Re: is it just me or...

[Rich Kulawiec via NANOG <nanog () lists nanog org>]

On Fri, May 23, 2025 at 01:49:07PM -0700, William Herrin via NANOG wrote:
> It's not just you. The "From:" header is being changed to:
>
> From: Your Name via NANOG <nanog () lists nanog org>
>
> I understand this has to do with DKIM crappiness where if any trace of
> the original email address remains in the From: header, the message
> may be rejected.

I believe it's DMARC, but: also crappiness.  Some instances of Mailman
are configured to only do this when it's necessary (based on the records
published by the domain); some are configured to do it all the time;
some are configured to never to do it.  I'm going to guess that this list
may be in the first category, but that's only a guess and I very much
defer to the list-owners.

As long as we're discussing this, let me note that there's an unfortunate
consequence of this DMARC...accomodation that sometimes arises when email
clients that auto-add contact addresses are in play.  Let me use the
message I'm replying to as an example.  It includes this header:

> From: William Herrin via NANOG <nanog () lists nanog org>

An email client that wants to helpfully populate the user's contact
list is likely going to extract:

        nanog () lists nanog org

and assign the text string:

        William Herrin via NANOG

to it.  This is not good.  And it becomes even more fun if the user (later)
composes a message addressed to nanog () lists nanog org, because the email
client may auto-populate it thus:

        To: William Herrin via NANOG <nanog () lists nanog org>

even though it's a new message in a new discussion thread.

OR if the user (later) tries to compose a message to William Herrin,
because they may start to type "William Her", the client will autocomplete it,
and if they're not paying attention to what the address really is,
they'll send it here instead of where they're trying to send it.

Unfortunately, this is not speculation.  (Well, this example is.)
I've seen it -- twice -- on other mailing lists.

But all this DMARC and DKIM complexity and bag-on-the-side-of-a-bag ad
hackery must be worth it, because surely they have solved the spam problem
(narrator: NO!) and forgery problem (narrator: NO, AYFKM?!).

---rsk

_______________________________________________
NANOG mailing list 
https://lists.nanog.org/archives/list/nanog () lists nanog org/message/2TUI7I62GQDWCTUK4XQFVZ63JCU2HZMI/