- From: Wilde, Erik <Erik.Wilde@emc.com>
- Date: Mon, 8 Jul 2013 16:33:03 -0400
- To: John Arwe <johnarwe@us.ibm.com>, "public-ldp-wg@w3.org" <public-ldp-wg@w3.org>
hello john. On 2013-07-08 19:37 , "John Arwe" <johnarwe@us.ibm.com> wrote: >issue-80 Accept-Post ... we'll need someone who will be here in summer >to put this through the IETF review process (we should hold off until >we're pretty sure we have all of them drafted though, like at LC draft >... I vaguely > remember we might have a second one somewhere in the queue of >resolutions, but not really sure). [1] has the process details, not very >scary + we're following an existing RFC's text modulo s/patch/post/* so >there should be less than average to go wrong. >... anyone reviewing -80 you can just search on -80, you will get >multiple hits. There was a bit of ambiguity in the record (should vs >must) so I drafted it as Should in the header registration (identical to >accept-patch) and > Must in the LDPC section 5.4.13 so it's symmetric to patch (where RFC >5789 says the header is a Should, and LDP says LDP Servers Must expose >it). i am writing I-Ds all the time, so i guess i could help with that. i am around all of summer, but that doesn't mean that things will move very quickly. IETF processes can be relatively slow, so we should not depend on this moving forward very quickly. it typically takes couple of draft versions to address the issues other people bring up. this slows down the process, but also means you get (typically very useful) feedback from experienced and interested people. and just to clarify: the idea is Accept-Post and not Accept-Post-Create as suggested in ISSUE-80, right? basically, this means http://tools.ietf.org/html/rfc5023#section-8.3.4 as a header, if i am not mistaken. Accept-Patch had it "easy" because it was specified as part of the PATCH method. for Accept-Post, it would certainly be useful to come up with some non-LDP use cases; AtomPub may be one, and there probably are others. and then of course the obvious question would be why the same spec shouldn't also add Accept-Put, but i guess that would make things harder and not easier. cheers, dret.
Received on Monday, 8 July 2013 20:33:45 UTC