| From: | Brooks Davis <brooks@xxxxxxxxxxxxxxxxxx> |
| Date: | Thu, 27 May 2004 10:50:59 -0700 |
On Thu, May 27, 2004 at 10:32:27AM -0700, Stephen Paskaluk wrote: > On Thu, 27 May 2004 18:58:40 +0200, Erik P. Skaalerud wrote > > Matthew Dillon wrote: > > > I hate spam. Sigh. I disallow the more common Content-Type: > encodings > > > used by spammers, like multipart/alternative, but I've been allowing > > > multipart/mixed because a number of DFly posters use it to > attach patches > > > and such. > > > > > > -Matt > > > Matthew Dillon > > > <dillon@xxxxxxxxxxxxx> > > > > Matt, have you considered using a anti-spam solution that has > > a better chance of catching these? for example, > > www.mailscanner.info content-scans for sendmail, and it > > supports spamassassin, clamav etc. > > > > Erik > > I can't say if it's doable or not, but using TMDA to challenge posts > from non-subscribers seems almost ideal for preventing this. > > For anyone not familiar with TMDA, it's a mail filtering tool that will > send a simple challenge back to the sender (ie. "Hi, I'm TMDA, please > respond to this to ensure you're a real person who wished to post to the > DragonFly kernel list"), and also provides whitelisting (so not everyone > needs to be challenged). It has more features, but that pretty much > sums up how it would be useful to keep spam off of a mailing list. > http://tmda.net is their home page. TDMA f'ing sucks for viruses as does anything that responds to messages containing viruses. It confuses two entirely different tests. What it actually tests is "is this a valid address?". What is wants to test is "is THIS MESSAGE from a real person?". As a result, I respond to all TDMA and similar challenges even if I didn't send the message so my address won't end up on a list of spammers. -- Brooks -- Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4
Attachment:
pgp00008.pgp
Description: PGP signature