Internetworkers: http://www.ibiblio.org/internetworkers/

[Steven Champeon <schampeo AT hesketh.com>]

on Fri, Aug 15, 2003 at 04:10:26PM -0400, Sil Greene wrote:
> 
> .:> Okay, spam is really going off the deep end now.
> .:
> .:It's just a hashbuster/bayes-poison. The idea is that if they stuff
> .:enough random garbage into the message, you won't be able to block it on
> .:the basis of its contents. You'll notice that there was probably a
> .:solicitation in there, too - the noise at the top (and usually, the
> .:bottom) is just there as noise to confuse Bayes filters and the like.
> 
> That makes sense, if all you're trying to do is defeat a filter technique.  
> Is that all this stuff is?  A grand experiment by spamsters to figure out 
> how to get the message through?  Seems like this technique would actually 
> hamper the spammer's efforts to _reach_ _his_ _audience_.  Or am I missing 
> something?

The bayes poison is usually stuffed into the MIME prologue, which is
not displayed by a MIME-capable newsreader, or into HTML as random cruft
set to the background color or hidden via stylesheets, etc. There are many
ways to hide the noise.

And yes, you'd think it might occur to some of them that delivery isn't
as important as comprehension and acceptance of the offer, but come on -
we're not exactly dealing with Harvard MBAs, here. (The spam-for-hire
folks who /send/ the stuff are clever, sure - but their customers are
largely the sort their families deny having spawned.)

-- 
hesketh.com/inc. v: (919) 834-2552 f: (919) 834-2554 w: http://hesketh.com
Book publishing is second only to furniture delivery in slowness. -b. schneier