Re: 9 July 2003 draft of "Client handling of MIME headers" available

On Thu, 2003-07-10 at 12:46, Roy T. Fielding wrote:
> >> At no point has the IESG ever been responsible for media type
> >> registrations,
> >> as is clearly documented in the RFCs that define the process for those
> >> registrations.
> >
> > On the contrary:
> >
> > "Registration in the IETF tree requires approval  by the IESG"
> >   -- 2.1.1.  IETF Tree
> >   ftp://ftp.isi.edu/in-notes/rfc2048.txt
> >  cited from
> >  "Application for MIME Media Type"
> >  http://www.iana.org/cgi-bin/mediatypes.pl
> >  <-- http://www.iana.org/protocols/forms.htm
> >   http://www.iana.org/
> 
> That isn't the process, Dan -- that is the end result: ALL
> non-informational RFCs are approved by the IESG as standard procedure.

Well, this is the part of the process where XENC and
SMIL seem to be stuck for ages.

[...]

> 
> There is no record of the W3C ever having submitted the "first step"
> as documented above for any of the following media types that people
> here have been complaining about not being in the standard configuration
> of Web servers:
> 
>     application/mathml+xml
>     application/rdf+xml
>     application/srgs
>     application/srgs+xml
>     application/xslt+xml
>     image/svg+xml

OK, I was talking about problems with XENC and SMIL; I see were
talking about different situations.

I can look into these others. Thanks for the details.


> And there are probably a lot more like them that I simply couldn't
> find via the W3C site search.  It would help if there was at least
> one page on the W3C site that listed the media types under development
> by all of the working groups.

Hmm... it seems more productive to make sure all those media
types can be found in the relevant IETF archives (e.g.
ietf-types) or provisional registries or whatnot.


> Let's pretend for a second that the first step was actually followed,
> placing the type on the radar screen.  After the review period, the
> application is sent to the IESG for review.   Given that they already
> approved RFC 3023, it is reasonable to guess that the IESG is not going
> to prevent a W3C group from registering in the IETF tree.

Well, XENC and SMIL show that this part isn't always so smooth.

>   All you have
> to do is publish an internet draft containing the same type of 
> information
> as provided in 3023 for the five types it registered -- mostly just
> references to the long-lived URIs provided on the W3C site for the
> format specifications -- which should be a simple copy and paste from
> the relevant W3C specification sections containing that form.
> Send a note to the IESG pointing to the I-D, placing it on the
> public IESG work queue -- this gives them something to approve
> and IANA something to refer to.  Finally, submit the form to IANA
> with reference to the IESG work item.
> 
> If you haven't been following the above process but rather dealt with
> the ADs directly, then it is also possible that the IESG has already
> approved registration of the type.  In that case, send the form to
> IANA with reference to the IESG minutes --- the IESG will not submit
> the form for you.
> 
> ....Roy

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Friday, 11 July 2003 18:20:32 UTC