Richard Clayton wrote in
<[email protected]>:
...
|>Internet-Draft draft-clayton-dkim-canmodify-00.txt is now available.
|>
|> Title: Permitted modifications to DKIM2 messages
For DKIM/ACDC (fwiw: next/last draft no more this week) there is
just one word: no.
DKIM is a byte-based protocol, which does not even get quoted-pair
stuff right, and it will remain like this.
Already for subject tag interpretation you need the full support
of RFC 2047 (as the "[TAG]" prefix gets enwrapped more often than
one would think, really).
Not to talk about MIME parts!
For DKIM/ACDC the *users* (sender and recipient), and therefore
its MUAs, are the thing. The recipipient's MUA may even process
a message, and therefore a DKIM/ACDC ingress signature, plus
differential changes (their necessity thus to be reduced) many,
many times, over years.
On the SMTP level it is only about "alignment", and mitigation.
The same mitigation that is used today, but official, undoable,
cryptographical verifiable up to first hop (in a perfect world).
For DKIM/ACDC('s next draft) there will be only *one* condition
that does not require mitigation, and that is just the very same
as of today, with DKIM/SPF/DMARC: if a mailing-list/xy does not
modify anything. There is a quite "popular" writeup "Actually,
DMARC works fine with mailing lists" [1] of the Linux kernel
i would think admin. Linux and NetBSD plus some more
mailing-lists do it like that.
This already needs special treatment, and it is already more than
complicated enough.
Everything else will be actively mitigated, protecting (the sender
and) the recipient from misinterpretations, also visually.
[1] https://begriffs.com/posts/2018-09-18-dmarc-mailing-list.html
P.S.: i do *not* concur with [1], in that *i* dislike DMARC and
everything surrounding how it was done. RFC 9057 gives very good
reasoning, better than i ever could (also: in english).
--steffen
|
|Der Kragenbaer, The moon bear,
|der holt sich munter he cheerfully and one by one
|einen nach dem anderen runter wa.ks himself off
|(By Robert Gernhardt)
_______________________________________________
Ietf-dkim mailing list -- [email protected]
To unsubscribe send an email to [email protected]
