Fwd: RE: [ChinLing] Searching for suitable database systems

[Guest guest]

Dear All

My query on the subject has begun to bear fruit.

I now have my hands on a Filemaker version for Taiwanese, but it will

take some time before I have control of it. But these seem to be much more

to this topic than I thought.

 

Here for those interested in the discussion, is an answer from a

similar mail to the ChinLing group on

ChinLing

they work on Sinitic Linguistics....

 

I am also very happy to have gotten to know that there is this FORUM

only for these kind discussions, so I plan to move this discussion

over theresince it would be more in place for that enviroment:

http://www.paradigm-pubs.com/forums/cyber/index.html

 

So thanks to Robert L. Felt for giving me this info.

Also many thanks to Wai Keet Hew at the ChinLing list for following

mail of info, I am almost hoping that he is giving answers to some of

Roberts topics, but this topic is so vast that there will always be

more

to it :

-----

In ChinLing , wk.hew@s... wrote:

Welcome to the list, Sebastian. I hope you will find it useful and

interesting, although it has been very quiet lately.

 

From your post, I gather that you have attempted to use a western

version of FileMaker encoding the Chinese in a two-byte encoding such

as GB or Big-5. At this stage, let me recommend Big-5, for GB is

missing many of the characters required for . Of

course, this deficiency has largely been remedied by the newer GBK

encoding, the extension to the original GB.

 

Having said this, I would suggest that you avoid such encodings as GB

and Big-5 altogether, for they do not work well together with most

European languages. English is more-or-less all right, although even

so, texts with mixed Chinese and English can sometimes prove

troublesome. However, combining Chinese with most of the other

European languages usually results in a disaster. In particular,

support for letters with diacrical accents, e.g., the grave, acute,

and circumflex accents and the diaeresis found in such languages as

German and French, not to mention the less familiar accents found in

other the languages, is less than perfect. In fact, the Big-5 and GB

encodings use the same code-space for Chinese characters that is

usually used for accented characters, so the two cannot co-exist

unless they are carefully separated by means of differing style or

font settings.

 

Fortunately, another solution is quickly becoming commonplace, i.e.,

Unicode. Although support for this universal encoding is not yet

universal, it is rapidly gaining acceptance. At the moment, obtaining

Unicode fonts with the full code set is none too easy, although I

expect the situation to change in the very near future. With Unicode,

there is no problem with conflicts between Chinese characters and

diacritical accents, for they are encoded in the same system by

unique codes. Moreover, Unicode has the advantage of encoding many

Chinese characters missing from GB, GBK, Big-5 and Big-5 Plus, which

encode some 7,000 13,000 21,000 and 19,000 characters respectively.

In fact, the latest version of Unicode includes every character found

the the historical and authoritative Kangxi Zidian

(±dº3|r¨å),

which contains over 45,000 characters. This is particularly important

for , which makes use of some fairly obscure

characters not usually found in small dictionaries.

 

I should therefore suggest that a Unicode compliant database, e.g.,

MS Access 2000, would best serve your needs. (I mention Access only

because it is the only one I am aware of which is Unicode compliant;

I am not recommending it, for I have reason to believe it is quite

slow and a little buggy. However, I am sure there would be others,

e.g., Paradox, Oracle, etc. Perhaps a new version of Filemaker would

be Unicode compliant.) I hesitate to recommend a particular database

system as I am not that familiar with them, and, the choice would,

moreover, depend on the size of your database. I am also unsure of

how well Macintosh systems support Unicode. Nevertheless, I think

Unicode is the best solution for your case where Chinese and western

languages need occur side-by-side. It overcomes the problem of mixing

double-byte and single-byte languages.

 

I hope this helps,

 

Wai Keet Hew.

--- End forwarded message ---