From: Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu>
a substantial subset of the readers of the ietf list subscribe to the
list via some other means - whether ietf-censored or some other sub-list
or a web archive. so a list of subscribers to the main ietf list doesn't
make a very good filter.
Moreover, the response I received to my complaint about one of the recent
incidents could be read as saying the perpetrator is (or was) a subscriber.
(The response from what seemed to be someone at the offending ISP was even
less clear than the spam.)
personally, I like the ietf-censored filtering model -
both because it doesn't place much burden on the secretariat and also
because it doesn't invite accusations of censorship toward IETF.
That's a good point, which might otherwise stated as saying the problem
is already almost solved for those who want the list filtered. All that
might be missing are:
- a note in the IETF subscription welcoming message for new subscribers
such as the person who restarted this thread this time directing them
to the ietf-censored list if they want to censoring.
- perhaps (or not) some spam body filtering on the input to the
ietf-censored such as the Distribute Checksum Clearinghouse (DCC).
There would be no spam problem if people would not just hit delete or
whine about spam where complaining is merely more noise (e.g. here) and
instead religiously complain about every unsolicited bulk message to its
responsible service provider, including going upstream as far as necessary,
and if you find to an unresponsive tier 1, filtering all of its port 25
output. However, at this late date, it's clear that's not going to happen.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com