John,

Thanks, I hadn't found that one. I've read it now, and the IETF 106 minutes.

It's the same family of idea, two domains pointing at each other. From the
minutes it seems to have stalled on a few things: no defined meaning for
"related" (the slides say it implies no trust or other semantics), a new RR
type, an optional signature scheme the authors themselves marked as at
risk, and no clear consumer.

DAC tries to avoid exactly those. It defines one meaning for one use (the
confirming domain operates or authorizes the other, for takedown and
reputation checks), uses TXT at an underscored name, adds no keys, and the
confirming side never publishes a list of its domains.

I'll write to Alex and Stephen as you suggest, and cite RDBD in -01 with
how this differs.

Dimitri

On Mon, Oct 26, 2026 11:44 PM, John Levine <[email protected]> wrote:

> I was wondering why this draft seemed familiar, and finally remembered why:
>
> Here's a draft from 2019 proposing the same thing that went nowhere.
> Perhaps you can
> check with the authors and see what happened to it.
>
> https://datatracker.ietf.org/doc/draft-brotman-rdbd/
>
> R's,
> John
>
_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to