Per Jessen said:
> I beg to differ.  I have to have _some_ idea about the data I'm using
> for my filtering.  Personally, I can't use your "suck it and see"
> approach. 

Hey, there is nothing wrong with your curiosity and I definitely understand 
that how a list gets its data and what steps it takes to ensure minimal to 
none FPs is important and can impact how a list is used and who might 
benefit from the list. But I think that my point was more that **results** 
matter **more** and one's lack of knowing the details about how a list works 
doesn't impact that list's quality. It is your right to not trust a dnsbl if 
you 
don't have enough information, but it is a mistake to assume that it must 
be bad if **you** don't understand it (I'm not sure that this was your original 
point, but I say this to be sure.)

Additionally, absent additional checks (i.e. FP-prevention filters), pure 
honeypot addresses don't necessarily make for a good DNSBL. For 
example, some spammers who send to harvested addresses send via 
their ISP's "shared" mail server. Also, honeypot-driven dnsbls are starting 
to miss some of the more shrewd spammers who have learned well how to 
listwash and/or are more shrewd in how they gather their addresses in 
the first place.

Rob McEwen
PowerView Systems
[EMAIL PROTECTED]

Reply via email to