RE: Mandatory

From: Paul Leach (paulle@microsoft.com)
Date: Thu, Apr 09 1998


Message-ID: <5CEA8663F24DD111A96100805FFE6587031E3DAA@red-msg-51.dns.microsoft.com>
From: Paul Leach <paulle@microsoft.com>
To: "'David W. Morris'" <dwm@xpasc.com>, hardie@nic.nasa.gov
Cc: ietf-http-ext@w3.org
Date: Thu, 9 Apr 1998 13:44:13 -0700 
Subject: RE: Mandatory



> -----Original Message-----
> From: David W. Morris [mailto:dwm@xpasc.com]
> Sent: Wednesday, April 08, 1998 10:29 AM
> 
> On Wed, 8 Apr 1998 hardie@nic.nasa.gov wrote:
> 
> > (Henrik)
> > I have two points here: First, I disagree that HTTP defines end-to-end
to
> > mean from the user agent to the origin server. For example, the text in
> > 13.5.1 says
> 
> Perhaps this reading might hold legally, but the understanding I've had
> from following the HTTP mail list is that end-end means 
> client to origin server. I too had/have a problem with use of end-end in
the mandatory
> draft. It was confusing at best.

I think that it should be noted that "end-to-end" means "client to server"
not "client to origin server" and that proxies are servers too. But I think
the use of "end-to-end" should mean what the HTTP spec says it means, even
if up until now 99.99% of the "ends" have been origin servers or clients.
(And, AFAIR, the mailing list never explictly said "end to end means client
to origin server" in a context where they were trying to speak
definitively.)

So I'd leave it the way it is in Mandatory. HTTP is so widespread because
people take creative advantage of things that are implicit in the protocol.

Paul