- From: Roy T. Fielding <fielding@kiwi.ICS.UCI.EDU>
- Date: Mon, 04 Oct 1999 20:07:27 -0700
- To: Larry Masinter <masinter@parc.xerox.com>
- cc: Mark Nottingham <mnot@mnot.net>, HTTP Working Group <http-wg@hplb.hpl.hp.com>
>I think that the HTTP working group mailing list >is an appropriate forum for discussion of >Mark Nottingham's document "Server-Side Roles in the HTTP". Only if it isn't talking about issues of compliance and using the key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in as described in RFC-2119. That is not what an implementation guide does, and I doubt that the IESG would want that in an Informational document. I have no objection to discussions on how to best implement HTTP, or further rationale to back up the HTTP requirements. What I object to is implementation advice for general-purpose server technology that is made-over to look like protocol requirements. The W3C does that sort of thing, but the IETF does not (at least not without clearly distinguishing between informational content and protocol requirements). ....Roy
Received on Monday, 4 October 1999 20:11:30 UTC