[Baypiggies] Challenge/Response email systems

jennyw jennyw at dangerousideas.com
Sat Jun 17 02:12:11 CEST 2006


Anna Ravenscroft wrote:
> Lastly, on a less "practical" plane and more emotional one: my
> emotional reading of it as a signal is - "oh, this person wants to be
> left alone. Okay, I won't bother them anymore. " and avoid emailing
> the person in the future. This may not be the person's intent (and may
> or may not be a common reading), but it's something users of C/R
> systems may want to be aware of.
>   

Yeah, I think the real problem is that people who you want to hear from
might send you an e-mail, get a challenge, and decide not to bother.
Then you don't get an e-mail that you might really want to have
received, and whoever it was at the other end might get annoyed with
you. The fact that people on this list need to be convinced that
challenge-response is a good idea does not bode well for someone using a
challenge-response system.

I think Bayesian filtering works great. I use SpamBayes (written in
Python) and it's been very infrequent that I've had a false positive. In
fact, I'd say it's more likely that I'd miss more legitimate e-mail
without SpamBayes (by accidentally deleting legit e-mail myself) than
with it.  My maybe folder gets mostly spam with a few good messages, but
I'm willing to look through the few messages that get classified as
maybe.  I've heard that DSPAM works even better than SpamBayes, but I
use SpamBayes because it was so easy to setup and use (at first I used
it through IMAP; now I use it through a combination of IMAP for training
and procmail for filtering).

Jen



More information about the Baypiggies mailing list