Using the vocabulary from my previous message, I believe Hadmut Danisch is
arguing as follows. He's agreeing that sender domain tracking is desireable,
but at the same time he thinks that MTA-chaing tracking is more practical. By
requiring all mail from a domain D to be injected via an MTA blessed by domain
D, MTA-chain tracking achieves sender-domain tracking. So he's asking, why not
impose such a requirement?
First, it seems to me that any scheme involving sender-domain tracking will
require messages "From:" domain D to eventually pass through an MTA blessed by
D. Thus, Hadmut's requirement of juswt making this the first MTA seems
reasonable from this perspective.
My only concern is this. Networks often block outgoing connections to port 25
(and for good reason). Pop-before-smtp and/or authenticated smtp don't help
with this problem. How do we deal with this problem? Maybe requiring
smtp-over-SSL in these situations will work (most MUAs seem to support this
now). However, this solution assumes most "ISP-like" networks will allow port
925 traffic to go out, I don't have 1st-hand experience, perhaps even this
won't work.
Other solutions? Or is this a show-stopper for requiring the first MTA be an
MTA blessed by the sending domain?
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg