Spam control -- low tech rules! (was: BitTorrent and Comcast?)

[Fred]

On Wed, 2004-09-29 at 22:43, Benjamin Scott wrote:
...
>   What's nastier is that spammers are now starting to include "real" text in
> their spam.  They seem to be either using public domain works (from Project
> Gutenberg, I'll bet), or fragments of stuff harvested from the
> web/Usenet/something.  Since this all that stuff is, in effect, the same
> language real people use (spammers are not people), filtering on word
> frequency and sentence structure alone will not work.  Yuck.

Nothing but nothing beats the power of whitelisting. I use it all the
time now.

I know, I know -- if a legit contact not on your whitelist tries to
contact you, you're hosed.

But I even found a "low-tech" workaround for that, which will work
splendidly -- at least until the spammers catch up:

My technique of putting an identifying string (like my "[hey]") in the
subject can be extended to the web page. Just fill in the subject line
on your mailto: and filter for that on reception. Viola! current
spambots drop the subject, thus they can be ignored. Works like a charm.

Of course once this little cute hack is in widespread use, the spammers
will exploit it. But for now enjoy nirvana.

Ah, I just thought of a way to even foil spammer's catching up with this
technique -- still fairly low-tech. Generate a unique, one-time-use,
same-day expire token to place in the header.

Better still, generate the mailto: with some Javascripting.

There is *always* a low-tech means to thwart high-tech. At low expense
to ourselves, we can make it far more expensive for them to operate. And
I'd love to see spammers try to work around whitelisting. No amount of
word stuffing will fool that! :-)

-- 
Fred -- fred at lrc.puissante.com -- place "[hey]" in your subject.
The mass of humans on planet Earth -- regard them as the ebbing 
seas in the winds of change. They ebb, they flow, they know not 
where to go.