CompuServe Messages

#Chinese nation

    02-May-95 04:01:06
Sb: #195423-#Chinese nation
Fm: Black Belt Systems 76004,1771
To: Bart Van Bockstaele 100574,2352
Bart, PC keyboards, like Amiga, Mac and certainly things like the venerable c64 that's wandered in and out of this thread, are all far too limited in keycaps to handle character sets like these. Hangul could produce (grabbing calculator) 14*21*14=4116 distinct characters from it's (14+21=35) basic character components, and that doesn't account for double characters and some of the rarer character components at all. So this problem exists everywhere – and a 4000+ key-keyboard isn't a solution, it IS a problem! <grin> The solution is to map the keys on the keyboard to produce the 35+ components and "build" the characters on the fly. There is a standard Hangul-component to keyboard mapping, and the software I use (on both the Amiga and the PC) use this. The difference is that once the character is built, the PC software can handle it, in general. The Amiga software has to be a specific application designed to do so. The Hanja (Korean term for Chinese characters) character set can only be described as huge. Some character dictionaries run to 50,000 (yeah, fifty _thousand_) characters! Can you imagine the keyboard? A more or less square keyboard would be 224×223 keys! Egads! Well, the problem's not as bad as it seems because most writing is done with a fraction of those 50000. But still, you want to be able to say "peanut brittle" when you need to… 🙂 There are multiple methods for handling this; the easiest way I've seen is one that relies on the root _sound_ of the character in Korean. You type in the Hangul for the syllable "Kwuhn", and the software will (if you press the appropriate key) pop up a windows with a scrolling list of Hanja that matches that sound. Probably straight Chinese would use a similar technique, in order to accomodate the limitations of the keyboard which you have correctly identified as the bottleneck in getting such a huge character set into play. Both Korean and Japanese have written language methods that use "letter" characters – they represent sounds as do our own familiar 26 or so letters, but both also have the Chinese set… one cool thing here is that when you learn the Chinese characters, you've gone a _long_ way towards being able to read Korean, Chinese _and_ Japanese characters… though it sure isn't all the way. Letter type writing is much easier to "fold" into a keyboard than is character-is-a-word types of writing. But both have to be done in order to serve in the task we've been talking about. And then the OS, and applications running under the OS, have to recognize and properly handle these huge character sets. –Ben …via AutoPilot