Katakana Middle Dot again (Was: tables-06b.txt: A.5, A.6, A.9)

[Yoshiro YONEYA]

Dear Vint,

> First, I assume that it is not intended to permit Katakana Middle Dot  
> without the presence of at least one other Han, Katakana or Hiragana  
> character

Right.

> Second, I assume the same would be true of the Ideographic Closing Mark

Right.

> Third, I have not been able to understand the utility of allowing a  
> label consisting only of IDEOGRAPHIC CLOSING MARK and KATAKANA MIDDLE  
> DOT.

I can't negate demand for such sequences.

Regards,

-- 
Yoshiro YONEYA <yone at jprs.co.jp>

On Fri, 7 Aug 2009 07:47:38 -0400 Vint Cerf <vint at google.com> wrote:

> Harald is asking the key question here:
> 
> > Is it reasonable to assume that there exists the reasonable desire  
> > to register labels that contain IDEOGRAPHIC CLOSING MARK and  
> > KATAKANA MIDDLE DOT, but no other Han, Katakana or Kana character?
> 
> First, I assume that it is not intended to permit Katakana Middle Dot  
> without the presence of at least one other Han, Katakana or Hiragana  
> character
> 
> Second, I assume the same would be true of the Ideographic Closing Mark
> 
> Third, I have not been able to understand the utility of allowing a  
> label consisting only of IDEOGRAPHIC CLOSING MARK and KATAKANA MIDDLE  
> DOT.
> 
> So I am as puzzled as Harald on this question.
> 
> vint
> 
> 
> On Aug 7, 2009, at 7:29 AM, Harald Tveit Alvestrand wrote:
> 
> > Yoshiro YONEYA skrev:
> >> Dear John,
> >>
> >>
> >>> U+3005 and U+3007 are identified as "Han" in the Unicode table
> >>> Scripts.txt, so need no special treatment.
> >>>
> >>
> >> Exactly!
> >>
> >>
> >>> U+3006 (IDEOGRAPHIC CLOSING MARK) is listed as in "Common"
> >>> script in that table.   Without understanding the use of this
> >>> character, it is plausible that it would occur in a label that
> >>> consisted only of it, the middle dot, and, e.g., Romanji? If it
> >>> is not going to be used except when other ideographic characters
> >>> are present, there is no need to make an exception, although a
> >>> comment might be in order.  Remember that, as you suggested, the
> >>> test now requires only a single character that is unambiguously
> >>> Hiragana, Katakana, or Han.
> >>>
> >>
> >> U+3006 (IDEOGRAPHIC CLOSING MARK) is some kind of simplified form  
> >> of U+7DE0.  U+7DE0 is sometimes substituted by U+3006 when it is  
> >> used for meaning closing, therefore treatment of U+3006 is the same  
> >> with Han.
> > I reiterate the question:
> >
> > Is it reasonable to assume that there exists the reasonable desire  
> > to register labels that contain IDEOGRAPHIC CLOSING MARK and  
> > KATAKANA MIDDLE DOT, but no other Han, Katakana or Kana character?
> >
> > Again, we are seeking a justification for overriding an Unicode  
> > determination - I don't understand the reason for the determination  
> > that placed U+7DE0 in script "Han" but U+3006 in script "Common",  
> > but generally, we have tried to reduce the number of special  
> > exceptions to the rules determined by looking at Unicode properties  
> > as much as possible.
> >
> >      Harald
> >
> 
>