Q2: What mapping function should be used in a revised IDNA2008specification?

John C Klensin klensin at jck.com
Tue Apr 7 20:39:47 CEST 2009

--On Tuesday, April 07, 2009 15:53 +0200 Alexander Mayrhofer
<alexander.mayrhofer at nic.at> wrote:

> (and, from a GUI
> perspective it would make sense that the case folding / lower
> casing output is visible to the user, much like today when a
> user enters http://www.IETF.org , the browser switches the
> location to http://www.ietf.org/ )

Please don't count on that transformation since almost all
original readings of the DNS spec, and many contemporary ones,
would claim that transformation is a bug if it is done as a DNS
matter ("required in the protocol") rather than a user interface

Also remember that, for IDNs, the equivalent conversion would
have the user type http://www.faßbier.at" and see
http://www.fassbier.at", which some users are likely to find at
least mildly astonishing.

> Ugh, more ratholes on the radar, i fear.. 

And that is only the first one.

Part of my problem understanding this is that, if I were a
registry in a German-speaking country, I'd assume that the
Internet was going to continue to expand (with organizations and
registrations expanding along with it).   I would therefore
assume that, over the next few decades, I'd far prefer to
explain to some folks that I previously told would have to live
without Eszett because of a bad IETF decision that the decision
had been reversed and that, if they wanted an Eszett-containing
name, they had better get it... rather than explaining, for the
next 20 years, why "Faßbier" is really spelled fassbier,
contrary to every dictionary in the country.  Explaining why
correctly-spelled nouns, starting in upper-case characters, come
back in lower case would just about as much aggravation as I
would want to wish for.

Obviously, you should make whatever decisions you are making on
whatever basis you want to make them.  But I am surprised that
the longer-term, continued-growth, perspective doesn't seem to
be entering into the discussion.


More information about the Idna-update mailing list