I think that's an overstatement or beside the point. While according
to standards theory, Sender: headers may have nothing to do with
whether the message is bulk, practice may differ. I'd want to look
at a few 1,000,000 private messages in a variety of locations and
check the behavior of a bunch of MTAs and MUAs before guessing whether
in practice "Sender" headers indicate "bulk."
Considering that we are a research group and not a standard's group, that
would be an excellent approach. We've found that the Sender header works in
many cases but many people pointed out additional methods...this isn't a
trivial issue.
From: "Eric Dean" <eric(_at_)purespeed(_dot_)com>
ok..very nice..so what do you propose for list detection?
It's not Dave Crocker's fault that Sender headers do not necessarily
indcate "bulk," and neither is it his responsibility to find a solution
to your problem if they don't.
You could but don't absolutely need to
thank him for pointing out a hole.
Thanks Dave
You certainly should not snarl.
Indeed...I did snarl as I wrote that response.
In your position, I'd hope to offer evidence showing that in practice
Sender: headers do work or some (real, not invented) words in an RFC
showing that Dave Crocker is mistaken. Failing that, I'd probably
fall back on "well, C/R isn't perfect, but it's good enough to help"
(if that were my opinion, which it isn't).
Well, I believe there is more work to be done in this area. We use Sender
but it appears that many other mechanisms are available..yet nothing
standard.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg