[ietf-clear] CLEAR Charter

[Douglas Otis]

On Thu, 2004-09-30 at 09:52, Ryan Ordway wrote:
> On Thu, 2004-09-30 at 07:54, John Levine wrote:
> > >|From CSV: "A sending SMTP client MUST supply a published
> > >|domain name as the parameter to an SMTP HELO or EHLO." If
> > >|that's consistent wiwth applicable RFCs, then I support
> > >|Glube's text.
> > 
> > I don't write RFCs, but I'm pretty good at reading them.
> > 
> > Section 4.1.1.1 of RFC 2821 says that the argument to EHLO or HELO is
> > the fully qualified domain name of the host, or a bracketed literal IP
> > address if the host doesn't have a useful name, e.g., it's in a
> > dynamic pool.  RFC 821 just says the argument is the domain name and
> > doesn't say anything about bracketed IP addresses.
> > 
> > We should make it explicit that CSV doesn't support hosts that don't
> > have names and isn't applicable to hosts that HELO or EHLO with a
> > bracketed literal.
> 
> 	Should there be anything done about bracketed IP
> addresses? The goal would be to get people to use
> fully qualified names for HELO/EHLO, should we react
> negatively to those who don't? Or should it be a neutral
> reaction? 

John Leslie and myself created a few paragraphs on this issue during the
creation of the proposal with cautions about brackets and ending periods
etc.  Dave Crocker argued successfully to keep the specifications to the
minimum needed for implementation. Going down the road of explaining all
these details did not add to the proposal.  I think Dave is right about
this point.  Less is more.

-Doug