Re: A last missing link for interoperable representation

From: Asmus Freytag via Unicode <unicode_at_unicode.org>
Date: Wed, 9 Jan 2019 03:03:52 -0800
On 1/9/2019 1:06 AM, James Kass via Unicode wrote:

Asmus Freytag wrote,

> Still, not supported in plain text (unless you abuse the
> math alphabets for things they were not intended for).

The unintended usage of math alphanumerics in the real world is fairly widespread, at least in screen names.

(I still get a kick out of this:)
http://www.ewellic.org/mathtext.html

I wonder how many times Doug's program has been downloaded.

Whether it's "abuse" or not might depend on whether one considers the user community of the machines which process the texts to be more important than the user community of human beings who author, exchange, and read the texts.

   It's "abuse" because all these extensions for symbols only ever cover the ASCII range. Couldn't do actual German Fraktur text with the math alphabets (other than selected words). Same for italics.

A good test might be whether something would require duplicating the entire Unicode range to achieve full coverage (or at least significant subsets like multiple, entire scripts).


Real humans are the user community of the UCS.  It's up to the user community to determine how its letters and symbols get used.  That's the general rule-of-thumb Unicode applies to the subset user communities, and it should apply to the complete superset as well.


There's a cost to providing multiple ways of achieving the same effect and it cuts against the "uniqueness" in encoding that Unicode set out to achieve ("unique", "universal" and "uniform" were the three mantras that launched the standard).

A./


Received on Wed Jan 09 2019 - 05:04:08 CST

This archive was generated by hypermail 2.2.0 : Wed Jan 09 2019 - 05:04:08 CST