W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1996

Re: draft-ietf-http-state-mgmt-03.txt

From: Benjamin Franz <snowhare@netimages.com>
Date: Thu, 1 Aug 1996 11:00:49 -0700 (PDT)
To: Bill Sommerfeld <sommerfeld%apollo.hp.com@hplb.hpl.hp.com>
Cc: dmk@bell-labs.com, montulli@netscape.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <Pine.LNX.3.93.960801103437.6506A-100000@ns.viet.net>
On Thu, 1 Aug 1996, Bill Sommerfeld wrote:

> > Implementation issue and *IMPOSSIBLE* to enforce. 
> 
> Take a look at the RFC1123 (host requirements) some time; you'll see
> subsections there like "TELNET/USER INTERFACE" "FTP/USER INTERFACE".
> There are a fair number of MUSTs in there..

All of the MUSTs and MUST NOTs in RFC1123 are protocal issues which have
the clear potential to break interoperability if not addressed. Even the
NVT issue is interoperability - not user display.  Failure to display to
the user when a cookie is set will not break the *operability* of the
protocal at any level. Failure to negotiate a minumum terminal type in
telnet would result in option negotiation breaking down fatally -
something that would clearly break interoperability. RFC1123 is *not* an
example of implementation issues coded into the MUSTs and MUST NOTs. 

> Or the ipsec rfc's; they specify similar issues (e.g., the user MUST
> be able to do manual keying).

You are going to have to be a lot more specific about which RFC you are
referring to. Context is everything and there are a LOT of RFCs.

-- 
Benjamin Franz
Received on Thursday, 1 August 1996 11:17:42 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:05 EDT