Stability of valid IDN labels

[Andrew Sullivan]

On Mon, Apr 21, 2008 at 04:47:13PM -0400, John C Klensin wrote:

> I believe that, if one is going to start making "forever"
> statements, that is the one you end up with.  At the same time,
> Eric is correct that registrations have never been treated as
> immutable, even though concerns about archival URI references
> are moving us down that path.

Yes, I see.  I actually have a fair degree of sympathy with the notion
of temporal scope -- I seem to recall arguing in favour of something
similar for "variants" in the past.  The problem is that, whereas I
had no problem seeing how one could enforce and check such rules at
registration time, I don't see how we can make this work on the
resolution side without replacing all the resolvers in the world every
time the list changes or (more likely) just building dns2.  Neither of
these seem to me to be palatable (and they're certainly off charter).

I nevertheless have a great deal of sympathy for the argument that a
registration, once its made (assuming that registration was correct at
the time) has to remain valid for as long as the registrant desires.
In the current domain name market, "approximately forever" is going to
be close enough to satisfy that.

> Yes, but where that argument leads is also to being as
> conservative as possible when adding things to what is now
> called the Protocol-Valid list, regardless of origins.
> Fortunately or unfortunately, we got rid of "MAYBE" as too
> complicated and not sufficiently justified, and our only real
> lever for being conservative about, e.g., things classified as
> letters went away with it.

Perhaps this is an argument for resurrecting MAYBE, then.  (I'm not
being glib; I just don't know.)

A

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