HTTP working group status

Here's my take on the current status. As usual, please let me know if
I've missed anything, or mischaracterized the state:

- HTTP/1.1: on the way to RFC, and Proposed Standard.
- digest: ditto.
   Yip yip hooray hooray, etc.

Both of these are going to be Proposed Standard. For the meaning of
'Proposed Standard', please read RFC 1602: 'Proposed Standard' is not
'Standard', it means that we've proposed it, and people should give
feedback when they go off to implement it. There were a number of
squawks about various ambiguities and things that need to be addressed
sometime between Proposed and Draft, some open issues (Age), and other
kinds of editing things. It's probably premature to start on this: I'd
like to see a HTTP/1.1 RFC before there's a new draft HTTP document
though. There's also some wrangling about whether digest will be
implemented as widely as some would hope.

cookies:  draft-ietf-http-state-mgmt-03.txt
   Last call expired August 20. No discussion that I can see.
   Expect IESG action soon.

SHTTP: draft-ietf-wts-shttp-03.txt
   I believe the Last Call expired; I've seen no discussion.
   Expect IESG action soon.

New work:

hit metering: draft-ietf-http-hit-metering-00.txt
    There's been some discussion on this on the list, and it's
    unclear that the facilities proposed here will be useful
    enough in the user community to go forward with this.

PEP:  draft-khare-http-pep-01.txt
    This is a 'discussion draft'. There was much dissatisfaction
    with the draft in terms of not being clear enough about
    what the protocol actually was. A 'specification draft'
    has been promised, but there's some ambiguity in W3C
    position as to whether this is actually being submitted
    as a 'working group' item.

content negotiation:
	draft-holtman-http-negotiation-02.txt
	draft-mutz-http-attributes-01.txt
    Koen and Andy are working to produce revised drafts of
    content negotiation, feature negotiation, and user agent
    characteristics.

sticky headers & header abbreviation:
    A draft was circulated on the mailing list (I don't see it in
    internet-drafts.) Some skepticism about whether this will buy
    anything.

referrals proposal: (no draft)
    draft was promised, but never appeared.

versioning: (no draft)
    A group on versioning and distributed authoring is working
    on a requirements document, may want to add optional header
    to GET, or optional methods (LOCK?), etc.

My analysis:

Content negotiation won't require a version change (I don't think).
It can proceed as a separate draft or dock. 

PEP may require a version change. Before the working group can
consider PEP, we need clarification from the author as to its status,
and clarification of the document.

Most everything else is stalled or else speculative.

There's no way we will be done by December, mainly because we need
more implementation experience with 1.1.

Recommendations:

- wait for new negotiation, feature & characteristics drafts &
  consider them. (This week?) 
- wait for clarification on PEP status. If it doesn't arrive this
  month, drop PEP from our charter.
- consider 'hit metering' and 'sticky headers' and 'header
  abbreviation' for Experimental, unless the authors think they
  should be dropped altogether.
- change milestone that we'll see a revised HTTP/1.x draft
  in December, but it won't be finalized until March. Goal
  is to submit it for Draft Standard.

Regards,

Larry

Received on Wednesday, 4 September 1996 02:48:31 UTC