Bundling

[Andrew Sullivan]

On Wed, Dec 02, 2009 at 11:35:21AM -0800, Lisa Dusseault wrote:
> I don't see where that principle begins and ends.  If it were entirely true,
> zone operators would be entirely free to register xn--garbage as a domain,
> entirely free to register DISALLOWED characters.

They are.  We have no DNS cops, no matter what people tell you about
ICANN.  And in fact, there are plenty of zones out there where you'll
find 8-bit labels, which turn out to be interpretable as labels made
up of UTF-8 or UTF-16 characters.  All we have is the ability to say,
"This is how you do it if you are doing IDNA2008."  We have to hope
that people play nice.  That said,

> Since it is in the purview of IDNA designers to make a character valid or
> disallowed,  wouldn't also be in the purview of IDNA designers to say
> something like "This character becomes valid on date 01/01/2020", right?  So

right.  So if we had a TRANSITIONAL category, then it would be natural
to say that, if you accept that character, there are some other things
you MUST or SHOULD do.  What we can't say, I think, is that your
existing zone MUST do _x_ or _y_, because then we're reaching past the
behaviour we're aiming for and into operation of existing zones.

At least, that's the sort of thing I'd expect from actual operators.
It's the sort of thing I'd have expected to say when I was an actual
operator.  I've more than once attempted to exhort the crowd over at
dnsop at ietf.org to pay more attention to this discussion, but if
someone else wants to try, right now would probably be an excellent
time to do so.

A

-- 
Andrew Sullivan
ajs at shinkuro.com
Shinkuro, Inc.