From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Tue Nov 09 1999 - 05:40:10 CST
In <19991108130956.M9805@demon.net> "Clive D.W. Feather" <clive@demon.net> writes:
>Charles Lindsey said:
>> There might be some benefit in a convention whereby, if a single cancel
>> message was directed only against the multiple instances of a single spam
>> (i.e. those that contributed to its BI) then you chose a single Message-ID
>> from them according to some agreed algorithm (the least, for example) and
>> use that to apply the $alz rule to.
>NO !
Indeed, I was not seriously suggesting it, unless the spam cancellers
themselves come up with a scheme.
>What if there are 21 articles meeting the criteria for cancelling, but each
>canceller only saw 20. They are likely to each end up with the same ID even
>though the cancels differ.
Actually, I think my scheme would have got it right for 20 of them - but
never mind.
I think the point we have arrived at is that the spam cancellers, and
others, do not want multiple cancels in the header (the single cancel will
remain, of course). So I have removed that feature.
The question then arises as to whether we now want an _additional_ block
cancel (with nocem-like syntax). The alternative is to leave the spam
camcellers to continue as they do, or to push for wider implementation of
nocem-on-spool. My present inclination is to leave it alone (unless the
nocem people specifically ask us to take nocem on board).
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Email: chl@clw.cs.man.ac.uk Web: http://www.cs.man.ac.uk/~chl Voice/Fax: +44 161 437 4506 Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5