On Tuesday, February 4, 2020 3:25:06 PM EST Dotzero wrote:
> Someone pointed to Sender-ID as an example experiment. A very poor example
> to choose. It was broken from the start. As an aside, I kept sending email
> to the folks at Microsoft using @Microsoft.com email addresses by using
> "Sender" to game PRA to get a neutral". Furthermore, it dragooned senders
> who had no intention of participating in the experiment by reusing their
> published SPF records in a manner they did not intend them to be used. I
> also point out how long it took to put a stake in the heart of Sender-ID.
> And yet even today we can find Sender-ID records littering the Internet and
> even a few places doing Sender-ID checks. For some definition of "We", we
> are good at additions and modifications but poor at deletes.

That was me.  I agree it was a horrible idea.  The point wasn't that Sender ID 
was great, it wasn't.  The point was that the AIB considered it reasonable as 
an experiment.  I think this is far less risky than that was.  I was trying to 
respond to the idea what the IETF doesn't support experiments where there are 
technical concerns about the nature of the technology.  The AIB's position, as 
I read it, was that such experiments are fine and if the IESG has concerns, 
they should add a note to document.

Scott K


_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc

Reply via email to