I very much agree with Ken.
>Hey, unless I am mistaken, some 10's of millions of Asian system users
>already do this kind of stuff all the time, and consider it just more of the
>typing. The key is how seamlessly the transition between input methods
>can be done, and there is a lot of ergonomic and UI design that goes into
>making this as transparent as possible to users.
>Frankly, I consider most Americans and Europeans to be "input-challenged"
>when it comes to thinking about how to make use of keyboards to quickly,
>efficiently and transparently get the characters they want into text,
>once the problem gets beyond the scope of adding a few accented Latin
>letters with accent dead keys.
Apple's Worldscript (official, see
http://www.apple.com/macos/multilingual/languagekits.html and third party,
see http://www.indigo.ie/egt/earra_bog/apple) does it easily. For
eight-bits hex entry on a Mac please download
John Cowan was arguing (I think) that it did make sense to have Arabic
hexadecimal equivalents just to prevent them from having to use a Latin
keyboard. I think that unless there are TRADITIONALLY-ACCEPTED conventions
in other scripts where non-Latin characters are used in hexadecimal
notation, that it is a real question as to whether such traditions can, or
should, be introduced. It poses real problems for normalization, doesn't
it? Does it mean that we can expect to find data with Gurmukhi digits 0-1
and Cyrillic A-BE-GHE-DE-IE as equivalent to strings of 0-F? I know, it's
just an "input method" but the error seen already with Cyrillic
A-BE-GHE-DE-IE vs. A-BE-TSE-DE-IE already highlights the problem, which is
how the _notional_ equivalence of Latin to "other" letters can differ from
language to language or person to person.
-- Michael Everson, Everson Gunn Teoranta ** http://www.indigo.ie/egt 15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland Guthán: +353 1 478-2597 ** Facsa: +353 1 478-2597 (by arrangement) 27 Páirc an Fhéithlinn; Baile an Bhóthair; Co. Átha Cliath; Éire
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:46 EDT