From: Brad Templeton <brad(_at_)templetons(_dot_)com>
...
Brightmail is not filtering as I normally consider it. They have live
humans identify spam attacks, and then publish tools to detect those
particular attacks. Thus no false positives.
If that were true, it would still fit the word "filtering."
It's clearly false to hope that Brightmail has no false positives.
For an extreme example that proves the point, consider a spammer that
includes its own drop boxes in its target lists. Brightmail will zap
the mail regardless.
All filtering involves some kinds of false positives, because
"mail I don't want" is fundamentally in the mind of the target.
To my mind the problem with filtering is it goes after a side issue.
I've said the same, but I suspect we disagree. Most filtering goes
after side issues such as the use of naughty words or words like
Viagra. However, the DCC is a filter but it hits directly on the
"bulk" aspect of "spam."
...
Commerce is the most common motive for spam, but not the only one, but
that is what laws go after.
Politics and religion are also common motives, albeit not quite as
common some sort of commerce. There's no hope that politicians might
ever outlaw their own spam.
Being at an ISP where people spam does not mean you are a spammer,
but that is what blacklists test for.
No, only some blacklist tests are about that.
Being an open relay isn't the cause of spam, but that is what relay
blacklists test for.
Agreed.
Being a stranger isn't the cause of spam, but that is what whitelists
test for. However, it is one of the key components, so this is more
on target, as long as there is a way for the strangers sending person
to person mail to get their mail through.
That works only for people who really want to send you mail.
It's hard to get strangers to jump through any hoops at all.
Free bulk e-mail is one of the causes of spam, so attempts to put
a cost on e-mail are going after the root problem, but they are
very difficult to implement and have other consequences.
Attack the root cause, not the symptoms.
agreed.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg