Re: combining: half, double, triple et cetera ad infinitum

From: Asmus Freytag <>
Date: Mon, 14 Nov 2011 12:29:11 -0800

On 11/14/2011 7:30 AM, Naena Guru wrote:
> Unicode was created for a commercial reason, particularly for the
> benefit of its directors.

This statement, not backed up by evidence, indicates a rather
rudimentary understanding of the forces that were behind the creation of
the universal character set. Coming as it does without details, it
doesn't add much to the discussion.

I can understand that some people are frustrated, when, over two decades
after the basic design was hashed out, the implementation is still not

The reason for that has to be sought in the inherent complexity of
writing systems. Even apparently very simple writing systems can have
surprising complexity when you try to support all areas of use and in
high quality typography.

Unicode is not just a character set, it provides a common framework for
organizing and formalizing much knowledge about writing systems. As a
result, there is now far more information and more accessible
information about writing systems than when Unicode was started. Had all
this information been available 20 to 25 years ago, there's a fair
chance that the design of a universal character set would have differed
in some aspects from what we now know as Unicode.

But even with the benefit of such hindsight, the sheer complexity of
writing systems remains. This complexity means that there will most
likely never be any implementation that supports all writing systems to
their fullest (highest quality). Every practical implementation will
subset somewhere, and that means there's no guarantee that any two
implementations will faithfully interoperate.

It might be that a different design would have made some implementations
easier, but I strongly suspect that the limitations I described here are
fundamental, so that the expectation would be that a different design
would have merely made other tasks more difficult while making certain
ones easier.

Received on Mon Nov 14 2011 - 14:31:24 CST

This archive was generated by hypermail 2.2.0 : Mon Nov 14 2011 - 14:31:25 CST