- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Mon, 12 Aug 2002 03:38:38 +0200
- To: "Larry Masinter" <LMM@acm.org>
- Cc: <uri@w3.org>, "Roy Fielding" <fielding@apache.org>
* Larry Masinter wrote: >I've been asked to propose a charter for a possible URI working >group to work on next phase topics. The working group should >be narrowly focused to insure viability and keep us from wandering >into unsolvable problems. For this purpose, I propose the >following items to be in scope: I'd like to have a BCP RFC documenting the recent practise for new fragment identification mechanisms like XPointer or as proposed in XFrames to use some functional notations in the fragment identifier just like document#xpointer(...) document#frames(...) document#xmlns(...)%20element(...) ... >OTHER TOPICS > >I think it would be good to finally 'obsolete' RFC 1738, >but it's necessary to do something, I think, with the >URI scheme definitions that are only documented there: >"file" is used widely but documented correctly nowhere. >"http" was defined in the HTTP spec and "mailto" has its >own RFC, "telnet" is still used, "news" and "gopher" >less so, and I haven't seen a "nttp", "wais" or "prospero" >in years. 'nntp' is less common, because most software allows the 'news' scheme to refer to NNTP servers, just like news://nntp.example.org/ followed by either a message id or a news group name. 'news' is rather common, though, but mostly on Usenet for obvious reasons.
Received on Sunday, 11 August 2002 21:38:34 UTC