On Mon, 18 Oct 2004, Frank Ellermann wrote:
jpinkerton wrote:
My comment was referring to that fact that Stephane seems to
thing that there is no spf community.
ACK, but to a certain degree that's true. The "SPF community"
asked Meng again and again to drop Sender-ID, and Meng in turn
asked again and again to (ab)use v=spf1 policies for Sender-ID.
It started some months ago when Wayne wrote "that's the split".
Meng is trying to do consensus building, and has been for some
time. This has its uses in bringing big players who can support
features at the MUA level or in various corporate SMTP server
packages. But it's not working, because the dichotomies between the
Microsoft approaches (sell licensed keys to individual companies or
users, stuff XML into headers for verification of authorized users
from a central key handling repository, patent it so it can't be
modified or extended by others, etc.) are so in contrast to the other
goals.
These other goals include extremely lightweight protocols, extremely
flexible policy publication and management, open source so every one
can use it freely and extend it as needed, and minimal impact on
already existing services so nothing already existing breaks. SPF does
this rather well, but its good name has been effectively been hijacked
by its its "press release advertised" association with Microsoft's
SenderID system.
So Meng? What would it take to convince you to dump Microsoft's
involvement and negotiations in this effort and tell them they can
ride in back along but they have to stop kicking the back of the
driver's seat?
--
Nico Kadel-Garcia
Systems Engineer
Mitsubish Electric Research Lab
<nkadel(_at_)merl(_dot_)com>