[SpamCop-List] Re: Er... purpose of .spam

[Mike Easter]

Skiwi wrote:
> ASIDE  - if it is "non-manually steered to a junk pile" how do you
> allows for false positives. Even for my quick reporting, I try to
> carefully look at all subject lines and if I am even a *little* in
> doubt I will preview it's source...

When I'm reading about a 'genuine' spamfighter 'reading' spam or its
subjects such as you've described, I'm not 'worried' or upset or
anything about thorough examinations.

The problem is that there is a *HUGE* population of 'spam whiners' and
'spam badmouthers' who are actually curious spamreading spambait.  I
would like to first convert all of the spamreaders of that ilk into spam
non-readers.  I would much rather have fewer spam whiners and
badmouthers and more spam non-readers.

Re false positives.  In a good 'anti-reading' program there /should/ be
false positives, either because some kind of unknown has been manually
'steered' into the junk because it didn't clearly 'jump out at' the
recipient as good mail or for some other reason.  The ideal filter would
be IDing almost all of the spam and steering it.  'Unknowns' getting
into the Inbox would be manually put into junk [causing some false
positives that way] and everything in the Junk would be examined by its
source /first/ - looking for clues of bogosity and spamosity from the
'inside'.

It's part of the game of scoring points for the spammer if you
unknowingly 'open' a spam.  There are no points for the spammer if you
happen to examine the source of a good mail because it wasn't obviously
good -- that is, it is better in the point scoring game to err on the
side of suspicion.

-- 
Mike Easter
kibitzer, not SC admin