>... probably a bad idea
I would go further: it's wrong. The theory of operation for Hebrew
in Unicode is that final forms are explicitly there or not. You do
not present (much less change to) final forms at the end of words
unless there's a final form there. I think this was more or
less clearly stated in the description of the Hebrew block in the
Unicode Standard book.
In the Hebrew writing system (as used by Hebrew, Yiddish, et al),
the Peh character in particular is used with its non-final form
to indicate the P sound, as opposed to the F sound. In the
Yiddish language this is extremely prevalent (e.g., the spelling
of the word "shlep"), but it's also encountered in Hebrew, e.g.,
for the spelling of the name Philip.
The only place to consider anything like this is if building an
input method (IME, front end processor, keyboard event handler,
etc.), where you might want to consider doing this. But then
your typers could not use a standard Hebrew keyboard layout (or
almost any of the variants), which always have explicit final-form
characters. It would be roughly like presenting an English keyboard
that was "smart" about capitals and you just typed in one case and
the input method figured out whether you wanted a capital or small
letter -- a lot of smarts, but does anyone really want this? (No.)
At any rate, the question was about character shaping, and I think
it's clear this should not be done for Hebrew in Unicode.
Mark David
Moderator, UYIP (Understanding Yiddish Information Processing)
http://www.uyip.org
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:46 EDT