> I too agree with Michael.

As do I. Had it not been the case that this change had already been
announced and vetted, I wouldn't have made it. The operating theory
is to plug the hole while we discuss whether it should be plugged or
not.

I think Michael's fascinating statistic (10 of 215,342, or 0.005% of
spam he gets is to the address he uses on TEI-L) helps make the
argument that public access (particularly via Google et al) is more
important than the privacy issues at stake here.

I'm investigating the possibility of running some software to alter
e-mail addresses to human but not machine-readable formats, but sadly
it is not a feature of LISTSERV, the software Brown runs (at least
not at version 1.8e or whatever). LISTSERV was written before spam
became a problem. (Although LSoft has added features to help prevent
LISTSERV from *sending* spam.) So it would have to be something run
as a cron job on the server, and thus would not be 100% protection,
as the last few items posted since the last time the job was run
would not be anonymized. This is a possibility, but not a sure thing.
(If anyone knows of a good utility for this, please let me know.)

However, I still think just going back to a public list, even if I
can't get anti-mining software run on the server, is the way to go.
Spam is here to stay, and while anonymizing software may help delay
the inevitable onslaught, it won't prevent it.