W3C processing of IETF registration; registries and extensibiliy

I'm still a little stuck on how to word what we would like to see done.

Would it help if the TAG were to "Recommend" to W3C to not be a "bad netizen"? 

There is an architectural component and an administrative one.
The architecture relies on extension points using "registration" in some sense:
  * registration of extension points in a publically available and maintainable registry
	* IANA for those elements shared with other (IETF-managed) application protocols
                * other registries & organizations (?)
  * registration using a URI-named namespace, with extension points defined by the "owner" of the URI (presumably)
  * registration using vendor prefixes, with the 'vendor' identified managing it

The architecture notes the tendency to want convey status in the name  (vnd., x- prefixes, or vendor prefixes),
And the difficulty of updating status because names can't change.

For IANA registrations, the administration is that W3C staff & working groups need to manage the registration information, and that the process itself needs revisions.
Other registrations have their own administrative procedure.


-----Original Message-----
From: happiana-bounces@ietf.org [mailto:happiana-bounces@ietf.org] On Behalf Of Bjoern Hoehrmann
Sent: Friday, November 18, 2011 1:49 PM
To: Ned Freed
Cc: happiana@ietf.org
Subject: Re: [happiana] [apps-discuss] font/* (and draft-freed-media-type-regs)

* Ned Freed wrote:
>> Besides vendor-specific types, the only registered font type that I 
>> am aware of is 'application/font-woff'. W3C WebFonts WG has submitted 
>> a registration for this subtype about a year ago, and the text of the 
>> registration is available as Annex B of the WOFF specification:
>> http://dev.w3.org/webfonts/WOFF/spec/#appendix-b
>Submitted to whom? It isn't on the IANA page so it hasn't been approved 
>unless approval was very recent.
>It's one thing if this registration is still bouncing around inside of 
>the W3C process - that's entirely the W3C's baliwick. But if this was 
>submitted to the IESG for approval, having an outstanding request for 
>"about a year" is COMPLETELY UNACCEPTABLE. And this is very much the 
>main problem this entire effort was intended to address, and which we 
>seem to have great difficulty focusing on.

I noted this a couple of times, being unable to find out the status of proposed registrations is one of the real problems with the registry.
Per http://www.w3.org/2002/06/registering-mediatype the W3C considers this to be in ietf-types review, but they don't really bother to main- tain their own list much, and they certainly don't care to review the status of registrations when advancing documents. `application/xquery` is a good example, it's listed there as "Next step is IESG review" but the W3C published the Recommendation four years ago noting "This media type is being submitted to the IESG" and published a Second Edition a year ago noting the same thing, with no registration attempt I know of.

With the RDF DAWG I negotiated many years ago that I would not object to their inclusion of unregistered media types in one of the drafts, on the condition that they formally make advancement of the document contingent on the registration of the types (or else remove them from the draft). They did that, but then there wasn't really an attempt to register the types, and the W3C advanced the document with no mention of them failing to meet the advancement criteria and without removing the types. That may have changed since thanks to others, but there is no point in even highlighting such problems if the W3C goes out of its way to be a bad netizen.

I also note that their own policy doesn't actually identify who is to take care of registration, Chair, Team Contact, Editor, or whoever, so nobody actually feels individually resonsible to do anything, and they do not require Working Groups to report on media type registration status when advancing documents, not de-facto at least, so there is no safety net to catch the fallout from nobody being responsible either.

If the registration process would make the registration status visible to the general public at all times, and if the W3C would identify in their policy who is individually responsible to see the process through and would highlight the status during transitions for those who get to decide whether a document should be advanced there, that would help a lot.
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de Am Badedeich 7 · Telefon: +49(0)160/4415681 · http://www.bjoernsworld.de
25899 Dagebüll · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/ _______________________________________________
happiana mailing list

Received on Monday, 21 November 2011 19:06:43 UTC