From: Ted Gavin <tedgavin(_at_)newsguy(_dot_)com>
...
First, I'm no Microsoft advocate. I was a mail administrator for some
number of years in the course of which, I had to deal with Microsoft
messaging products.
As such, you should look around at other systems and perhaps even
read what people have repeatedly said here to discover that your
the problem statement is entirely off the mark.
Those persons who are responsible for managing Microsoft Exchange
implementations should know that Out-Of-Office responses, as well as
anti-virus application auto-notifications can be given permission to
send to the Internet, just as they can be DENIED permission to send to
the Internet. ...
What is that "send to the Internet" business? The vacation message
problem has nothing to do with "sending to the Internet", but with
not doing as has been said many times here.
Those who have ignored the repeated descriptions of the bug here could
look for the problem elsewhere. They could start at
http://www.FreeBSD.org/cgi/man.cgi and then
http://www.FreeBSD.org/cgi/man.cgi?query=vacation&apropos=0&sektion=0&manpath=FreeBSD+4.2-RELEASE&format=html
and which contains
No message will be sent unless login (or an alias supplied
using the -a option) is part of either the ``To:'' or
``Cc:'' headers of the mail.
The fault is not in the software. The fault is with the users and the
administrators.
You are wrong. The fault is first in the software for not doing
what has been repeatedly described here. The fault is with the
users and administrators second for unthinkingly accepting a silly
description of the problem and doing as people in Redmond do and
not paying attention to anything that didn't originate in Redmond.
...
I agree. Not everyone's software does it, but not everyone's software
gives people an equal level of administrative control. If you
implement Exchange, you're not looking for the complexity or
specificity that sendmail or some other product offer you. You're
basically purchasing McMail. You're looking for whatever benefit you
feel that Exchange provides. At that point, what any other products do
becomes irrelevant.
Again, for the umpteenth time, this bug has nothing to do with sendmail
or any MTA, as most people distiguish MTA's from MUA's. From what some
Microsoft advocates have said, that is even the case with this Microsoft
bug, that the out-of-office abuse system is outside the Exchange MTA.
...
So you are essentially stating that persons who work for companies
that use Microsoft Exchange Server are no longer welcome in IETF
participation unless they use a mailer and get another account of
their own? If so, just come out and say it.
I'm saying that people who are too lazy or witless to pick software that
does not cause them to make pests of themselves have no place trying to
develop network protocols. Those who are cannot be bothered to pay
attention to years of repeated descriptions of a serious problem in their
software but instead merely repeat the party line of their chosen vendor
are worse than useless in any design effort, including the IETF's.
So, since Exchange users
will now have to pay for another mail account, why not just let them
use Exchange, but charge them a membership fee and use the cost to
subsidize a moderator?
That is more of the all too common IETF-as-a-social-club-and-tutorial
nonsense.
Your point, while making it nice and convenient for everyone, is
directly contrary to the notion that the internet is for everybody.
The Internet is not for everybody, but only for those who pay their own
freight. Besides, I explicitly said that this bugware should be banned
from the Internet but only banned from subscribing to IETF lists. The
IETF is not the Internet.
Inasmuch as IETF exists under the ISOC umbrella,
That is quite wrong.
and since there are
both valid and valued contributors to IETF who use Exchange (I was
once one, thankyouverymuch), perhaps there's a more palatable way to
do this.
The notion of "*valid* contributors to the IETF" sets the hackles of anyone
who thinks about it. "Valued contributors" are something else, although
that has related dubious aspects. Those who claim to be valued IETF
contributors tend to be more valued in their own eyes than anyone else's.
Is the vacation feature of the package separate from the
main thing?--perhaps but that's irrelevant.
I disagree. I think it is entirely relevant It's only different
because it's a client-driven feature as opposed to a server-driven
feature. ...
Nonsense. How some marketeers decide to package things does force
the world to accept abuse. "Tying" and "packaging" are forces of
nature only for consumers who let any single vendor's marketers do
their thinking for them.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com