- From: Henrik Frystyk Nielsen <frystyk@w3.org>
- Date: Tue, 13 May 1997 16:31:49 -0400
- To: Larry Masinter <masinter@parc.xerox.com>, "David W. Morris" <dwm@xpasc.com>
- Cc: Koen Holtman <koen@win.tue.nl>, http-wg@cuckoo.hpl.hp.com
At 10:23 AM 5/13/97 PDT, Larry Masinter wrote: >So I'm wary of using URLs in PEP or in feature negotiation >as a way of identifying content types and features. It's not >that it is unworkable, but the issue needs to be addressed. But URLs don't all have to point to "http://my-flacky-company.com/~henrik". Well-proven extensions can point to (hopefully) stable things like IANA, IETF, ISO, W3C, and the like. I can see no reason why URLs can not be just as reliable as registered tokens. To me, this is a question of policy, not technology :-) Used together with redirections and metadata in general, I think we have a much better way of transitioning than with the "x-" hack/kludge. Also, tokens and URLs can go hand-in-hand much more than is currently the case. We could introduce relative URLs being used as identifiers and define a default, absolute URL for resolving it (for example one of the locations above). If we chose it wisely then existing tokens could be interpreted as relative URLs. There is no reason why tokens can't fit within the URI space. However, I agree that careful wording needs to be thought about. Some of the issues are touched in the current PEP spec, but more may be needed. Thanks, Henrik -- Henrik Frystyk Nielsen, <frystyk@w3.org> World Wide Web Consortium http://www.w3.org/People/Frystyk
Received on Tuesday, 13 May 1997 13:37:24 UTC