On Wed, 12 Mar 2003, Vernon Schryver wrote:
Consider the RBL+, SPEWS, and SBL as models for how popular a favorite
spam defense might be after several years. All of those are said to
have significant good effects immediately and if everyoone used any
one of them, the spam problem would be solved overnight. They don't
require mail senders or other mail recipients to do anything. All of
them are trivial to install. SPEWS and the SBL are entirely free
other than quite modest per mail message bandwidth and other costs.
The monetary costs of the RBL+ are less than many proposed alternatives.
The RBL and SBL are said to have very low false positive rates. They
are all reported to be at least 20% and according to some more than
70% effective (spam rejected divided by total spam). In summary,
all three are exemplars of what many spam defenses would like to be.
Just some figures for this, on my home system (which I admit isn't
statistically significant, but we don't use these DNSBLs at work) the SBL
stopped 7% of my total spam this week, wirehub (open proxies) stopped 32%,
and monkeys.com (another open proxies list) stopped just over 1% and ORDB
stopped 1%. I have a feeling wirehub beat monkeys.com simply on speed, and
so looks better, but there should be no overlap with the other lists.
The rest was stopped by a combination of different content filters (and a
few sneaked through the net). Oh, and zero false positives - these DNSBLs
have very low false positive rates generally.
This doesn't prove much except that different people get different results
with their DNS blacklists.
[This has nothing to do with why you posted the above I know, but people
earlier were asking for some figures so I thought I'd post some].
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg