In a message dated 96-08-01 19:13:49 EDT, [log in to unmask] (Aldo-Pier
Solari) writes:
 
> OK, that looks like a good  equation  to list owners:  You would not
>  track which services & nodes are being overloaded or disrupted  with
>  AOL  bounces  "because of the privacy of customers which neither pay
>  nor log in".  A very good trade-off for AOL, I would say.
 
Please don't put words in my mouth. I asked how you would propose that we
track member subscriptions to mailing lists without violating their right to
privacy. My concerns are:
 
(1) We would literally have to track all outgoing mail from all accounts
(2) There is no standard naming convention for mailing lists, so we cannot
guarantee to be able to automatically catch all list subscription requests
(3) Some members subscribe to lists which would be considered controversial,
at least by U. S. standards, and would object to anyone but them (and others
on the list, possibly) knowing of their membership.
 
> An  auto-rutine  (or
>  robot) cannot violate any privacy at all.
 
That might be debatable, again at least in the U. S.; further, since AOL is
an international concern, we would almost certainly have to have even more
complicated sets of rules to determine whether or not a given member, at a
given time, could be tracked so as not to violate national statues on
privacy.
 
>  L-Soft has a great technical  expertise on networking.  E. Thomas is
>  the  right  person  to  answer  that  question.   However,  I  could
>  speculate that if LISTSERV has the capacity to  deal  with  all  AOL
>  bounces,  it  certainly  has  the  capacity to deal with the signoff
>  commands :-)
 
I'll defer to Eric on this, but I'm fairly sure he told me at least once that
using NETWIDE signoffs was a bad idea.
 
--David