The paper clearly states that the study does not attempt to
be authoritative and the main conclusion is that someone should
do a proper study.
In fact I suspect that the more interesting questions would be
about the way that the spam changes over time, the number of
distinct items of spam etc.
Phill
-----Original Message-----
From: Jim Youll [mailto:jim(_at_)media(_dot_)mit(_dot_)edu]
Sent: Saturday, April 26, 2003 8:40 AM
To: Hallam-Baker, Phillip; Vernon Schryver; asrg(_at_)ietf(_dot_)org
Subject: RE: [Asrg] A New Plan for No Spam / Velocity Indicator
At 20:15 -0700 4/25/03, Hallam-Baker, Phillip wrote:
Like 89 is not a big number, but it is a lot bigger than 0.
But it really depends on the quality of the sample... Intuitively it
"feels" like you'd need more than that to really represent the
situation, but looking at the sample by itself, there's no way to
know.
Perhaps this is one of those situations where it may be necessary to
take a very large sample that helps to determine how big a sample
size is really needed, and to judge the quality of those already
in-hand.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg