Around 22 o'clock on Jun 29, Yu Shao wrote:

> >Tagging GB18030 fonts as suitable for traditional chinese seems like a 
> >mistake; the glyph forms are more likely simplified, and it would be 
> >
> Agreed.

This is reassuring.

> As gb18030 is compulsory from government, I think we should just treat 
> gb18030 as Simplified Chinese, and all fonts from now on should gb18030 
> compliant. For these fonts, the new included Chinese  minority Yi and 
> Tibeitan characters would do.

The trick that I use to distinguish between simplified Chinese and
traditional Chinese targeted fonts is not whether they cover a significant
fraction of the Unicode codepoints mapped from gb18030, but whether they
cover nearly all of the Unicode codepoints from mapped from Big5.

The algorithm looks like:

        if (covers_much_of (gb18030))
                font supports simplified Chinese
        if (covers_almost_all_of (Big5))
                font supports traditional Chinese
                font does not support simplified Chinese
        if (covers_almost_all_of (JIS))
                font supports Japanese
                font does not support simplified Chinese
        if (covers_almost_all_of (Korean Wansung))
                font supports Korean
                font does not support simplified Chinese

Nearly all Han fonts cover as much of GB18030 as those targeted for 
simplified Chinese, but (in my limited sample) simplified Chinese fonts 
cover only a small fraction of all of the other Han encodings.  Except for
Arial Unicode, which covers all of the encodings nearly completely.  

Remember that this whole mess is only needed for fonts which don't have 
any OS/2 codePageRange bits set; the hope is that new fonts covering more 
of the Unicode range will be provided in TrueType or OpenType format so 
that this particular hack can be avoided.

> But the very popular Microsoft's Chinese simsun font now, is actually a 
> gbk font.

This is a TrueType font and so the above hacks don't apply.  Are there new 
GB18030 fonts being distributed in formats that don't include the OS/2 
codePageRange bits?

Keith Packard        XFree86 Core Team        HP Cambridge Research Lab


_______________________________________________
Fonts mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/fonts

Reply via email to