Re: [dbound] department of poor memory, was Fwd: New Version Notification for draft-dcrocker-dns-perimeter-00.txt

Setting aside tree walks for holistic solutions, I see per zone utillity in
a structured format that could be implemented by a zone operator...

Us PSL volunteers often require a custom txt record as a validation of
authenticity/authority of requestor on patches that get submitted.

Not saying it swaps right in to place, but (just theorizing here) one could
enhance the integrity of sections being from an authoritative administeator
AND validate the patch by matching requests against these entries within a
zone as part of the process to tighten up a match.

On Thu, Apr 4, 2019, 09:01 John Levine <johnl@taugh.com> wrote:

> In article <09cdb203-f926-70cd-3f7b-b06cdc48a11b@dcrocker.net> you write:
> >On 4/4/2019 6:43 AM, John R. Levine wrote:
> >> If you want to compare apples to apples, you might want to adjust the
> >> draft to compare your prefixed TXT records to my prefixed TXT records.
> >
> >What specific wording changes to the draft are you suggesting?
>
> Honestly, now that I look at it again, it's clear that none of the
> proposed hacks to avoid tree walks will work, and anything that needs
> tree walks is dead on arrival, so there's not much point.
>
> If you want to keep at it, as far as I can tell, the main differences
> between my 2016 proposal and this one is that I use code numbers to
> identify different kinds of boundaries and you use strings, easy
> enough to change either way, and yours requires walking up the tree
> one level at a time which can be a lot of lookups if people send
> hostile requests while mine does one lookup per defined boundary.
>
> You might check and see if I missed anything.
>
> R's,
> John
>
> _______________________________________________
> dbound mailing list
> dbound@ietf.org
> https://www.ietf.org/mailman/listinfo/dbound
>