[Freeciv-Dev] Re: Localization: %s, %s, show me your secrets!

[Tuomas Susi <tsusi@xxxxxxxxx>]

On Sat, Mar 03, 2001 at 12:19:06PM +0100, Sini Ruohomaa wrote:
> Hi,
> 
> I'm working on the Finnish translation of Freeciv, and was wondering if
> there is some relatively convenient way of looking up what this and that
> '%s' can become, and where the possibilities are (for translating them) -
> the latter is mostly solved by knowing what %s can become and making a
> little string-search I suppose.

Using the cross reference at http://www.freeciv.org/lxr/source/?v=cvs
might help. You can try to back-track function calls with the links and
find out the possibilities of what some %s can become.

> Another form of the problem appears when %s gets (probably) replaced by
> only a handful of words ("it's either Engineers or Settlers"): if I could
> locate where all those words were when translating a phrase which includes
> them, I might actually be able to 'bend'(*) words. I'm after information
> like 'all these (few) words bend the same way, I can just add an 'n' in
> the end to make genetive'.  Eg. "%s report on the LARGEST Civilizations in
> the World" (here the names were easily located, too) - I changed the
> sentence to have the %s in its base form (come to think of it, I can't
> quite remember why exactly I did that, maybe habit of avoiding bending),
> and that required being able to translate all the names in non-genetive,
> too.
> 
> (*) This is a relatively big problem for a non-analythical language. The
> words like to change - if x is an object in a sentence, it looks a bit
> different than if it isn't, and lots of prepositions in English are
> replaced by adding endings. Eg. a fish: kala, I got a fish: sain kalan.
> Finland: Suomi, to Finland: Suomeen. As a result, for the general
> statements, I spend time avoiding having any objects in the sentence (you
> might have a vision on why this is a bit hard). ;)

I was thinking about translating Freeciv to Finnish myself, but I think
I'm more useful as a C programmer than as an English-Finnish-translator.
I'll study the code, maybe I'll be able to hack something useful. I had
the same experience with the translation - the cases in Finnish (and many
other languages) are tough to handle if the software isn't originally
written to cope with languages that use suffixes (Finnish) instead of
prepositions (English). Genetive is not the only case that causes trouble.

There are AFAIK basically two ways to solve this problem. You can make all
the sentences such that they only use the base cases (or whatever they are
called in English), which results in a bad style in languages like finnish.
Then there is the option to make the code have (for all nouns) different
strings for all the needed cases. At this point I don't know the source
very well. Could this be done without unreasonable effort?

Finnish translation would be very nice. Please make your translation
public even if it hasn't got any fancy case handling. It can be improved
later.

-- 
Tuomas Susi <tsusi@xxxxxxxxx>

A man is a success if he gets up in the morning and gets to bed at night,
and in between he does what he wants to do.
        -- Bob Dylan