Re: [idn] draft-ietf-idn-ace-eval-cn-00

[Martin Duerst <duerst@w3.org>]

At 12:07 01/07/17 -0400, Keith Moore wrote:
> > Not an ACE, but probably worth mentioning anyway:
> >
> > UTF-8       21
> >
> > With the additional advantage that it's much easier to explain
> > to anybody why some names are too long and others are short enough.
>
>is it really?

Very significantly, yes.

>would an average Greek, Russian, Chinese, Japanese,
>or Korean speaker really understand the mapping between characters
>and length of UTF-8 representation any better than the mapping
>between characters and length of (say) DUDE representation?

Greek and Russian may be a bit easier for some of the ACEs,
but for Chinese, Japanese, and Korean, you either just use
some program and believe the output, whatever it may be, or
you go look up the numbers and do the calculations by hand,
which may take you something like 20-30min.

All these cases are very easy with UTF-8.

>seems like if the limits are specified in any other form than
>"maximum number of symbols" then users will find this mysterious.

Well, because different users may even see different numbers
of symbols, there will always stay some bit of mistery.


>and for many languages UTF-8 requires different octet lengths for
>different characters.

Yes. But it's something that isn't too difficult to know, if
you want to know. The toughest case is probably Vietnamese,
because you have ASCII (1 byte), Latin-1/Latin extended A
(2 bytes) and Latin Extended Additional (3 bytes).

Regards,   Martin.