby seems positively upbeat:

To me a very important intangible is a willingness to participate openly in venues like mailing lists, weblogs and wikis. To me, that is more important than the fact that the W3C has an office in Morocco. And on this account, on this particular proposed standard, and at this point in time, the advantage clearly goes to the W3C. Perhaps this is not your fathers W3C after all.

Tim Bray is not against the proposal:

Personally, I think we could get a good result in either organization and would not push back against the community if many voices were raised in favor of switching horses to the W3C. Looking at the pros and cons above, my leaning is towards the IETF.

both Sam and I have willingness to take it on and sign-off from our employers, wherever it is.

Asbjørn Ulsberg - reservations aside - supports the idea:

I have some doubts about this, but if I just put it aside, I have to agree with and repeat after Danny: "I wholeheartedly support Atom becoming a W3C-based specification".