W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1994

Re: Comments on the HTTP/1.0 draft.

From: Larry Masinter <masinter@parc.xerox.com>
Date: Tue, 13 Dec 1994 01:47:22 PST
To: mvanheyn@cs.indiana.edu
Cc: fielding@avron.ICS.UCI.EDU, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <94Dec13.014729pst.2757@golden.parc.xerox.com>
You know, Internet protocols don't define "conforming implementations"
but rather "conforming behavior on the net"; you might want to define
a "conforming implementation" as "an implementation that never
exhibits non-conforming behavior", but of course, that seems to be a
site/configuration issue as much as anything.

The conformance requirement on data streams is that senders only send
conforming data streams to recievers (unless somehow the protocol
allows for the recievers to indicate that they're willing ot accept
non-conforming data streams.)

We should talk about senders and recievers rather than servers and
clients, because when clients send data to servers, the roles are
reversed and the conformance requirements reverse.

No sender should send an unregistered type to a reciever that hasn't
indicated a willingness to accept it. The 'practice' that allows HTTP
to be somehow more free with MIME types is that the protocol allows
for negotiation.

That "text/html" wasn't registered in the official MIME registry is
somewhat of a red herring: it was certainly registered in the web
documents. 
Received on Tuesday, 13 December 1994 01:49:16 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:09 EDT