Tables: BackwardCompatible Maintanence

[Erik van der Poel]

Yes, of course they are different in kind. All I'm saying is that if
the Unicode community decided that their change was important enough
to make, then the DNS community may decide that it is important enough
to make a change from PVALID to DISALLOWED or vice versa. I.e. the DNS
community may decide that the old value was wrong. (Things have been
wrong before -- see eszett, final sigma and ZW*; can we guarantee not
to make mistakes this time?)

Shouldn't the DNS community be given some time to consider such a
change before automatically updating the BackwardCompatible category?

Erik

On Wed, Dec 10, 2008 at 3:04 AM, Harald Alvestrand <harald at alvestrand.no> wrote:
> Erik van der Poel wrote:
>>
>> Hi Patrik,
>>
>> Currently, the Tables draft is a bit asymmetric with respect to
>> BackwardCompatible and Context Registry:
>>
>> http://tools.ietf.org/html/draft-ietf-idnabis-tables-04#section-5.1
>>
>> For BackwardCompatible, it says "appointed expert" and for Context
>> Registry, it says "IETF Review". It would be nice if they both said
>> the same thing, and I would personally prefer IETF Review.
>>
>
> actually, as the later discussions on BackwardsCompatible show, they are
> different in kind. The BackwardsCompatible is generated more-or-less
> mechanically, and an appointed expert should just look over it to see that
> it looks right; the Context registry will require human judgment, and there
> may be substantial disagreements about the change to be made.
>
> so I think they should be different (but it needs to be clear why it's that
> way).
>
>             Harald
>
>