Re: [Idna-update] [Ext] FWD: Expiration impending: <draft-klensin-idna-rfc5891bis-01.txt>

[Andrew Sullivan <[e-mail]>]

Hi,

Having failed to hold up my end of the log on this work, I guess I
should say something here.  Also, I was on the IAB for some of this
and lead on the i18n program(s) during some of it, so I should take
some blame.

On Mon, Mar 05, 2018 at 07:34:22PM -0800, Asmus Freytag wrote:

> (1) A small, but significant number of both *existing* code points and
> combining sequences exhibit the same issues as the code point that lead to
> the IAB recommendation to halt the update of IDNA2008 tables.

Yes.  Note the IAB couldn't actually halt anything (Asmus didn't
suggest we at the time did, but I want to make that clear).  What the
IAB did was ask IANA not to do more updates until there was some
consensus.

> (2) Objectively, halting the process did and does nothing about existing
> "troublesome characters".

Sure, but I don't think anyone thought it did.  I think the idea was
that the IAB observed a problem that it believed had been solved by
IDNA2008, but which had not been.  That problem remains unsolved.

> repertoire elements is generally a poor way of mitigating the issues -
> mainly because doing so would arbitrarily favor letters over digits, or one
> language or writing system over another.

I am not sure I am prepared to concede that such limitations is
"generally poor", since in fact general arbitrary limitations are what
we have in the "host name" or "LDH" rules, historically.  STD13 is at
some pains to point that out, so I don't think it is controversial.

The problem is when that arbitrariness rubs up against both commercial
and political realities.

> require different mitigation approaches well within the scope of
> registration policies.

But registration policies are not uniform and can't be assured.  This
is how the "DNS is distributed" and "everyone should be safe"
principles come into conflict.

> (6) By freezing the update of IDNA2008 tables, IAB effectively declares that
> IDNA2008 is "stuck in the past". This incrementally increases the pressure
> on / temptation for various operators to unilaterally move beyond IDNA2008.
> If such "wild catting" can cloak itself in the moral mantle of support for
> some minority languages, it provides cover for those cynically selling emoji
> labels.

Worse, of course, IDNA2003 provides no principled reason _not_ to add
emojis, and I think this argument from Asmus is an important (and
too-overlooked) point.  Even if IDNA2008 draws the line imperfectly,
the identifier/non-identifier code point line is _way clearer_ in it
than in IDNA2003.

Best regards,

A

-- 
Andrew Sullivan
[e-mail]