From: Erik van der Poel (erik@vanderpoel.org)
Date: Mon Mar 07 2005 - 01:19:54 CST
Doug Ewell wrote:
> I agree that ℂ is unnecessary in domain names. I also feel that it's
> harmless.
It certainly doesn't seem like a big problem at first glance. But, in
networking, it is sometimes bad to have too much flexibility. It has a
way of coming back to haunt you later.
> There are hundreds, if not thousands, of characters like this
> in Unicode.
And that's supposed to make me feel better?
> Accepting NFKC lock, stock and barrel for this purpose is probably
> better than second-guessing the work of the UTC.
Actually, the Unicode database does contain some info that we might be
able to use in Nameprep. Double-struck C has the tag <font> in its
Character Decomposition Mapping, while the Japanese wide characters have
the tag <wide>. So Nameprep could choose to normalize just the <wide>
characters and maybe some of the others, but not <font>. For domain
names, that may be the way to go.
Erik
This archive was generated by hypermail 2.1.5 : Mon Mar 07 2005 - 01:21:32 CST