Re: [idn] IDN charter update - revision 5

[Dan Oscarsson <Dan.Oscarsson@trab.se>]

>4.  A standard track specification on normalization of domain
>name identifiers for the purpose of string comparisons. This document
>may
>include case folding, map outs, prohibited characters, and what is
>acceptable as hostnames.
>
>Goal & Milestone:
>
>Dec 2001        domain name identifiers normalization draft last call

I still is not satisfied with the above.
The draft above is which?

Point  4. above says a specification for string comparisons.

I need a specification defining the normalised form for domain names,
so that domain names can be handled easy,
This form may not include case folding.
It is so that software can easy manipulate the character data.
Typically this is Unicode normalisation form KC or C.
If unnormalised character data is allowed,  software will fail in
many places (including DNS, LDAP and web browsers).
It is a wast of CPU resources to allow non-normalised character data.

I still think point 4 ned to be split into two:
- One defining a normalised domain name so it can efficiently and
  easy be handled by software.
- And one defining the canonical form that can be used when
  doing binary comparing of domain names to check for equality.

   Dan