W3C home > Mailing lists > Public > xml-dist-app@w3.org > November 2003

Re: Update: experiences with SOAP media type registration

From: Chris Lilley <chris@w3.org>
Date: Mon, 3 Nov 2003 03:41:19 +0100
Message-ID: <6637179341.20031103034119@w3.org>
To: Mark Nottingham <mark.nottingham@bea.com>
Cc: public-ietf-w3c@w3.org, "Xml-Dist-App@W3. Org" <xml-dist-app@w3.org>

On Monday, November 3, 2003, 2:44:28 AM, Mark wrote:


MN> Since sending the message below, the IESG considered and rejected
MN> the registration of the application/soap+xml media type,
MN> apparently because the draft did not actually contain its
MN> registration (instead, it referenced the registration in the SOAP
MN> specifications),

In future, (once there is a 'Standards Tree'), having the registration
in the W3C spec and pointing to it will be the way to go. Until Ned
Freed's ID that revises the registration process becomes a Rec,
though, the registration information has to be copied into a separate
ID not pointed to.

MN> and because it referenced works in progress (e.g., the SOAP 1.2
MN> specifications, which were not REC at the time).

Its a Rec now. What was it at the time?

There is a fundamental problem in that

a) Media registrations need to reference stable documents
b) Stable is defined as Rec
c) To get to rec you need to meet your CR exit criteria
d) To do c) , you need to demonstrate  interoperable implementations
e) implementations need to use a media type
f) goto A

In future, it should be stated that the CR is a 'stable document' for
the purposes of media type registration (it has after all successfully
left last call), and thus that a media type can be registered at that
point. If subsequent changes affect this registration, then the
registration should be updated when you get to Rec.


MN> Unfortunately, the IESG did not notify the authors of the  
MN> Internet-Draft of the decision made by e-mail; Ned Freed has said that
MN> the rejection was communicated to people "on the W3C concalls," but
MN> this information didn't filter down to the WG. (If anyone has  
MN> information about this, please contact me; we still need to resolve the
MN> registration of application/soap+xml).

'Telling W3C' is rather like 'telling IETF'. Clearly (and not just for
W3C working groups) the authors of a registration need to be told
whether it has succeeded or not and if not, what they can do about it.

-- 
 Chris                            mailto:chris@w3.org
Received on Sunday, 2 November 2003 21:41:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:15 GMT