🚀 go-pugleaf

In article <1993Jan8.094119.6795@prl.dec.com> boyd@prl.dec.com (Boyd Roberts) writes:
>In article <1993Jan7.065611.15193@fcom.cc.utah.edu>, terry@cs.weber.edu (A Wizard of Earth C) writes:
>> 
>> Again, sorting can not be safely tied to character set lexical order for
>> all languages.  I disagree with Boyd here.  Localization with Unicode is
>> a piece of cake.  Unicode allows it to be entirely data driven, with no
>> locale-specific algorithms or hard-coded data.
>
>Maybe you should check your attributions, but I have never advocated
>that the code values can be used to do lexical sorting in the general
>case -- far from it.  Your above paragraph contains a contradiction
>as I read it.

Sorry if I have misattributed anything.  The disagreement was regarding
the ease of localization.

>Tell me how I sort on stroke count in Unicode without ``locale-specific
>algorithms or hard-coded data''?

Locale-specific sorting can be done with a generalized algorythm which
is itself data-driven.  The *data* doing the driving, on the other hand,
is *entirely* locale specific.

Localization (in terms of providing a native language environment of
commands, interface text, and error messages) can be totally data driven
using locale specific message catalogs.

None of this requires either direct manipulation algorithms (ie: not data
driven) or hard-coded data (as in constant character strings compiled into
programs.

In previous posts, I (and others) have pointed out why it is fundamentally
impossible to take all sorting issues into account with simple lexical
ordering because of multiple sorting procedures within a given language.
This is because the locale specific character set binds the sorting order
to the lexical order within a single language (unless multiple character
sets are provided for each language with more than one sort order, and
the character set is thus bound to the locale, not to the lexical ordering
within a particular character set for each language.

Your stroke count example is well taken, since it is not the only sorting
order in Chinese (for instance, sorting on radicals is frequently used),
but stroke-count (and direction) sorts are highly useful even in English
for handwriting recognition systems.

>Re: `san'
>
>    It can be used on either the family name of the first name.
>    This assertion is based on current common practice in Japan.
>
>    If you knew Ohta-san reasonably well (ie. friend or colleague)
>    and were contemporaries you could use `kun' instead of `san'.

So, out of curiousity, is "Ohta" his family name or his first name?  I had
assumed it was being used as an honorific (ie: Mr.), in which case the
correct usage in his particular case is still dependant on whether he
was ordering his name for Japanese or for English in his signature.


					Terry Lambert
					terry@icarus.weber.edu
					terry_lambert@novell.com
---
Any opinions in this posting are my own and not those of my present
or previous employers.
-- 
-------------------------------------------------------------------------------
                                        "I have an 8 user poetic license" - me
 Get the 386bsd FAQ from agate.berkeley.edu:/pub/386BSD/386bsd-0.1/unofficial
-------------------------------------------------------------------------------