On 28/Jan/11 17:18, Michael Thomas wrote:
On 01/28/2011 07:53 AM, Esa Laitinen wrote:
It doesn't really hurt much if somebody is in "Crazy Al's
left-handers block list": badly managed services will lose their
users pretty quickly.
But not quickly enough. All it takes is a few clueless companies
or isp's to use badly run blackholes to make them a nuisance. Worst
of all, is when they claim you're on some blackhole but don't tell you
which one, and it's not one of the ones that the various blackhole
aggregator/test sites knows about. I've been having problems with
some Swiss ISP that does this.
We (improperly) call reputation the judgment of self appointed block
list operators. Spamhaus has a fame of being very good at it, but
people who've been erroneously bitten don't agree.
But it's the whole process that lacks any standardization or best
common practices as far as I can tell. If you're a legitimate sender,
you really don't want to waste your time sending stuff that readers
don't want to see. But the feedback loop is either nonexistent, very
murky or hasn't scaled down to where medium to small (read: have
day jobs doing other things than email delivery) can wire them up.
Hm.. Actually, there are some papers about how to run DNSBLs.
However, the semantics of a list is what it is: the unappealable
opinion of its operators. Would it be better if it were the result of
a poll, say, or the average opinion of thousands of mailboxes operators?
It seems to me that improving that situation would dramatically
help the false positive rate as well as the problem of accidentally
falling into blackholes of the clueful blackhole operators.
That would imply a radical change in attitude, from individual
survival tactics to free cooperative strategies.
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg