At Wed, 28 May 2014 12:57:55 -0400,
Ted Lemon <[email protected]> wrote:

> What you are proposing is essentially a management function, not a
> naming function. Using the DNS to provide that function can work,
> and may even make sense in some cases, but I don't think it's the
> right thing to do from an architectural standpoint.

On a quick read of the draft and the thread discussion, I tend to
agree with this.  If this were just another minor but ordinary RR
type, it may make sense for some people and is probably worth
standardizing to let the market decide.  But the proposal includes a
lot of other technical complexity in the DNS protocol handling, such
as a special rule for DNSSEC or zone transfer and exceptional cases
for negative answers.  It also makes the content of zones even less
public, which might make sense in the era of NSEC3 and dnspriv, but
will certainly require other new considerations such as encrypting
zone transfers (just refusing xfr or normal query for NOTE wouldn't be
enough in terms of security considerations).

So, overall, it seems to me the gain of this proposal is not worth the
added complexity.

--
JINMEI, Tatuya

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to