Orie, Peter,
Unsurprisingly, I agree with Peter. I slightly disagree with his
priorities even though the question that prompted this thread was
about the PRECIS profiles. To me, the conversation and clarity in
the document about the differences in purpose is the critical one.
Based on conversations with Asmus and others about what is going on
in the Unicode Consortium, they are (IMO finally) moving in the
direction of better defining usable strings with comparison,
confusion, etc., as criteria. Going in that direction suggests that
they have finally figured out that, e.g., the Unicode "Identifiers
and Syntax" spec (UAX#31) did not go nearly far enough for use
outside the very limited contexts of programming languages in which
it originated. That is interesting because, had they made those
moves 20 or 25 years ago, we might have been able to take a very
different course. I hope they will come up with a better
characterization of the differences that I/we have been trying to
describe, but it won't be soon enough for a decision on this document
in the next several days.
With that distinction in place, and explained in the document, I
think the answer to the PRECIS question will be obvious, and
obviously negative. I think the explanation would benefit, as Tim
suggests, by retaining a reference to RFC 8264 (not 7564; I'm
relieved to have company in being confused) and maybe UAX#31 with it.
Without the distinction, I (still) think there is danger of the
document causing serious harm by sending the message (however
inadvertently) that PRECIS is complicated and lots of work and that
implementations that don't want to go to the trouble can just use one
of these subsets/profiles instead.
john
--On Sunday, March 2, 2025 08:13 -0700 Peter Saint-Andre
<stpeter@xxxxxxxxxx> wrote:
> Orie, thanks for your input.
>
> In John Klensin's note on February 16 [1] and my reply to John on
> February 17 [2], I sensed that we were moving in the direction of
> greater clarity regarding differences in purpose between
> draft-bray-unichars and the IDNA/PRECIS standards. I'd like to see
> the authors engage with those messages, for two reasons:
>
> First, I suspect that doing so will lead us in the direction of
> removing the PRECIS profiles.
>
> Second and more importantly, it will help us and future readers
> understand what draft-bray-unichars is trying to accomplish and why
> internationalization regarding packaging formats differs
> significantly from (and can co-exist with) internationalization for
> addresses and identifiers.
>
> If I can help nudge that conversation along, please let me know.
>
> Peter
>
> [1]
> https://mailarchive.ietf.org/arch/msg/last-call/jRltNz8fBjEqO2NAAg7
> Odje7Nys/
>
> [2]
> https://mailarchive.ietf.org/arch/msg/last-call/Mg0w20A0g9qAAh8jYzc
> aPoBLpuA/
>
> On 3/2/25 7:57 AM, Orie wrote:
>> Hi,
>>
>> Based on the discussions here, I have a slight preference to
>> remove the profiles.
>> I'd say that it would be better to convince Peter that they are
>> needed, than myself.
>> I agree with his comments here:
>> https://mailarchive.ietf.org/arch/msg/
>> last-call/Mg0w20A0g9qAAh8jYzcaPoBLpuA/
>> <https://mailarchive.ietf.org/
>> arch/msg/last-call/Mg0w20A0g9qAAh8jYzcaPoBLpuA/>
>>
>> Regards,
>>
>> OS
>>
>> On Sat, Mar 1, 2025 at 4:13 PM Tim Bray <tbray@xxxxxxxxxxxxxx
>> <mailto:tbray@xxxxxxxxxxxxxx>> wrote:
>>
>> On Feb 28, 2025 at 6:14:57 PM, Peter Saint-Andre
>> <stpeter@xxxxxxxxxx <mailto:stpeter@xxxxxxxxxx>> wrote:
>>> IANA has asked me (as the designated expert for the PRECIS
>>> Profiles registry) to review the IANA registrations in
>>> draft-bray-unichars-10. However, my sense from list
>>> discussion is that the next version of the
>>> document might remove the PRECIS profiles. Are the authors
>>> leaning in that direction? See below for more context.
>>
>> Our feelings on the subject aren't strong, but we agree that it
>> seems the path forward is smoother without the profiles. We'd
>> like to retain the section 2.2 language citing RFC7564 and
>> encouraging readers to consult it. At this point consensus is
>> Orie's call and we're OK with whatever he says.
>>
>> * Tim & Paul
>>
--
last-call mailing list -- last-call@xxxxxxxx
To unsubscribe send an email to last-call-leave@xxxxxxxx
