W3C home > Mailing lists > Public > ietf-http-wg@w3.org > October to December 1996

Re: REPOST (was: HTTP working group status & issues)

From: Larry Masinter <masinter@parc.xerox.com>
Date: Mon, 30 Sep 1996 20:03:47 PDT
To: FisherM@is3.indy.tce.com
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <96Sep30.200347pdt."2759"@golden.parc.xerox.com>
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/1671
In an offline discussion with Foteos Macrides, we explored the various
pros and cons of REPOST vs. Idempotent: yes. I'm not sure we've
reached closure yet, but there was a side topic that I thought I'd
bring up:

One of the concerns about bookmarking 'POST' requests, whether
idempotent or not, whether the client warns or not, is that some POST
requests contain passwords or other 'hidden' information.

Even if a POST result were identified as idempotent, clients might not
want to quietly save all of the input data on a hotlist in such
circumstances.

Maybe there's more about a POST result than just 'idempotent' that's
actually useful, as in 'is it safe to bookmark the request'
independent of 'is it safe to redo the POST'.

Larry
Received on Monday, 30 September 1996 20:07:43 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 2 February 2023 18:43:00 UTC