Re: [DNSOP] PSD records, was Verifying TLD operator authorisation

[Tim Wicinski <[e-mail]>]

John

I totally agree with you we need to Nerd Harder one the problem space, and
folks are trying.
But one way we Nerd Harder is to do experiments like this and study how it
works, and
use that moving forward.

I've been talking to folks over the past months on this problem space
trying to grapple with
quantifying what other folks see.   Part of this reason is that my employer
has this problem,
and while my boss and I have noodled on the idea off and of for the past
year, my feeling
is that our Java developers will decide to solve this problem for their use
case, and it will
be done in their way makes their life easier and everyone else's life
harder.   I'd like to
be in front of the problem, leading them toward my promised land than
chasing them from behind
nipping at their heels.

Talking as a chair, I see this more of an Application of DNS Problem than a
DNS Operations Problem,
but I'm willing to be convinced otherwise.  This means, the work may be out
of scope for DNSOP.

I do feel that whatever solution the IETF works we, we have to first accept
that it will *not* replace
the Public Suffix List initially.  The PSL is a W3C entity, and while I
feel they will look honestly at what
the IETF may create, they are free to adopt or not adopt it for the PSL,
and the IETF needs to
accept that.

We've not only have to solve the problem, we need to make a convincing case
for the W3C to adopt.
I find the IETF is great at the former, and struggles at times with the
latter.

Tim
(mostly speaking as myself)

On Fri, Jun 21, 2019 at 11:08 PM John Levine <[e-mail]> wrote:

> In article <CADyWQ+HVYAfVE0WSEPAG7Zz_OL+uM4DmHRLn6Y=
> [e-mail]> you write:
> >-=-=-=-=-=-
> >
> >If y'all care what gets published in a TLD, please take a look at
> >https://datatracker.ietf.org/doc/draft-ietf-dmarc-psd/
> >which is an experimental draft that will go into WGLC last call soon.
> >This was driven by wanting to add _dmarc records
> >into TLDs, per ICANN rules it needs to be an RFC.
>
> I'm not thrilled about it since I would prefer that we nerd harder on
> the general domain boundary problems (the ones for which we all use
> the PSL), but PSD for its particular use case of name trees seems
> pretty harmless.
>
> R's,
> John
>