[Date Prev][Date Next][Subject Prev][Subject Next][
Re: languages, unicode
- Subject: Re: languages, unicode
- From: Dorothy Day day@xxxxxxxx
- Date: Tue, 23 Apr 1996 10:44:38 -0500 (EST)
On Tue, 23 Apr 1996, Nathan Sivin wrote:
> I agree with Wolfie that users ought to have a choice of IBM or ANSI
> character set. Locking the old one out is not a good move in an
> application that sells itself largely on customizability.
It would be a good idea, and not very difficult, to provide
import/export options to convert among various character sets as needed.
Nota Bene's ability to import from/export to national alphabets in
Cyrillic, Greek, or Hebrew is an example of how readily this can be done
(and how beneficial for exchanging vernacular email with persons who
regularly use those national character sets). Mac WordPerfect's ability
to convert between the ANSI character set used on the Mac and that in
Windows is another. And MSWord's miserable failing at the same task is a
model of how not to do it.
Already, users of tools like ProCite have to deal daily with how to
import particular accented characters from/export to word-processors
that are not geared toward cooperation with external programs. At least
Ibid is (at least currently) aware of the character set produced by
XyWrite. But move to some other program you find useful, and the the
conversion hassles arise afresh.
Standardizing the handling of ANSI character sets would be a prudent
first step towards eventual incorporation of Unicode capabilities. Begin
with the most common ANSI/ISO sets, then add others as demand occurs.
Eventually each of those codes can be coverted directly to ISO/UTF-8
2-byte equivalents for use in a 2-byte-enabled version of the program.
It is of course possible for individual users to cobble xpl routines to
convert particular characters they regularly encounter; but it's far
more useful (not to mention elegant) for Xy-other to build in character
> I would be delighted with a format in which each tag had a
> corresponding tag turning the feature off (like using MD-IT after
> MD+IT in the current version). That would greatly simplify
> conversions, including that to HTML. I believe KF's list of new
> features specified something of that sort.
Yes, definitely. As long as the option of complete reset with MD+NM is
available, being able to simply turn off a single attribute would
greatly increase flexibility and convertability.
School of Library and Information Science