John C Klensin
klensin at jck.com
Thu Jul 17 20:21:17 CEST 2008
--On Thursday, 17 July, 2008 19:32 +0200 JFC Morfin
<jefsey at jefsey.com> wrote:
> At 18:03 17/07/2008, John C Klensin wrote:
>> But it does seem to me that there is a more general
>> question here at least for the future and especially if the
>> reason for deprecating a character is to replace it with a
>> different preferred (or normalized) form.
> This seems to indicate that the proposed unique text should
> not only be clear to developpers but also to operators. If a
> character is replaced, should it not be suggested that the
> domain names using be automatically updated ?
"Automatically" would be very bad news, for reasons that should
be clear to anyone who has read the specs and considered the way
that IDNA is necessarily implemented, the way the DNS works, and
the associated transitional effects. In that sense, I think it
is obvious that Mark is correct -- since deprecation does not
remove code points, we are safe in ignoring any deprecation of
characters and just continuing whatever we are doing.
My question was about whether best practices might tend in the
direction of discouraging further registrations with deprecated
characters and whether changes made outside the IDNA
environment might make such characters harder to access, enter,
display, etc., in the long term.. and, if so, whether we need to
make any provisions for those possible trends.
More information about the Idna-update