Am 25.01.2016 um 15:25 schrieb Axb:
On 01/25/2016 03:17 PM, Reindl Harald wrote:masschecker is not worth an argument in 2016: 02-Jan-2016 00:22:49: SpamAssassin: No update available 03-Jan-2016 00:32:25: SpamAssassin: No update available 04-Jan-2016 00:03:32: SpamAssassin: No update available 05-Jan-2016 01:45:11: SpamAssassin: No update available 06-Jan-2016 01:28:29: SpamAssassin: No update availableI don't find your masscheck contribution in the logs...
you simply would not want my anonymized and header stripped samples contribute to the masscheck
your typical reaction if someone points out a potential problem "why don't you do and no i don't see something problematic", nothing new.........
if it would be only about you i could just disable rules or score them different and be done, consider that i care also about other users which are not you
You are aware that you can run your own personal local score generation process based on your corpus, right? All the code is available in SVN
i don't need it and it won't work with anonymized and stripped samples and they do their job pretty well (yes i know that you don#t care about bayes)
BAYES_60 354 1.23 % 7.70 % (OF TOTAL BLOCKED) BAYES_80 380 1.32 % 8.27 % (OF TOTAL BLOCKED) BAYES_95 308 1.07 % 6.70 % (OF TOTAL BLOCKED) BAYES_99 3541 12.34 % 77.11 % (OF TOTAL BLOCKED) BAYES_999 3173 11.05 % 69.09 % (OF TOTAL BLOCKED) SPAMMY 4583 10.21 % 99.80 % (OF TOTAL BLOCKED)i just find it amusing that you come up with "there is no problem, masscheck will fix that" while you *know* that masscheck don't fix anything currently
signature.asc
Description: OpenPGP digital signature
