[Asrg] Re: Spam send/receive ratio

Jim,
You wrote:
>> One of the distinguishing characteristics of a spammer is the
>> imbalance between mail sent and mail received.  Unfortunately I do not
>> see a convenient way to penalize people who fall into this category.

I don't agree with that characteristic as generally defining, but with some 
large exceptions, it could be. I'll accept it for purposes of discussion.

>   One idea about this (although I don't know how scalable it is) would 
> be, if you could authenticate the sender ie DomainKeys or the like, then 
> every sender would have a send/receive ratio (somewhere), which would be 
> entered into the mail header in some way that it could not be spoofed 
> (more protocol changes though).  Then if a mail is received by the 
> end-user's inbox, it could check the ratio, and decide whether to 
> accept/reject/quarantine it.

Not too many protocol changes, if the entire check were out-of-band. The 
only problem with it is that there either needs to be some trusted 3rd-party 
that would keep the counts (hah!) or some distributed system that could 
produce accurate, Internet-wide, statistics on demand (and double hah!). It 
would be nice and elegant, it's just a complete impossibility.

>   But I sill don't see a way to penalize senders who are rejected too 
> often.  Although perhaps it could be integrated into the 'delay' in 
> creating the email-stamp/introduction-stamp (the more higher your 
> send/receive ratio goes, the more it "costs" to send new stuff)  Another 
> possibility is of course law enforecement, but this would seem to 
> require some centralization of the ratio information, and whose 
> rejecting what.

You would either increase their costs/delays to send, or you would reject 
outright.

>   The send/receive ratio might also be a problem for mailing lists - 
> maybe, I'd have to see statistics.

Mailing lists, in a hybrid system, could be easily white-listed by any 
number of means. The registration messages should, in theory, have a perfect 
1:1 send:receive ratio. Of course, if someone were trying to make 
subscribing to a particular list impossible, he could get it to send out 
numerous registration confirmation messages to skew the ratio.

Sincerely,
Philip Miller

_______________________________________________
Asrg mailing list
Asrg@ietf.org
https://www1.ietf.org/mailman/listinfo/asrg