Changing the values of domain names and the need for mapping

Andrew Sullivan ajs at shinkuro.com
Fri Feb 20 20:54:44 CET 2009


On Fri, Feb 20, 2009 at 09:23:54AM -0800, Paul Hoffman wrote:

> In the IETF, making protocol changes that affect operations of
> something as valuable as the DNS and not giving any operational
> advice is rare. If we want to do either (a) or (b), we either need
> to come up with the operational changes ourselves or hear from the
> affected parties about what they are thinking. I believe the latter
> is better.

I think there is a serious potential for confusion in the contemplated
change.  I also think Paul has a point that changing IDNA such that ß,
which used to be mapped, is now to be PVALID amounts to a
bits-on-the-wire change.  That sure looks like it at least approaches,
if not crosses, the line laid by the WG charter.  Nevertheless, I
think calling this a change to the operations of the DNS is stretching
the point.

Nothing about the DNS protocol or any operations relating to it will
need to change as a result of the change.  

There may be particular groups of people whose expectations of what
happens when they type certain letters into a "resolution context".
That resolution context is _not_ DNS resolution, but something above
DNS.  It so happens that the plan is to change what happens there. 

There likely will be things that operators of DNS zones need to do;
but I have to agree with Cary that these are already things such
operators need to do whenever they make policy changes about what may
appear in the zone.  This one happens to be more complicated because
it's a strictly speaking incompatible change, with some potentially
rather unpleasant side effects, in which both versions have to
interoperate for some time.  But the policy makers for those zones do,
I think, have the capacity to make policy such that nothing changes in
the zone itself the day IDNA2008 is turned on.  (For instance, one
could simply make a policy that ASCII-compatible encodings of "ß"
aren't allowed.)  This is the basis for the hand-wavy "sunrise like"
provisions.  There will surely be pain for operators of zones that
need to do these things in a competitive context, and I bet a very
good lunch that we'll see at least one "hijack" claim over one of the
possible example cases before it's all through.  The WG appears to
have decided that that's ok.

I do believe, however, that there is way too little involvement of
some of the technical and policy brains who are going to have to deal
with the effects of this change.  If someone has some mechanism by
which those people could be engaged, it'd be nice.

A

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


More information about the Idna-update mailing list