From: vunzndi@vfemail.net
Date: Thu Mar 22 2007 - 09:54:33 CST
It makes some sense to me in so far as it does not duplicate any
existing VS, not however if it sets a precedence for future
collections. That is why I suggested the non-duplicate VS for any one
code point, but rather the same VS to be allowed to be named in
different collections.
Quoting Andrew West <andrewcwest@gmail.com>:
> On 21/03/07, vunzndi@vfemail.net <vunzndi@vfemail.net> wrote:
>>
>> In IVD Adobe-Japan1 many codepoints with just one glyph are included,
>> which since adobe is the first to registar a collection makes sense to
>> me.
>
> I'm not at all convinced it makes any sense to define 12,040 variation
> sequences for characters that do not have any glyph variants (i.e.
> 12,040 variation sequences where the glyph defined by <U+XXXX,
> U+E0100> is the same or almost the same as the representative glyph
> for U+XXXX, and there are no other variation sequences defined for
> U+XXXX).
>
> The reasons given for this bloat of variation sequences are:
>
> 1) For the convenience of Adobe -- so that all Adobe Kanji can be
> "uniquely and explicitly identified without referencing their default
> (IVS-less) encoding" (why? -- maybe it would be better to encode a NO
> VARIANT character that could be applied to any character without need
> for registration or a defined variation sequence)
>
> 2) For the sake of forward compatibility -- "because kanji may be
> added in future Adobe-Japan1 Supplements that may be variants of such
> kanji" (we could cross that bridge when we come to it)
>
> I'm not convinced that either or both of these reasons necessitate the
> encoding of 12,040 redundant variation sequences. It may be convenient
> for Adobe, but I strongly doubt that it will be in Unicode's best
> interests.
>
> Andrew
-------------------------------------------------
This message sent through Virus Free Email
http://www.vfemail.net
This archive was generated by hypermail 2.1.5 : Thu Mar 22 2007 - 09:57:15 CST