- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Wed, 02 Aug 2006 11:59:41 +0200
- To: suma@soe.ucsc.edu
- CC: w3c-dist-auth@w3.org
Hi Suma, there's quite a lot of previous discussion on the HTTP WG mailing list that you should review, for instance around the thread <http://lists.w3.org/Archives/Public/ietf-http-wg/2004OctDec/thread.html#msg4>). So I would approach the topic this way: - discuss on the HTTP WG mailing list, not here. This is not specific to WebDAV, - realize that APPEND is a special case of PATCH, - also realize that PATCH itself doesn't need a lot of work. By submitting a PATCH request a client asks the server to modify the existing resource based on the instructions in the message body. Full stop. How the message body is to be interpreted depends on it's Content-Type (so you don't need "Patch-Type" at all). All the spec really needs to specify is at least one very simply diff format, and a MIME type for it. Speaking of which I see use cases for at least two patch formats (one for textual diffs such as in the Unix diff command output), and one for modifying parts on binaries (seek/write/etc). <http://greenbytes.de/tech/webdav/draft-dusseault-http-patch-04.html> of the PATCH draft actually used that model, and IMHO it was a mistake to move away from it. Quite logically, draft 05 it good serious pushback from Roy Fielding (<http://lists.w3.org/Archives/Public/ietf-http-wg/2004OctDec/0011.html>). Best regards, Julian
Received on Wednesday, 2 August 2006 10:06:24 UTC