It might be interesting for IAB to think about the estimated half-life
of well-known port numbers in the Internet architecture, since we've been
seeing
(1) firewalls that limit traversal based on port numbers,
(2) ISPs that have opinions about what services go where, based on port numbers,
(3) URL schemes that support alternate port numbers fairly easily,
(4) Mechanisms like SOAP that use HTTP as a substrate (ne: RFC 3205)
specifically
to avoid (1) and (2), and
(5) Servers running at alternate ports specifically to avoid (1) and (2)
Knowing what "port 25" means seems like something our children won't
understand...
... although they may wonder why everything EXCEPT web access is running over
port 80!
Spencer
-----Original Message-----
From: Theodore Ts'o [mailto:tytso(_at_)mit(_dot_)edu]
Sent: Thursday, February 27, 2003 8:33 AM
To: RL 'Bob' Morgan
Cc: IETF
Subject: Re: IAB policy on anti-spam mechanisms?
On Thu, Feb 27, 2003 at 12:41:41AM -0600, RL 'Bob' Morgan wrote:
Many sites, including my university, support STARTTLS+AUTH on the
Submission port (587, RFC 2476), which I believe is the recommended
service for clients to use to submit mail in any case (though not
well-supported among MUAs, to my knowledge), and also is
effective at
getting around ISP blockage of port 25. Of course if it
becomes very
popular the misguided ISPs will block it too.
Yup, the problem with well-known ports is that well-known port numbers
get either (a) blocked by misguded ISP's, or (b) transparently proxed
by misguided ISP's. Since I have no idea what sort of stupidity I
might encounter at various different hotel, conference, or 802.11
hotspot networks, it's more convenient for me to use a non-standard
port.