RE: "end-to-end" headers in HTTP/1.1 [was Re: Mandatory]

From: Yaron Goland (yarong@microsoft.com)
Date: Fri, Apr 24 1998


Message-ID: <3FF8121C9B6DD111812100805F31FC0D02971236@red-msg-59.dns.microsoft.com>
From: Yaron Goland <yarong@microsoft.com>
To: "'David W. Morris'" <dwm@xpasc.com>, Henrik Frystyk Nielsen <frystyk@w3.org>
Cc: Ted Hardie <hardie@thornhill.arc.nasa.gov>, mogul@pa.dec.com, hardie@nic.nasa.gov, ietf-http-ext@w3.org
Date: Fri, 24 Apr 1998 12:01:45 -0700
Subject: RE: "end-to-end" headers in HTTP/1.1 [was Re: Mandatory]

I agree with David.

Common usage of the term "end-to-end" means client to server. It is
irrelevant what the HTTP spec did or did not mean by that term. Usage
defines meaning.

I remember when folks were still trying to save the word "hacker." That was
a lost cause as well.

		Yaron

-----Original Message-----
From: David W. Morris [mailto:dwm@xpasc.com]
Sent: Friday, April 17, 1998 10:40 AM
To: Henrik Frystyk Nielsen
Cc: Ted Hardie; mogul@pa.dec.com; hardie@nic.nasa.gov;
ietf-http-ext@w3.org
Subject: Re: "end-to-end" headers in HTTP/1.1 [was Re: Mandatory]




On Sat, 18 Apr 1998, Henrik Frystyk Nielsen wrote:
 
> - everything else which we often call end-to-end. 

I continue to disagree with the notion that we (the HTTP-WG community) use
'end-to-end' to refer to 'everything else'. There will be much less
confusion of the 'mandatory' specification uses a different term.

Dave Morris