Complete.Org: Mailing Lists: Archives: freeciv-dev: September 2000:
[Freeciv-Dev] I can do the Swedish part (was: Re: Re: Compound City Nam
Home

[Freeciv-Dev] I can do the Swedish part (was: Re: Re: Compound City Nam

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
Cc: Freeciv developers <freeciv-dev@xxxxxxxxxxx>
Subject: [Freeciv-Dev] I can do the Swedish part (was: Re: Re: Compound City Names)
From: Erik Sigra <freeciv@xxxxxxx>
Date: Tue, 26 Sep 2000 10:48:52 +0200

Hi,

Reinier Post wrote:
> 
> [xconq German city names algorithm]
> 
> > This is the most elaborate solution I have ever seen. (I am not at all
> > familiar with the syntax.) It looks very promising.
> 
> It is elaborate, but not generic: it is built around some ad hoc
> knowledge of German city names.

You probably mean, what Stan posted here yesterday is not generic.
That's true. But that was the German file. However, the parsing
algorithm is generic, if I got it right.

> I prefer Falk Hueffner's algorithm, because it is completely generic: it
> assumes no knowledge of the language.  As a practical advantage: it's
> shorter, and it doesn't need language specific tuning for each new nation.

I don't think the need for knowledge is a problem - because there is
plenty of knowledge availible. For example, each translation requires
good knowledge about the target language. Still there are many good
translations for Freeciv. <irony>You wouldn't prefer a general
translation algorithm that requires no knowledge of the target language
either, would you? ;-) </irony>

I can do the language specific tuning for Swedish. I could also do it
for a bunch of other languages if neccessary (it requires far less
language knowledge than a translation), but I'm sure it won't be too
hard to  find native speakers of thoose languages to do the job.

> But I also like the names it generates better: a much wider range of
> plausible names is generated, some of which are really funny.  This goes at
> the expense of generating some implausible names, but I'd prefer that to a
> boring naming scheme.

I haven't yet seen any generated results from Stan's algorithm, but I
don't think it would be too boring for me at least.

I would also like to see Falk's algorithm included. But not neccessarly
as the only algorithm.

> --
> Reinier Post

Erik



[Prev in Thread] Current Thread [Next in Thread]