| Main Archive Page > Month Archives > spamassassin-users archives |
>On 12/21, darxus@chaosreigns.com wrote:
>> But your question may actually be a very good one, which might, in the
>> end, come down to: Why isn't RP_MATCHES_RCVD flagged as mutable and
>> having optimal scores generated based on masscheck data?
On 21.12.11 13:37, darxus@chaosreigns.com wrote:
>I take that back, I think all the scores in 72_scores.cf are mutable /
>subject to regular optimal re-scoring. So the answer is some combination
>of:
>
>1) Because you don't contribute data via masscheck, so we don't have the
> benefit of your data to improve the usefulness of scores.
should we contribute spam that hits RP_MATCHES_RCVD ? :)
(unluckily, not mentioning mailing lists, I'm not getting much of mail)
>2) Because the re-scorer thinks this rule is useful enough in blocking
> spams that it's worth the occasional missed spam it causes. Would you
> rather get as many spams as you do, and have them all hit
> RP_MATCHES_RCVD, or not use RP_MATCHES_RCVD, and miss twice as many
> spams? "Twice" is not intended to be a solid number, just an example of
> how the re-scorer might be thinking. Since it always aims for 1
> wrong non-spam in 2500, so what varies by rule usage is the number
> of spams it catches.
Actually, it has negative score, so the rule unblocks ham... and causes
FNs.
--
Matus UHLAR - fantomas, uhlar_at_fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
My mind is like a steel trap - rusty and illegal in 37 states.