Consensus Call Tranche 8 Summary - Addendum

Andrew Sullivan ajs at commandprompt.com
Wed Oct 22 17:08:00 CEST 2008


On Wed, Oct 22, 2008 at 08:53:16AM -0400, John C Klensin wrote:

> 	of the tree and below, the current documents carefully
> 	do not distinguish among registries at various levels of
> 	the tree (a characteristic that I believe is very
> 	important and that, given the DNAME-related issues that
> 	have been problematic for some Bidi approaches, possibly
> 	an absolute necessity).  

I emphatically agree that this is the current state in the documents,
and that to the extent that refusal to distinguish is not clear, it
needs to be made clear (I think John's done good work on this front,
so thanks to him).  I also agree strongly that there are strong
reasons not to treat the "top level domains" as somehow different in
kind from any other level of the tree.  (If one needs an example why
trying to tie some sort of semantics to the position of a label in the
tree is a bad idea, I suggest looking at the train wreck that the
browser cookies specification has caused.  Or, perhaps I should say,
"train wrecks" and "continues to cause".)

> characters".   I believe that recommendation is common sense,
> although a hypothetical particularly greedy registry or
> registrar might not agree.  If people feel that recommendation
> should be made explicit in the documents somewhere, please
> suggest text and where to put it.

If rationale mutates into "advice for registries" or something like
that, then text of this nature could be included.  But I don't think
there's a home for it today.

A

-- 
Andrew Sullivan
ajs at commandprompt.com
+1 503 667 4564 x104
http://www.commandprompt.com/


More information about the Idna-update mailing list