Glenn Adams wrote:
>The fundamental question is whether to encode distinctions to be made
>during the rendering process. A smart font could easily distinguish these
>behaviors without code distinction as long as either a language or font
>binding were present (in the case of mixed Arabic/Uighur text), or simply
>use one default behavior for mono-lingual Uighur text.
I was told that there are similar differences also in other cases of
languages written in the Arabic script.
I think that the correct solution is to leave the precise rendering rules to
a higher level protocol that would provide a language attribute. For example,
the proposed HTML I18N protocol. One of the roles of the language attribute is
precisely this, to provide for rendering differences between languages.
Thus, we would not hurt the main principle, that Unicode codes characters, not
shapes.
Jonathan
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:31 EDT