[lug] Stopping the New Generation of Spam

[Sean Reifschneider]

On Sun, Dec 03, 2006 at 12:19:02PM -0700, Bill Thoen wrote:
>Over the last 2-3 months I've been getting a *lot* more spam than ever

I've noticed a lot more spam over the last week.  We have pretty tight spam
protections, but I went from getting maybe 4 per day through to my main
box, to getting around 10.  I figure this is just the holiday spam season.

The random words messages are killers for Bayesian filters.  We aren't
really using bayes anyway, but these really go towards poisoning bayesian
filters.  What you really want to do is OCR the GIF image that's included
with the random text, and use that OCRed text to feed to the filter, and
also to filter against.

I've written up some journal posts on what we are doing and planning for
anti-spam, if you want more details:

   http://www.tummy.com/journals/entries/jafo_20061203_134948
   http://www.tummy.com/journals/entries/jafo_20061203_143406

Sean
-- 
 Her eyes were like two brown circles with big black dots in the center.
Sean Reifschneider, Member of Technical Staff <jafo at tummy.com>
tummy.com, ltd. - Linux Consulting since 1995: Ask me about High Availability
      Back off man. I'm a scientist.   http://HackingSociety.org/