Re: UTF-8, ISO C Am.1, and POSIX

From: Keld J|rn Simonsen (
Date: Thu Aug 28 1997 - 15:48:18 EDT

Glenn Adams writes:

> A good case can be made that the only reason they needed a code set
> independent design was because Unicode did not exist at that time. It's
> existence now obviates the POSIX design philosophy. Eventually, all
> systems will migrate to Unicode/10646 as their default character set.
> As they do this, they will rely on the defined semantics of Unicode/10646
> and begin migrating away from the hands-off approach taken by the POSIX
> model. The POSIX model will eventually become irrelevant.

I also see more and more systems going to UCS and using the
semantics defied on UCS, and further definitions on UCS such as
conversion and sorting. But I do not see this as eliminating POSIX,
I see POSIX and UCS as a perfect match and POSIX can be used to promote
UCS and definitions on UCS to other systems not using UCS. And I believe
that systems not running UCS will be around for a long time to come.

> Of course the pace of this transition is certainly an arguable (and unknown)
> datum at this time. The global expansion of the Internet and the Web will
> be a strong catalyst in this process. Closed systems that don't care about
> or don't want global access can continue to benefit from the POSIX model.
> I suppose it's ironic that a model designed in the context of "open systems"
> will eventually be usefuly only for "closed systems."

Well, I see POSIX systems as the heart of the Internet now, and
actually also as the birthplace of Internet. And I believe POSIX
will stay as a very significant part of the Internet, certainly
upgrated to support UCS.

I see POSIX and UCS as complimentary, not antagonistic.


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:36 EDT