- From: <keshlam@us.ibm.com>
- Date: Wed, 24 May 2000 15:01:03 -0400
- To: "Tim Berners-Lee" <timbl@w3.org>
- cc: xml-uri@w3.org, "Tim Bray" <tbray@textuality.com>, "Simon St.Laurent" <simonstl@simonstl.com>
>(Something which you compare as a string is not a relative URI.) Reminder: Some of us -- Tim Bray among them -- have been saying for the past month that namespace names _AREN'T_ relative URIRefs (*), and that the Namespace spec needs an erratum saying that these names are strings... whose recommended values just happen to follow the URI-reference syntax. If we were doing this over from scratch they would probably be URIs (or perhaps URNs!), thus forbidding the relative syntax entirely and eliminating the point of contention. As I noted elsewhere, it's a question of which of three words you consider to be the error... and whether we can minimize the damage done by correcting that error at this late date. Even after all this chatter, the leading proposal for an effective solution still seems to be Literal-and-deprecate... which can also be described as Forbid-with-migration-path. (* Can we please be careful about terminology? URIs are absolute by definition. URI References may be relative, as well as permitting a locator suffix. Forgetting that difference is what got us into this hole in the first place.) ______________________________________ Joe Kesselman / IBM Research
Received on Wednesday, 24 May 2000 15:01:46 UTC