Re: temporal URI fragments

S. Mike Dierken wrote to <mailto:uri@w3.org> on 20 March 2003 in "Re:
temporal URI fragments (was: URIBOF at IETF meeting S.F.)"
(<mid:OE38fY4jEbfdMwgdl1D00002fa8@hotmail.com>):

> I believe that the intent of a fragment was to address a subpart of a
> retrieved representation.
> By definition it is offset from the start of the local data that was
> actually received, not offset from the start of the remote data.

I agree.

> If you address anything on the Web, it is a resource.

URI references address things on the Web, things that are not resources by
RFC 2396.

>> The question then boils down to: what is it that we have to change in
>> the "URI generic syntax" standard to enable our suggested use of
>> temporal fragments? As it turns out: not much!
> I think a lot changes, and unnecessarily so.

I agree.

>> While this usage prescriptions may be appropriate for html pages, it is
>> not good for Web resources that consist of large volume data, of which
>> the user is only interested in receiving a small subpart.
> Then make that small sub-part addressable.

The fragment identifier proposal does make the sub-part addressable, at the
cost of mangling standard URI semantics. That this mangling is both bad and
avoidable is the opinion of at least two people on this list.

> How about using a ';' character instead of '#'?

Yes! As I contemplate the issue, I reject the query proposals as well as the
fragment identifier proposals. Extracting media runs is essentially a
protocol parameter: it has standard meaning in RTSP regardless of the server
and of the media type. Thus the expression of the extraction should be a URI
parameter, [";" param] in the grammar of RFC 2396.

Received on Tuesday, 8 April 2003 04:34:16 UTC