Re: [DNSOP] draft-liman-tld-names-04

[Masataka Ohta <[e-mail]>]

Some non-ASCII name wrote:

>> The reality is that, with ISO 8859/1, which is what you are using
>> in your mails, plain 'Y' without diaeresis is used as a capital
>> form of both plain 'y' without diaeresis and 'y' with diaeresis.
> 
> No, that statement is not correct. The mapping you might want
> depends on context, and that not only because of language,
> but also where the language is used. I.e. it depends on what
> locale you are using.

IT'S ENOUGH THAT YOU ADMIT "IT DEPENDS". Remember that you wrote
"Because the matching function can not be changed." in your
mail on November 13.

And, under ISO 8859/1, it does not even depend, because ISO 8859/1
does include 'y' with diaeresis but not 'Y' with diaeresis, which
means all CAPITALIZED representations (see above for an example),
all you can use is plain 'Y' without diaeresis.

>> That's the case insensistivities of the real world, regardless
>> of how Unicode defines case insensitivity. Worse, localized
>> domain names must but can not handle the real world issues.
> 
> And that is why case insensitivity is NOT part of IDNA2008.

OK. You admit that your statement in your mail on Nov. 12:

>> Unlike RFCs 1034 and 1035, it lacks discussion on case
>> insensitivities of non-ASCII characters, I'm afraid.
> It does:

is wrong.

Then, we have agreed that localized domain names in IDNA2008 can
not handle case insensitivities.

> Instead only case folded code points are allowed.

That's not the point.

						Masataka Ohta