Re: An IANA Registry for DNS TXT RDATA (I-D Action: draft-klensin-iana-txt-rr-registry-00.txt)

[John C Klensin <[e-mail]>]

--On Saturday, August 31, 2013 02:52 -0700 manning bill
<[e-mail]> wrote:

> given the nature of the TXT RR, in particular the RDATA field,
> I presume it is the path of prudence to set the barrier to
> registration in this new IANA registry to be -VERY- low.

That is indeed the intent.  If the document isn't clear enough
about that, text would be welcome.  I'm still searching for the
right words (and hoping that the discussion will interact in
both directions with the 5226bis effort
(draft-leiba-cotton-iana-5226bis), but our thought is that the
"expert reviewer" will provide advice and education about the
desirability of good quality registrations back up by good
quality and stable documents.  But, if the best we can get is
registrant contact info, name of a protocol, and a clue about
distinguishing information, then that is the best we can get.

> Or is the intent to create a "two" class system, registered
> and unregistered types?

In one sense, that result is inevitable because some of the
locally-developed and used stuff that lives in TXT records will
probably not be registered no matter what we do.  That is still
better than the current situation of a one-class system in which
nothing is registered.  But the intent is to get as much
registered as possible.

Again, if the I-D isn't clear, text would be welcome.

   john