"Ryan Malayter" <rmalayter(_at_)bai(_dot_)org> writes:
Here is a link to the analysis:
This is very useful, and most of what I was looking for, but I'd still
like to see an analysis done with hand-filtered email.
I agree, this is very useful information.
I found it interesting that, despite several comments to this list,
the SpamCop DNSBL came out looking very good. It blocked more spam
than any other DNSBL and yet its false positive rate isn't a huge
amount worse than the highly regarded CBL DNSBL.
The SpamCop DNSBL has been around for a long time and was one of the
first automated DNSBLs. Its algorithm has changed a lot and the
size of the reporting base for SpamCop has increased greatly. I
wonder how much of the bad reputation is due to old data/experiences.