Browser IDN display policy: opinions sought

[John C Klensin]

--On Saturday, December 10, 2011 21:51 +0900 "\"Martin J.
Dürst\"" <duerst at it.aoyama.ac.jp> wrote:

>...
> Also, now that we have non-ASCII TLDs, that gives us some new
> ideas. We should be able to assume that ICANN wouldn't be open
> to visual spoofing at the TLD level, such as e.g. not allowing
> whole-script confusables in Cyrillic or Greek. That should
> mean that cyrillic.cyrillic and equivalents are safe to
> display. And these are incidentally the domains where IDNs are
> really at their best, and where the growth should go.
>...

Having suggested yesterday that we should try to get ICANN to
fix a large fraction of the underlying problem, and maybe could
succeed, and had my sanity questioned (in good humor; no offense
taken), a comment on the above...

I note that you say "should be able to assume".  It seems to me
that is correct.  It also keeps you out of this wagon I'm riding
to the nut house :-)

As to how realistic that assumption is, you might consider:

(1) ICANN's Board apparently (minutes have not yet been posted)
passed a resolution on Thursday exempting IDN variations on .EU
from review for visual spoofing and other forms of
confusability.   An optimist would assume that is a special
case, never to be repeated, and that EU itself will be careful
to avoid potentially-confusing strings even if doing so results
in unnatural translations.  A pessimist would assume that anyone
with sufficient power and leverage can get such an exemption and
that one should be careful about what protections one assumes
will come from that direction.

(2) The rules about what will or will not be accepted are
contained in the "Applicant Guidebook", which I strongly
recommend as recreational reading for anyone who is trying to
figure out what one should expect from ICANN in this area.
http://newgtlds.icann.org/applicants/agb
As a guide to relevant material in this 352+ page document
(maybe much longer-- the pre-assembled version seems to have
some sections omitted), see Section 1.1.2.10 on "String
Contention".  Those of you who took statistics courses may find
it particularly interesting to contemplate what a definition
actually means that is based on whether the combination of two
things "create[s] a probability of user confusion".

If you have any strong opinions about either that document or
exemptions from confusability reviews, remember that this list
is not the right place for the discussion.  The members of
ICANN's current Board of Directors, including the CEO, are
listed at http://www.icann.org/en/general/board.html should you
want to express your support (or offer other opinions).

    john