Path: utzoo!utgpu!water!watmath!clyde!bellcore!rutgers!ucsd!ucsdhub!hp-sdd!hplabs!hpda!hpcuhb!hpcilzb!hpcea!hpausla!cjh
From: cjh@hpausla.HP.COM (Clifford Heath)
Newsgroups: comp.emacs
Subject: Re: US PC programmers still live in a 7-bit world!
Message-ID: <380001@hpausla.HP.COM>
Date: 18 Jul 88 05:00:04 GMT
References: <1988Jun22.223158.1366@LTH.Se>
Organization: HP Australian Software Operation
Lines: 27

> characters I've missed any reference to the X-OPEN groups Native
....
> only a matter of time before the American computer manufacturers have to
> follow .

Americans (like anyone) can be pretty parochial, but don't forget that
X-OPEN designed its standard around **Hewlett-Packard's** implementation
of Native Language Support, which was a working system six years ago.

X-OPEN is a good standard as far as it goes, but when you try to build a
DBMS that uses it, you run into problems that are (mostly) not present
in HP's implementation, mainly due to the lack of a distinction between
data language and user language, and an inadequate definition of how
such multi-lingual systems can operate.  Remember that when you sort
something, it only appears sorted as long as you look at it using the
same collating sequence.  The Btree indices used in most databases
become corrupt if you update them using an incorrect collating sequence,
so you MUST distinguish between user language (for messages etc) and
data language.

In any case, for anyone wanting to sell software outside their own
country, X-OPEN should be compulsory reading.

Clifford Heath, Hewlett Packard Australian Software Operation.
(UUCP: hplabs!hpfcla!hpausla!cjh, ACSnet: cjh@hpausla.oz)
I didn't get paid to say this, so if you don't like it, don't tell HP.
Don't tell me, either.