John Levine <johnl(_at_)taugh(_dot_)com> wrote:
Under the current setup, any domain that has an A record is presumed
to be a mail domain,
Let us pause for a moment, and consider that statement.
It is correct in that it talks of "domains". DNS queries are for
Resource Records for a "domain". And email (excepting the @[IP] case)
is intended to go to a "domain."
It's slightly less correct to say "is presumed..." The domain
might also have an MX RR -- in which case the presumption is denied.
The "presumption" means that any SMTP server asked to send to that
domain must try to open a port 25 connection, and continue to
"presume" until that attempt times out.
and if it's not, there's no good way to advertise that fact.
Clearly, waiting for a timeout is not "a good way".
(Current crocks include running a stub server that rejects everything,
and the MX 0 . proposal.)
It would seem that neither of these are likely to be done by admins
that now can't be bothered to publish a MX RR because "it's working
just fine as it stands."
In a world where the fraction of mail that is spam is 95% and growing,
Has anybody else noticed that we were saying "80% spam" just a year
One can argue whether the spam volume is doubling every six months
or whether it takes a year... Let's assume it takes a year: in the
ten-year expected lifetime of 2821bis, that would project to 99.995%
the load due to unwanted mail to non-mail domains is significant.
... meaning that today's 20:1 would become 20,000:1. :^(
As Keith recently pointed out, since it is no longer 1983, the vast
majority of names with A records are not mail servers nor are they
domains that occur in valid mail addresses.
I suspect we'll see Zipf's law at work (frequency of occurance is
roughly inversely proportional to the rank position), with roughly
100 domains accounting for at least half the email, and nobody bothering
to set up email beyond the 100,000th domain. ISC.org estimated roughly
500 million hosts on the Internet at the start of 2008. Thus, we seem
to be very roughly at one in 5,000 right now.
As other people have pointed out, a no-mail default is far more robust
than the current default, since a fair number of non-mail hosts turn
out to be running some sort of default SMTP server which will swallow
and lose mail.
I can't estimate this effect. Sorry...
Or even if they aren't running an SMTP server, it can take a week for
the message to time out and bounce.
I can testify to that.
It'd be much better for such messages to fail immediately so the
sender will notice and can do something about it.
This does seem more user-friendly to me.
I continue to gaze in amazement at the postings I see which deny
that we can or shuold do anything about these facts.
John Leslie <john(_at_)jlc(_dot_)net>