Slashdot Mirror
Eliza for Spam
Saint Aardvark the Carpeted writes "Check this out for sheer genius...This guy has posted to Perl Monks a script that uses the Perl Eliza module to respond to spam. Check it and contribute your suggestions for improved vocabulary." The downside of course is that spammers never set their reply correctly (which I think is forgery, and should be treated as such) so this is probably more academic then useful, but its definitely funny.
? just gave me an idea.
The downside of course is that spammers never set their reply correctly (which I think is forgery, and should be treated as such)
hy not have mail systems check back with the return address to verify it as sent / approved to be sent from that address. Then label the mail accordingly - approved by labelled sender or not. Then receivers can choose whether they want unverified mail or not, in advance of receiving it. Hmm. Of course this could be detrimental to anonymous communications hich have they own set of (very) useful qualities. (Specially against corrupt governments.)
This guy has a lot of interesting ideas. I would think it would be fairly simple to write a program that did the following things:
1: If a SPAM message contains a form, you forward it to the program (which would create a local SMTP server to receive messages only from you). It would fill the form with random data, but somewhat intelligently, by parsing the HTML and figuring out what is valid for the form.
2: If a SPAM message goes to a web site with a form, go to that form in your browser. Run the anti-spam program and click a button to auto-fill the form. You could configure it to fill out the form multiple times, in the background, without user intervention.
3: If the SPAM message contains a link to geocities, or other free web hosting services, forward the e-mail to your anti-spam program's local SMTP server. It will grab the link and then forward the e-mail, header, and link to the appropriate abuse@ address.
There are probably other things like this that could be done. If someone wrote this program, I'd use it religiously, and I'd imagine a lot of other people would too. If it was easy enough to use and as unobtrusive as possible, people would be using it like crazy. The spammers would get wiped out.
Believe me, if I had the spare time, I'd start writing this program today.
you can find out how you can help give spammers a hard time here http://www.lenny.com/spam please help fight spam
http://Lenny.com
4 great justice!
My friend and I have been kicking around an idea to use Eliza to reply to all e-mail. You could give it an AI that looks at a real response to a similar question, keep a database of those replies and then only forward messages to you that require a reply because the algorithm doesn't have a context to reply from. I already have a script that ssh's into machines that I maintain and do sudos and greps and such. My boss thinks that you have to constantly futz with things and be logged in all the time. He is the last log \ /var/log/messages king. He'd probably can me if he knew that I wasn't at the switch constantly, but then again it took me a week to explain how to even check logs.
Comparing it to Windows will be a moot point, since El Dorado is going to have a 40% larger code base than XP.
Now all it has to do is click on the link in the spam and fill out forms with data to clog their databases. More information can be found here:
http://lenny.com/spam/index.html
There is no
you can tell procmail to exit with whatever exit code you feel like it. there's an exit code for "no such user" if you can detect your spam with procmail and any combination of scripts, you can force it to send a bounce message saying that the user doesn exist. if you want, you can even force that sendmail-generated bounce message, and still receive the mail.
consider the following recipes:this would have you still able to read your spam, if you're so inclined. (deliver it to /dev/null if you're not.) i would tend to think that a "user does not exist" bounce message would be better for preventing future spam than an annoying little eliza-generated email
#define F(x) int main(){printf(#x,10,#x);}
F(#define F(x) int main(){printf(#x,10,#x);}%cF(%s))
I agree it's nasty, but hardly forgery. It's no more forgery than writing "Dr. Nikolai Pantsanundies, 6th planet of the Gastric System, 7th Dimention" in the return address portion of an envelope. What we NEED instead of more stupid laws (we have enough idiotic ideas about computer "crime" in law enforcement already, Taco) is legal recourse for the recipient. I say, if we can track down who sent it, we should have a legal right to send them a bill, and sue if it's not paid. I mean a law that explicitly states this, not just a sig tagline like some guys have tried over the years.
jX [ Make everything as simple as possible, but no simpler. - Einstein ]
It's a cool service. Offered free, but I recomend you pay for it... doesn't cost much and it's such a great service.