Mixing languages and macOS X was: Sorting lines with diacriticals

Manuel Companys mcompanys at mac.com
Sun Jun 9 15:21:23 CDT 2002

Le 6/06/02 17:40, « Mark Mitchell » <mark_mitchell at kmug.org> a écrit :

> Manuel writes:
> But we have had this for near 15 years! I started with a mac 128; when the
> 'mac +' popped up, we could parameter the language in the control panel
> --------------------
> Not built in to the OS we haven't! (depending on your language.)
Right! Of course I meant the alphabetical scripts, and more specially
"western european languages". All "western european languages" used to get
exactly the same fonts, including all their characters, and the little flag
menu allowed to change on the fly, fitting the national keyboard. For
another simple-alphabetical language (cyrillic, greek) you had just to put
the fonts, and the kchr resources (flag, keyboard mapping etc.). THIS
facility, was not always shipped in the US: you had to ask for it

Moreover these fonts also had all the diacriticals needed in almost all the
other languages written with the latin alphabet (or in cyrillic, if we are
speaking of cyrillic fonts). They were not always accesible from all
national keyboards, and they printed as a separate letter. But you could use
resEdit to set a "0 onset" so that the diacritcal will print upon the
letter, or even use a simple utility as (FontMonger) to create the
diacritized chars, give them the ascii value of some unused char, and use
resedit, very user-friendly, to make a new ressource (and little flag) for
your keyboard.

> For Japanese 
> and other 'difficult' languages cumbersome
> 'language kits'  were  required and they seldom worked very well.  Under OSX
> you can run, for example, traditional Chinese,
> Hebrew and Japanese, (oh, and English of course, and have the main menus in
> any one of these languages...) switching between
> them with relative ease.  I'd like to have seen you attempt that 15 years ago!

Are you not confusing macOS X vs macOS and Unicode vs conventional system?
With the conventional 1 byte coding you have 256 spots (only 128 in DOS and
Windows) which leaves more than 200 hundred possible chars.

UNICODE uses TWO bits per char which extends the possibility to over 40000!
If your script system is syllabic (katakana / hiragana, devanagari) OR needs
a lot of contextual variants for each letter (arabic) OR mixes the two
difficulties (tibetan), you need two bytes; and if you use some thousands of
ideograms (chinese, kandji) the need is still worse.

We are pretty happy that the Phenicians came up with the alphabet!*

You can switch between ANY language with macOS using UNICODE.

> I repeat, OSX is a multi-lingual marvel.

I agree!   UNICODE certainly works much better with macOS X than with system
9, and, besides, switching between the menus was impossible.

Although I still think that the real multi-lingual marvel was the early mac
typing 12 mixable western european languages, with the system, the
application and the files fitting in a single floppy: in a 400 bytes one!
And with the technology available in 1984  for that matter!



* But Bill Gate's Entourage obliges us to use UNICODE for the Euro symbol
(ASCII 219) which we have had in our macs for years and even since the
beginning with the symbol "currency" it replaced. And this while HE accepts
the $ (36), the ¥ (180) and even the £ (163) which is bound to disappear: we
are MAD! ;-)

More information about the use-livecode mailing list