On 29/07/2003 10:46, John Hudson wrote:
At 06:11 AM 7/29/2003, Karlj�rgen Feuerherm wrote:
Well, that was precisely the question. Are we talking about a mere
preference of visual effect or an actual difference in (original)
text--that
is, an intended semantic differentiation?
A good question, and one for which I would like to know the answer. I
have Unicode text from Libronix, derived from the Westminster
Theological Seminary text, that clearly encodes holam_vav distinctly
from vav_holam, indicating that someone thought it was important
enough a distinction to carefully make during the original WTS
transcription. Fonts for this kind of text encoding need complex
contextual lookups to prevent the holam from attaching to the
preceding consonant. The same fonts will also display the vav_holam
encoding correctly, i.e. without a distinction. So from a display
perspective, this is one issue that is already solved: the question is
one of document encoding and comparison.
John Hudson
So, let me clarify. You are proposing that the order vav-holam be used
for the consonant vowel sequence, and that both vav-holam and holam-vav
be considered valid encodings for the vowel only version? OK I suppose
if we can somehow ensure that these are treated identically for
collation, searching etc (though we cannot of course make them
canonically equivalent). An even more clever font would then have the
option of detecting which vav-holam sequences are actually the vowel and
displaying accordingly, thus meeting the objection that the visual
display should depend on the font etc rather than on the choice of
otherwise equivalent encodings.
--
Peter Kirk
[EMAIL PROTECTED]
http://web.onetel.net.uk/~peterkirk/
