[sword-devel] To cantillate or not to cantillate
Fri, 22 Jun 2001 06:05:59 +1000
Chris Little wrote:
> The shortcomings of Unicode support for Hebrew have been mentioned
> before, and most of them relate to its support for cantillation marks.
> Font support for those marks that have Unicode codepoints is very poor.
> Code2000, Cyberbit, and Arial Unicode MS are the only 3 fonts I have
> found that support them. The greatest problem in Windows is that
> Windows will automatically select another font that includes support for
> a character set when the current font does not support the set. But
> Times New Roman, Miriam, David, and many other fonts include support for
> the consonants and vowel points, so they claim to support Hebrew despite
> not supporting its entirety. But they render blocks instead of
> cantillation marks.
> In the future I plan to add render filters to toggle cantillation
> points, vowel points, and accents in Hebrew and Greek. But we have a
> release coming up now, so I'd like to know what others think we should
> Do we release a text that does not include cantillation marks despite
> being able to include them now simply because they will not render
> correctly by default and we haven't yet built a mechanism for turning
> them off? Or do we include the marks and set the font to "TITUS
> Cyberbit Basic"?
We had a discussion about this a while back on the Logos beta newsgroup
(lbs.titus), and Bob insisted that the 'right way' (TM) was to use
Unicode. What it means is that the Greek and Hebrew texts are unusable
(IMO) until a complete font is available.
I think that if we have a text that looks right now, we should maintain it
in its existing form until suitable technology is available to convert it
to a more portable mechanism. In the meantime, the non-portable, smaller,
correctly-marked text is preferable. (Of course, if it's a new text which
hasn't been released in either form, then people can probably live with a
less-than-ideal display until the fonts are available.)