Re: 1 in 1000

From: Robert A. Rosenberg (bob.rosenberg@digitscorp.com)
Date: Mon Apr 24 2000 - 14:56:17 EDT


At 06:29 AM 04/24/2000 -0800, Elliotte Rusty Harold wrote:
>Probably less than one person in a thousand today speaks a language that
>cannot be reasonably represented in Unicode.

Do you mean this to mean that "over 999 people in a 1000 can speak at least
1 language that is reasonably represented in Unicode", "less than one
person in a thousand today speaks only languages that cannot be reasonably
represented in Unicode" or "less than one person in a thousand today speaks
a language or languages that cannot be reasonably represented in Unicode
(ignoring the question of if they also speak one of those that can
reasonably be represented in Unicode)"? They are not the same question
since each counts a different characteristic. Case 1 only asks about
Unicode Languages (if you can talk one you fit), Case 2 asks only about
non-Unicode Languages (if you only talk them, you fit), while Case 3 asks
about the ability to talk a non-Unicode Language (you fit if you can talk
one even though you are also part of Case 1).

Case 2 is more restrictive than Case 3. It would be the choice of statistic
to use if someone wanted to stop adding new scripts ("why do so - there is
limited 'need' for new scripts since over 999 out of 1000 can get along
with the current set of scripts") while Case 3 is a better representation
of "Native Speakers" since it ignores the Multi-Lingual situation (it talks
about how many need added support to handle _A_ language that they speak).



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:02 EDT