Re: Mississippi Hißes

John C Klensin klensin at
Fri Dec 11 19:29:24 CET 2009

--On Friday, December 11, 2009 10:09 -0800 Mark Davis ☕
<mark at> wrote:

> I agree with you that Michael can put whatever he wants into
> If, however, no client software ever ends up
> going there, it doesn't matter much what he does. So if he
> wants to have <$> (which equals
> <>), there isn't anything stopping him -
> it just won't work in practice.

But this is exactly where the relationship between some of our
discussions and Internet reality turns problematic.   There are
a few open source browsers and several more open source mail
clients, and clients for other applications, out there.  Even
for some of the things that aren't open source, plugins work and
often aren't hard to construct.   So such names might be of use
only to Michael and whatever friends he got to run compatible
software.  But that might be exactly what he would prefer, or he
might intend the names only for internal use.  That is not a
made-up case: for internal (not intended to be used by unrelated
parties over the public DNS) use, UTF-8 labels are very common.
See the slides and/or audio from the Hiroshima tech plenary
and/or draft-iab-idn-encoding.

The use of such names with unmodified clients that are strictly
conformant with IDNA is going to be very limited, and Google
probably won't index them, but, depending on his hypothetical
goals, there is considerable distance between those restrictions
and "won't work in practice".

Besides, the assertion Shawn made, and to which Michael and I
responded, was about registration, not utility.

> On trademarks, however, what you imply I said about them
> doesn't seem to have much to do with what I actually did say
> about them.

I didn't mean to imply that anything you said on the subject --
see my more recent note to you about that.  I was merely
referring to the fact that you did comment, with authority, on
the general relationships.


More information about the Idna-update mailing list