From: Christopher John Fynn (cfynn@gmx.net)
Date: Wed Dec 17 2003 - 22:52:55 EST
"Dean Snyder" <dean.snyder@jhu.edu> wrote:
> Recently I have had second thoughts about encoding complex signs.
> Modification of base, or simple, signs was a productive process for
> making new signs in the earlier periods of cuneiform usage, and included
> such modifications as adding or subtracting wedges, rotating signs,
> infixing signs, etc. (For some examples of how the ancient scribes
> modified base signs to form new complex signs see <http://www.jhu.edu/
> ice/basesigns/>.)
> Instead of encoding all 875 post-archaic, base and complex cuneiform
> signs, we could instead encode the 280 base signs plus a dozen or so sign
> modifiers. (I am not including in these approximate figures the 75 or so
> numerical signs being proposed for encoding.) This would be somewhat
> analogous to encoding "a", "e", the acute accent, and the grave accent
> instead of encoding "a with acute", "a with grave", "e with acute", etc.
This fits in best with the Unicode charater encoding model and is definitely
the way to go, particularly if the script was productive.
If additional complex signs are found you will then be able to represent them
straight away and won't have submit a proposal to add an additional character,
wait for it to be accepted & get encoded, and then wait support for it to
appear in applications and fonts (a proccess which usually takes several years)
> Encoding base signs with modifiers would more closely mirror, in the
> encoding, the way the script system itself actually worked and it would
> more easily accommodate modern research in archaic cuneiform, a stage in
> cuneiform script development we have all decided not to encode for now
> due to the current provisional state of its scholarship. By providing in
> the encoding the base signs along with their modifiers cuneiformists
> working in archaic and other periods could generate newly discovered or
> newly analyzed complex signs ad hoc, without having to go through the
> time-consuming and expensive Unicode/ISO standardization process.
> Compound and complex sign realization would then simply be a matter of
> the coordination of input methods with fonts, something now doable by end
> users with modern computer operating systems. (This, of course, assumes
> that we are more likely to find new combinations and modifications of
> existing base signs than to find new base signs themselves. At any rate,
> when we do find new base signs we need to encode them anyway.)
I think it is always a good idea to closely mirror in encoding the way a script
system actually works - and break it down into primitives or base characters,
combining marks and modifiers
It might be helpful to at how "smart-font" systems like OpenType and AAT/ATSUI
are already used for rendering complex scripts and to try and think of the
features and lookups a Cuneiform font using this sort of technology might use.
> To most cuneiformists, of course, the encoding underpinnings would all be
> hidden by input methods and fonts. One would simply type the expected
> SHUD3 and the input method would map it to 3 code points, KA INFIX and
> SHU (mouth sign with hand sign infixed), and the font would render it as
> one complex sign (meaning "to pray").
This is perfectly feasible.
> And from a practical point of view encoding only the base signs and their
> modifiers would be easy for us to do - we need only remove the complex
> signs from our lists and add the 13 or 14 modifiers.
This seems to be the right approach.
- Chris Fynn
This archive was generated by hypermail 2.1.5 : Wed Dec 17 2003 - 23:38:32 EST