It appears that Baptiste Carvello  <[email protected]> said:
>while this might be true, it also raises questions: isn't authentication
>of _direct_ mail flows between good faith actors precisely what DMARC is
>supposed to achieve. How could this be messed up?

One of the large providers is ARC signing mail that was sent to them, then
forwarded outside and forwarded back, not realizing that the external link
means that you can't tell what happened.  They're not supposed to do that
but the practical message is that ARC is too hard or too underspecified
to produce useful signals.

>It does seem that the main reason why ARC supposedly failed is that it
>was only hurriedly and half-heartedly discussed on this list.

ARC was invented in 2015.  If a decade isn't long enough for people to
figure it out, how long do we need?  The large mail providers all implemented
it and have a great deal of experience with it and don't find it useful.

R's,
John

_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to