[idn] IDNA: a micro layer above DNS or DNSinternationalization ? (was: Re: I-DACTION:draft-ietf-idn-idna-08.txt)

[John C Klensin <klensin@jck.com>]

The second issue of my attempted dissection really has two
parts.  The first one was: 

	Is IDNA closely bound to internationalization of the DNS
	or, as you note above, is it really "about a micro layer
	above DNS and below applications" and hence "has nothing
	to do with putting anything into RRs"?  

I have long read the charter of this WG more broadly than Dave
has, but I see nothing that authorizes the WG to develop general
internationalization tools or new micro layers.  So, if we take
the "nothing to do with the DNS" reading of what IDNA is about,
then identifying IDNA documents as products of this WG was
improper, the Last Call was procedurally improper since it
claimed, I believe, that IDNA was a WG product, and any
discussion of IDNA itself in this WG is out of charter and
improper.  

Of course, the introduction and abstract of the IDNA
specification claim just the opposite -- that it is a [proposal
for a] "standard method for domain names to use characters
outside the ASCII repertoire" and that it is "only meant for
processing domain names, not free text".

I don't believe that the community would be well served by
hairsplitting about charter boundaries in this case. But, if we
are flexible in that regard, and accept IDNA as DNS-related
enough to be within this WG's scope, it seems to me that the DNS
implications of IDNA, including whether it is properly specified
for DNS use, become part of the WG mainstream.  

Logically, the alternative would be for the Last Call to be
withdrawn and repeated as a "not a WG product" one and for this
WG to start work on a DNS internationalization profile for IDNA
that normatively references the "micro layer" IDNA specification.

The second part of this issue flows from the above:

	[And] are Eric's concerns properly addressed to the IDNA
	specification or should they be focused somewhere else?

I am increasingly convinced that most of them do not belong in
the IDNA specification, but that

	(i) The WG has not completed its work to the extent that
	anything is appropriate for interoperable use with the
	DNS until the complementary specification is written and
	placed on the Standards Track.
	
	(ii) Separating them may require a substantive change to
	the way the IDNA specification is written, but not
	necessarily the way that IDNA operates.

In either case, that new specification is definitely part of the
necessary subject matter for this WG, even if IDNA is not.

   john