W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1995

Re: Possible optimization to State-Info proposal

From: <Harald.T.Alvestrand@uninett.no>
Date: Fri, 25 Aug 1995 10:27:43 +0200
Message-Id: <199508250827.KAA01366@dale.uninett.no>
To: Shel Kaphan <sjk@amazon.com>
Cc: bobwyman@medio.com, dmk@allegra.att.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
You know, the "HTTP/1.0 is state of the world" idea has some good sides.
Now we can say, with a fair degree of sense, that "HTTP/1.0 says that
GET and HEAD are mostly idempotent", while HTTP/1.1 says that
"GET and HEAD are idempotent, and so can be cached, UNLESS the following
parameters are put on the document by the server...."

Since 1.0 is descriptive, it has to say that the idempotency of GET
is "by convention", while 1.1 is prescriptive, and can cast the idempotency
into stone (rather soft stone, but stone nonetheless).

             Harald A
Received on Friday, 25 August 1995 01:30:02 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:26 EDT