On 2008-04-05 01:30:19 -0000, John Levine wrote:
ps. While surveys are always useful, I'll repeat the observation
that changing the query model for SMTP is out of scope.
Dave, you're assuming your conclusion.
What we're trying to settle here is whether the existing query model
is synthesize from A or synthesize from any address record.
Or synthesize from neither. That's the original algorithm in RFC 974: If
the list of MX records is empty, an MX record containing the domain on
the right side is synthesized. Whether an A or AAAA (or other) record is
present, is irrelevant at this stage. Address lookup only happens later
and will fail in the absence of an A or AAAA record regardless of
whether the hostname came from a real MX record or not.
(Yes, I know that RFC 974 is historical. But the language in RFC 2821
doesn't make a difference in an IPv4-only world, and I think it
needlessly complicates the algorithm).
_ | Peter J. Holzer | It took a genius to create [TeX],
|_|_) | Sysadmin WSR | and it takes a genius to maintain it.
| | | hjp(_at_)hjp(_dot_)at | That's not engineering, that's art.
__/ | http://www.hjp.at/ | -- David Kastrup in comp.text.tex
Description: Digital signature