On Jan 10 2005, Peter J. Holzer wrote:
Only the list expander knows the subaddress. Michael's MUA sends mail
to the main address, which replies with a CAPTCHA.
That's not the way I understood that it works.
ISACS rewrites all outgoing mails to contain a unique subaddress for
each recipient. Thus all your mails to the mailing-list will contain
your subaddress for the recipient asrg(_at_)ietf(_dot_)org, (e.g.
<laird(_dot_)123(_at_)(_dot_)(_dot_)(_dot_)>),
not your main address <laird(_at_)(_dot_)(_dot_)(_dot_)>.
When Michael hits the group reply button, his MUA will take the
addresses from the headers and compose a mail to <asrg(_at_)ietf(_dot_)org>
and
<laird(_dot_)123(_at_)(_dot_)(_dot_)(_dot_)> (It doesn't know that this is a
subaddress, nor what the
main address is).
I didn't think it would be this way, as that would make spam attacks so much
easier.
Consider the following: I give a subaddress
laird(_dot_)123(_at_)(_dot_)(_dot_)(_dot_) to
the ASRG mailing list. I now send a message to the list.
1) If the ASRG expander forwards my subaddress
laird(_dot_)123(_at_)(_dot_)(_dot_)(_dot_) to everybody
on the list, then everybody knows my subaddress and can send me mail without
being filtered. Moreover, the public archives at
http://news.gmane.org/gmane.ietf.asrg.filtering/
now list the subaddress laird(_dot_)123(_at_)(_dot_)(_dot_)(_dot_) , so anyone
on the internet can send
me mail without being filtered.
2) If the ASRG expander censors my subaddress, ie claims the mail is from
laird(_at_)(_dot_)(_dot_)(_dot_), then anybody replying to me privately must
still pass a CAPTCHA.
But at least the list expander is the only entity which can use the unfiltered
subaddress.
--
Laird Breyer.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg