On Tue, 4 Mar 2003, Hadmut Danisch wrote:
On Tue, Mar 04, 2003 at 10:49:50AM -0700, Vernon Schryver wrote:
I don't like text analysis while receiving the message. And I am
not really convinced about such content methods.
A rarely mentioned but very important advantage of doing content analysis
during message transfer is that suspected spam can be rejected rather than
discarded. In the rare case of a false positive identification of spam the
legitimate sender will get a clear indication that his message was not
received. You can't really send a DSN later, since most such notices (for
genuine positives) would go a forged address and be little better than
spam themselves.
BTW:
I am attracted to hashcash, but how would desired bulk mail be
whitelisted? By connection IP address? By From address? By subject line?
None of those seem ideal to me, but perhaps connection IP is possible.
That would mean whitelisting the server and not the list (probably a good
thing, but when a mailing list changed servers users would miss mailings).
Could some public spirited citizen keep a publicly directory of opt-in
mailing lists that MTA operators might want to whitelist? That would seem
to be essential if whitelisting mailing lists was not to become a clerical
burden. Would some form of address authentication be required? If so, then
hashcash is not a substitue for RMX but a complement.
Dan Feenberg
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg