Subject: Re: [Patch] More zh-TW translation updates from ThizLinux :-)
From: Andrew Dunbar (hippietrail@yahoo.com)
Date: Wed Dec 05 2001 - 16:45:09 CST
--- Rui Miguel Seabra <rms@multicert.com> wrote: > On
Wed, 2001-12-05 at 12:28, Andrew Dunbar wrote:
> > > Actually, Macau still has a _lot_ of portuguese
> > > speakers :)
> > Well in that case we should also have pt_MO fall
> back
> > to pt_PT or even pt_BR if need be. But I remember
>
> Falling back to pt_BR has three big disadvantages:
> 1) the standard Portuguese is from Portugal
> 2)
> [rms@roque abi]$ ./dumpstrings.pl pt-BT >
> ../.strings.html
> Missing strings:
> pt-PT: DLG_PageSetup_ErrBigMargins
> pt-PT: DLG_Lists_Hebrew_List
> en-US:
> pt-PT: 2
>
> [rms@roque abi]$ ./dumpstrings.pl pt-BR >
> ../.strings.html
> Missing strings:
> (...)
> en-US:
> pt-BR: 713
>
> 3) in light of my efforts (check 2 -- oooh... 2 new
> strings) I'd
> consider an insult to see the fall back be pt-BR :)
>
> > talking about this before for cases such as
> British
> > English and Canadian French.
>
> We'd have to have an intelligent way of knowing
> which is the standard
> for the language, in order to do it properly.
Well the idea is that it's theoretical. If for some
reason the best locale hasn't been installed or is
broken in some way, fall back to the next best, keep
falling back till we have something.
Of course it's better to fall back to pt-PT if it's
there and does work but if it not, falling back then
to the 3rd best, pt-BR is theoretically the better
algorithm for the general case. Some point in the
future we may have a more up-to-date BR than PT and
still no MO and it's better to have algorithm which
works all the time than tuning the algorithm to the
current data files every so often.
You grok what I'm saying don't you? (:
Andrew Dunbar.
=====
http://linguaphile.sourceforge.net
________________________________________________________________
Nokia 5510 looks weird sounds great.
Go to http://uk.promotions.yahoo.com/nokia/ discover and win it!
The competition ends 16 th of December 2001.
This archive was generated by hypermail 2b25 : Wed Dec 05 2001 - 16:45:11 CST