W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2010

Re: Questions (errata?) about caching authenticated responses [#174]

From: Jeffrey Mogul <Jeff.Mogul@hp.com>
Date: Wed, 02 Jun 2010 10:56:23 -0700
Message-Id: <201006021756.o52HuNiF032696@pobox-pa.hpl.hp.com>
To: Mark Nottingham <mnot@mnot.net>
cc: Henrik Nordstrom <henrik@henriknordstrom.net>, HTTP Working Group <ietf-http-wg@w3.org>, Duane Wessels <wessels@packet-pushers.com>, JeffMogul@acm.org, Jeff.Mogul@hp.com
   Picking this issue back up and CC:ing Jeff for his
   recollections (see <http://trac.tools.ietf.org/wg/httpbis/trac/ticket/174>
   for background).

Sorry, I don't have any clear recollections.  I did find some
relevant emails that I had sent:

(email 1) (Thu, 02 Jan 97 10:55:03 PST) in private (not to the
mailing list) that contains these paragraphs:

   The current issues in the HTTP WG related to caching, for HTTP/1.1
   in particular, are limited to a few issues that seem particularly
   pressing:

        (1) there is an apparent bug (or at least a "poor design
        choice") in the current draft re: the "proxy-revalidate"
        Cache-control directive, probably my fault.  We will likely
        end up defining a new directive to avoid the problem, and
        I've volunteered to write up a brief draft on this.
        
        (2) Several related documents will need to be revised to
        deal with this bug.

(email 2) (Mon, 21 Apr 97 19:19:18 PST) to Larry Masinter and
Jim Gettys (also not to the entire list):

       I thought the http-versions draft was ready, but looking at

       ftp://ietf.org/internet-drafts/draft-mogul-http-revalidate-00.txt

       I see that contains two alternatives, and clearly cannot
       progress until we've reached consensus on which one.

   You didn't read it carefully enough.  Section 3 lays outs
   a number of alternatives, to show that they were considered,
   and includes an analysis to show what their benefits and
   drawbacks are ... but does not propose any of these directly.

   Section 4 precisely defined a specific proposed change.  This
   is the "normative" part of the proposal, i.e., a set of edits
   to be applied to RFC2068.  If Jim finds anything in section
   4 ambiguous, then I need to fix this; otherwise, not.

       Were you intending to release a new draft so I can 'last call' it?

   The only glitch is that Roy, during a hallway chat at the WWW6
   conference, casually suggested that he wanted me to replace the
   term "proxy [cache]" with "shared cache".  He hasn't ever sent
   email to this effect, but I have no objections to making this
   change.  I consider it to be an editorial issue (clarification)
   and it seems pointless to spin another I-D version just to clarify a
   single word.

I don't remember what Jim did as a result of this exchange.
(See http://tools.ietf.org/html/draft-mogul-http-revalidate-00 for
the draft mentioned above.)

Also, Duane sent an email on 19 Jul 2000 to the http-wg list
   Subject: Questions (errata?) about caching authenticated responses
which brings up some questions about proxy-revalidate; see
<http://ftp.ics.uci.edu/pub/ietf/http/hypermail/2000/0142.html>

That's all I can remember at this point, but I think a careful
reading of http://tools.ietf.org/html/draft-mogul-http-revalidate-00
might clarify much of the history.

-Jeff
Received on Wednesday, 2 June 2010 17:57:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:20 GMT