Ancestor-descendant distance

[Sean Crist]

On Tue, 24 Aug 1999, Dr. John E. McLaughlin wrote:

> The only problem with Unicode is that it isn't uniformly supported by
> computer programs.  Any Unicode-based effort would be restricted to the
> Internet Explorer browser as Netscape Navigator doesn't support it.  You
> either wish for the future and sacrifice present compatibility, or you make
> some compromises.

Well, I guess you've got to consider what the options are.  I think they
are the following:

1. Unicode.  Current drawback: not yet universally supported.

2. Make do with ASCII or the extended ASCII set (ISO-8859-1 Latin 1),
which forces you to do things like use @ for the Old English a+e ligature,
etc.  This really cramps you; even aside from the legibility problem, it
would be a challenge to work out a set of conventions for encoding data
from all the IE languages in a uniform and usable way with so few discrete
characters.

3. Use a hodge-podge of other encoding schemes.  This suffers the same
problem that Unicode currently does (no encoding scheme beyond ASCII has
universal acceptance), and suffers from the additional problem of
conflicting character codes, etc.  I'm sure that there are needed
characters included in Unicode but not included in any other standard
encoding scheme (e.g. the Gothic "hw" character, which does have a code in
Unicode but not in any other encoding standard I know of). Further,
support for these other encoding schemes it likely to gradually go away as
Unicode support grows.

4. Do nothing and wait until Unicode support is broader.  Drawback:
unnecessary delay to the development of useful materials!

Many of the major players in the industry are backing Unicode, and it's
pretty clear that support for Unicode will continue to increase.  If
predictions about the computer industry are worth anything at all, then
it's just a matter of time before Unicode editing, display, and text
processing capabilities are conveniently available to everyone.

As a short-term stopgap measure, one can always write scripts to convert
data from a Unicode-encoded database to some other encoding for users who
need it; and the programmer CDs from the Unicode consortium come with
extensive conversion tables for exactly this purpose.

In case you hadn't noticed, I think Unicode is the greatest thing since
sliced bread. :-)

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]