Terry,
I'm so grateful we've got a statistician-like brain (yours) on this
list. However, dare I say that your open-ended question below will
prompt only guesses from the many on this that love a good debate but
are oh-so-slow to step up and get some real work done.
So, I say this: you raise a good question. Lets research it. But lets
not debate it based on " my years and years of using email".
To all: Who will step forward and assist us w/ this research? We've laid
out the rough sketch before. If you want to help, pls let us know.
But if all we're going to do is debate based on guesses, do it on
slashdot and give us the link.
Peter
-----Original Message-----
From: Terry Sullivan [mailto:terry(_at_)pantos(_dot_)org]
Sent: Saturday, August 09, 2003 12:44 PM
To: asrg(_at_)ietf(_dot_)org
Subject: RE: [Asrg] RE: 2.a.1 Analysis of Actual Spam Data -
Titan Key reduces spam attacks
Two things:
1) To the extent that my ignorance of mechanics of
SMTP may have caused a certain... off-topic drift
over the last couple of days, I offer a sincere
apology to the list as a whole.
2) I'm still troubled by an anomaly in the notion
that 550 -> less spam.
All other things being equal, I'm usually the
first one on the "We need more data" bandwagon.
But it seems to me that there is a very large
base of information (though not true "data")
pertinent to the question and ready-to-hand.
As the 'Net user base has grown over the last,
say, 5-10 years, I daresay that the number of
550s thrown between servers has also grown. I
would hazard a guess that the number of 550s
has, at the very least, grown as some (unknown)
monotonically-increasing function of the number
of users.
(In the only case for which I have detailed
knowledge, there are at least 10 different email
addresses that legitimately throw a 550 for me,
because I've changed ISPs, left the company, etc.
There may be more, but there are at least 10.)
Now, for all I know, a 10:1 ratio may very well
be atypical [high or low]. But the thing that
keeps buggin' me is this: if there is a robust,
substantive inverse relationship between 550s
and spam volume, how has it managed to elude
detection this long, given the sustained growth
in 550s? (Call me cynical, but I sorta doubt
that it's the result of great QA on those
every-email-address-on-the-planet CDs.)
(Though I might be wrong...) :-)
- Terry
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg