W3C home > Mailing lists > Public > www-tag@w3.org > March 2002

[uriMediaType-9] New Internet Draft: Resrep Type

From: Sean B. Palmer <sean@mysterylights.com>
Date: Wed, 27 Mar 2002 13:58:28 -0000
Message-ID: <00c801c1d597$7a2f9040$ceb90150@localhost>
To: <www-tag@w3.org>
A new Internet Draft has been published that is relevant to the
uriMediaType-9 TAG issue, and vaguely relevant to the "HTTP
range" debate/issue (yet to be assigned an issue identifier).


http://www.ietf.org/internet-drafts/draft-palmer-resrep-type-00.t
xt
   - Resource and Representation Typing in HTTP [1]
   Sean B. Palmer, March 2001, Internet-Draft (work in progress)

It specifies two new HTTP headers: "Resource-Type" and
"Representation-Type". Representation-Type allows people to type
a document using a URI rather than just a MIME type, stopping
centralization around the IANA. In fact, the proposal in its
current state allows for multiple representation types to be
given, possibly along with a content type.

Note that the wording in the draft strongly implies that HTTP
URIs can identify any resource given the definition in the RFC
2396. This implication is, in my view, appropriate for the draft
since it gives one the ability - via. an authoritative method -
to give the type of any resource that is identified with an HTTP
URI; I would like to state explicitly that the value is not
semantically limited, pending a TAG decision on the range-of-HTTP
issue. In fact, the current wording is technically not
incompatible with the view that HTTP URIs can only identify
generic documents.

I note that the draft findings for the uriMediaType-9 issue
contain the following line:-

[[[
the W3C TAG would prefer that MIME media-types map into the HTTP
URI scheme with some well-known base URI
]]] - http://www.w3.org/2001/tag/2002/01-uriMediaType-9

Personally, I don't agree with that finding; it completely
ignores the fact that other schemes are dereferencable (which, is
my objection to HTTP-deification). It should at least defer to
the "is HTTP special?" issue raised on www-tag.

In any case, the res. and rep. type header values are defined as
being any URI-view (URI, or absolute URI with fragment), and
therefore include non-HTTP schemes.

Note that the main purpose for the draft was to provide the
"Resource-Type" header; "Representation-Type" was added mainly as
an afterthought.

[1] N.B. The present version is archived at
<http://web.archive.org/web/20020327053715/http://www.ietf.org/in
ternet-drafts/draft-palmer-resrep-type-00.txt>.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://purl.org/net/swn#> .
:Sean :homepage <http://purl.org/net/sbp/> .
Received on Wednesday, 27 March 2002 08:59:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:05 GMT