From: "Hallam-Baker, Phillip" <pbaker(_at_)verisign(_dot_)com>
Vernon's arguments are becomming increasingly obtuse.
NNTP is a Pull system. Each spam message sent to usenet is sent to the
system as a whole and not to the individual users. It is the individual user
who decides whether or not to pull a message onto their personal machine.
If that were true, then current Internet email would also be a "Pull system."
In fact the only way individual users decide whether to "pull a message
onto their personal machine" is with a protocol that is distinct from
the netnews server-to-server protocol, and that corresponds to POP.
Of course, there is just as much "pulling" of non-spam (or spam) onto
individual user machines in Internet mail and NNTP, which is not really
any at all as the notion "pull" is intended by its advocates in this
mailing list. Internet email is not a "Pull system" and neither is NNTP.
There are still plenty of people who read netnews without using NNTP
and do not in any sense pull push, slide, or otherwise move netnews
articles "onto their personal machines." The classic newsreaders that
use active files and spool directories are still in use. Users of
those classic programs are just as protected by real netnews spam
defenses (as opposed to wild guesses and surmises) as users of NNTP
based newsreaders. This is another way of saying that the supposed
"pull system" of NNTP does not figure in stopping netnews spam.
Spam mitigation in usenet has the major advantage that it is only necessary
to delete a message once per server and it dissappears for all the users at
that site.
That is between irrelevant and wrong. Netnews spam mitigation is
not based on merely deleting messages on individual servers but on
"cancelling" messages. That cancelling affects spam on news servers
even before it reaches the servers. It is true that article files
are deleted, but the same could be done to spam mail in some
implementations of mailboxes.
The terms push and pull are well understood to the network comunity in
general as are forged headers, authentication and every other term Vernon
has invented disputes over.
Yes, those terms are reasonably well defined, and but they are evidently
difficult to use based on wild guesses about network protocols.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg