Suresh Ramasubramanian ops.lists at
Mon Feb 22 02:20:22 UTC 2010

On Mon, Feb 22, 2010 at 12:08 AM, Joel M Snyder <Joel.Snyder at> wrote:
> but the false positive count jumped by 112 messages per 10,000 (because
> APEWS was somehow having a lousy month).
> In general, the more reputation services you include, the more likely it is
> you're going to have false positives.

Christ.  You pick APEWS as a reputation filter.. and then even bother
to *count* the false positives?

That's not a list that's particularly designed to minimize FPs, to put
it very mildly.

Suresh Ramasubramanian (ops.lists at

More information about the NANOG mailing list