W3C home > Mailing lists > Public > uri@w3.org > October 2006

Re: 4395 historical schemes

From: Tom Petch <nwnetworks@dial.pipex.com>
Date: Mon, 30 Oct 2006 10:09:52 +0100
Message-ID: <02b801c6fc2e$a49ebf00$0601a8c0@pc6>
To: <uri@w3.org>, "Frank Ellermann" <nobody@xyzzy.claranet.de>

I think your question too general to be taken as purely a URI one.

Where there is an application level dialog that can negotiate the use - or not -
of TLS (or SSH or ....), then only one port is needed, only one scheme is
needed.

Where there is no such dialog, then we are likely to need two ports and two
almost identical URI schemes in perpetuity.

syslog is using TLS, has no such dialog, will ask for a new port, could then
need two almost identical schemes.

Tom Petch


----- Original Message -----
From: "Frank Ellermann" <nobody@xyzzy.claranet.de>
To: <uri@w3.org>
Sent: Friday, October 27, 2006 4:27 PM
Subject: 4395 historical schemes


>
> Hi, if an URI scheme x is related to a protocol y using port z, and
> if there's an almost identical scheme sx using y over SSL port sz,
> is it a good idea to register scheme sx as "historical" in the IANA
> consideration of a revised x spec. ?
>
> I have some necessary pieces (x, y, z, sx, sz) plus an old I-D for
> sx (= snews) plus a new RFC (4642) declaring sz (563) as bad idea.
>
> RFC 4395 says in chapter 4 "In some circumstances, it is appropriate
> to note a URI scheme that was once in use", and from that I _guess_
> that it's in many cases not appropriate.
>
> Frank
>
>
>
Received on Monday, 30 October 2006 15:20:33 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:10 UTC