tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: A draft for a multibyte and multi-codepoint C string interface

Steffen Daode Nurpmeso wrote:
> It would have been a good thing if there would have been consensus
> long, long ago, maybe in the mid-90s, resulting in a common i18n
> solution alongside of Unix.

There have been a consensus, since mostly the beginning of the `90s.
And that consensus is _against_ locales; they are seen like some other
"committee inventions" (like trigraphs) such as a necessary by-products
of the international standardization process which are basically best to
be forgotten.

There have been numerous proposals for ``better´´ solutions. As you can
easily guess, none of these proposals had long-term success; while they
were marginally better for some aspects of i18n, all also suffered of
flaws, drawbacks, lack of support for some domains or features, etc.,
which were used to put them down before being standardized.

My idea is that it does not seem that i18n+l10n can be solved by a
one-fits-all solution. And the fact the existing solution is perceived
as "wrong" by most people makes it even harder to design efficient
improvements to the current state of affairs.

> It is a pity that not even Universities put some effort into this.

It has been my experience that students was one of the two bigger
sources for such proposals (the other being software companies going
more global.)
However I have yet to see a rigorous study done by human sciences
(rather than computer science) students, from several sub-fields (rather
than by one individual), to serve as a base for improvement.

It could easily could have been what you thought about, just that I was
not sure.


Home | Main Index | Thread Index | Old Index