RE: Re: Fwd: Three bits on MediaTypes and IANA

Joseph,

There are some people who wish that IANA would keep 
www.iana.org organized in a way that URLs of the
form "http://www.iana.org/.../media-types/..." could
be used as abstract URIs to identify MIME media types.

However, there is no policy that IANA should do so;
IANA is free to maintain its web site as it sees fit
as a convenience to those who would look up registration
information informally, without any constraint about
the structure of the URLs used within it.

I, Graham Klyne, Ted Hardie and Michael Mealling did a
complete review of the IANA directory to see what
might be used to create URIs for IANA-registered protocol
elements, the results of which are at:

http://www.ietf.org/internet-drafts/draft-mealling-iana-urn-02.txt

After considerable reflection and discussion with IANA,
it seemed inappropriate to try to change IANA's work practice
for organizing its published information merely to create URLs
that would be useful for other purposes.

Personally, I believe this reflects the common confusion
between 'the stuff in a web site' and 'the concept described
by that stuff'; cf 'tdb' vs 'duri' in
http://www.ietf.org/internet-drafts/draft-masinter-dated-uri-00.txt.

For the particular purpose of identifying a complete
media type designation including parameters and
parameter values, Eastlake's 'content-type' scheme in
http://www.ietf.org/internet-drafts/draft-eastlake-cturi-03.txt
seems just right.


> -----Original Message-----
> From: Joseph Reagle [mailto:reagle@w3.org] 
> Sent: Tuesday, January 15, 2002 9:35 AM
> To: w3c-policy@apps.ietf.org
> Subject: Fwd: Re: Fwd: Three bits on MediaTypes and IANA
> 
> 
> 
> Noting that, "A cursory glance of this directory suggests that any
media 
> type registered through an RFC, doesn't get its own media type."
> 
> ----------  Forwarded Message  ----------
> 
> Subject: Re: Fwd: Three bits on MediaTypes and IANA
> Date: Tue, 15 Jan 2002 12:34:11 -0500
> From: Joseph Reagle <reagle@w3.org>
> To: Mark Baker <distobj@acm.org>
> Cc: www-tag@w3.org
> 
> On Tuesday 15 January 2002 01:29, Mark Baker wrote:
> > Hmm, odd, it seems some types get their own URI and others don't.
> > For example, RTF gets one;
> >
> > http://www.iana.org/assignments/media-types/text/rtf
> >
> > A cursory glance of this directory suggests that any media type
> > registered through an RFC, doesn't get its own media type.
> > How's that for a disincentive?! 8-)
> 
> Yes, it seems that the link for a specific sub-type points to an email
> request, but if it is in an RFC the link isn't provided.
> 
> > I'm all for more HTTP URIs, though I don't know that Eastlake's
encoding
> > (or any standardized one) is required.
> 
> When you're encoding characters in a URI you sometimes might have to
do
> some escaping and such.
> 
> > The relationship between;
> >
> >   http://www.iana.org/assignments/media-types/text/plain
> >
> > and
> >
> >   
> http://www.iana.org/assignments/media-types/text/plain?charset="foo"
> >
> > (or whatever URI structure IANA decides to use)
> >
> > should be made explicit through linking.
> 
> What do you mean through linking?
> 
> 
> --
> 
> Joseph Reagle Jr.                 http://www.w3.org/People/Reagle/
> W3C Policy Analyst                mailto:reagle@w3.org
> IETF/W3C XML-Signature Co-Chair   http://www.w3.org/Signature/
> W3C XML Encryption Chair          http://www.w3.org/Encryption/2001/
> 
> -------------------------------------------------------

>

Received on Wednesday, 16 January 2002 11:21:45 UTC