In the end we get a bunch of hacks to make the system work (in this regard
Microsoft is no better: they have to create new LCIDs to capture this
complexity too). This isn't surprising, given the nature of languages and
cultures.
Differences in writing systems are much more problematic than the
Norwegian example. The Simplified/Traditional Chinese thing leaps to mind,
of course, and variants are very unsatisfying (how long before the @EURO
variant goes away? what happens in locales with a two variant
requirement? And, of course, variants aren't based on *any* standard...)
On the other hand, it *is* good that the main system is predictable and
based on easily available standards. So long as everyone agrees to what
things mean we'll be fine.
I imagine that the nb/nn codes will eventually predominate... except that
this means many Norwegian locale structures are going to be messy: Java,
for example, will default from nb to default, not from nb to
nn-if-it-exists or from nb to no. On the other hand, the Bokmal/Nynorsk
mess now *has* a solution that is a standard. It's up to implementers to
make sense of it.
Best Regards,
Addison
On Thu, 31 Aug 2000 Marco.Cimarosti@icl.com wrote:
> Addison P. Phillips wrote:
> > This is a weakness of the locale model used on the Web and most UNIX
> > systems: the hierarchy is based on the ISO 639 language codes
> > and the ISO 3166 country codes. It doesn't cover such minutiae as
> > "inside-a-country" variation easily nor does it deal well with subtle
> > issues like this.
>
> This may be even more serious when such sub-locales have an impact on
> display.
>
> It is the case of Malayalam, that has two typographical styles
> ("traditional" and a "reformed") which are both in current use.
>
> For instance, the Open Type specs for Indic scripts
> (http://www.microsoft.com/typography/otspec/indicot/default.htm), in
> Appendix A (http://www.microsoft.com/typography/otspec/indicot/appen.htm)
> defines the two different styles:
>
> "MAL " Malayalam (Old Style)
> "MLR " Malayalam (Reformed)
>
> But there is not much use to it until it will be possible to tag the text
> correspondingly (e.g. ml-OLD ml-NEW).
>
> _ Marco
>
>
>
>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:13 EDT