W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > January to April 1997

Re: 1xx Clarification

From: Larry Masinter <masinter@parc.xerox.com>
Date: Wed, 16 Apr 1997 13:36:39 PDT
Message-Id: <33553857.68C@parc.xerox.com>
To: Yaron Goland <yarong@microsoft.com>
Cc: "'doug_crow@cacheflow.com'" <doug_crow@cacheflow.com>, http-wg@cuckoo.hpl.hp.com
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/3085
> I believe we should leave this issue open. I feel that Roy's attitude is
> best, "If you don't understand it, then dump it." If others need to
> solve this problem they are free to add headers, new 1xx messages with
> bodies, etc.

I don't have a problem with adding something which has
a well-defined meaning with the caveat that some recipients
might not "understand" it. The problem is that allowing for
1xx responses outside of the context of a request would
make them ambiguous where they are not now ambiguous.

Now, if a client has sent two pipelined requests, gets one
response, and then a 1xx response, it would be ambiguous as to
whether the server intended the 1xx response to be associated
with the second request or to be interpreted outside of
the context of that request.

If we (continue to) disallow 1xx responses outside of
the context of a specific request, then the context of
the response is not ambiguous.

Received on Wednesday, 16 April 1997 23:52:57 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:40:19 UTC