From: Rick McGowan (firstname.lastname@example.org)
Date: Wed Jan 26 2005 - 14:35:05 CST
Simon Josefsson wrote,
> Nowhere in the current document can I find any text that say that
> internal-idempotency was a design goal or even a requirement.
It may not be in the document, but I don't think anyone developing this
would have decided that idempotency is not a requirement. Obviously, it
should be a requirement, as illustrated by the sample implementation.
Perhaps everyone just assumed that. And UTC also sees it that way,
otherwise they would not have approved the change to D2.
In any case... PRI #61 is about the document UAX #15, and the changes in
it, not about this normalization issue per se. The PRI for the
normalization inconsistency (change in D2) was decided long ago (June 2004)
under PRI #29, which is closed:
UTC decided that the documentation was inconsistent with the sample
implementation, and voted to make them consistent. Whether this actually
constitutes a "change" depends on your point of view, and what you actually
implemented. See the PRI #29 review document for information on which
implementations were found to be inconsistent.
This archive was generated by hypermail 2.1.5 : Wed Jan 26 2005 - 14:37:03 CST