[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: German translation in Fraktur
From:       Gerrit Sangel <z0idberg () gmx ! de>
Date:       2008-02-06 21:00:51
Message-ID: 200802062200.52163.z0idberg () gmx ! de
[Download RAW message or body]

Hello,

thanks for the answer!

Am Middeweken 06 Februor 2008 schrieb Federico Zenith:
>
> Ok, to sum it up your problems are:
> 1) Use mandatory ligatures (OpenType);
> 2) Convert to old German spelling;
> 3) Switch font for acronyms and all-caps text in general.
>
> In 1), you may want to substitute these ligatures diretly in the text with
> some regular expressions ("s/ch/whatever it is/g"), if they are included in
> Unicode. I could not find the ligature for "ch", though. Support of
> OpenType sounds more of an X11 issue to me.

That is a bit of a problem, because there are some ligatures in Unicode, 
e.g. "ff", but most of them are just for Latin. I already asked once at the 
Unicode ML, and they said that one should use Opentype ligatures for that. 
There will not be codepoints for other ligatures.

I am wondering a bit, because Firefox does support Opentype ligatures (at 
least Version 3). Do they use another rendering engine or so? I am not quite 
familiar with it.

> Anyway, while programming an application, I noticed that a particular font
> (I think it was FreeSerif on Kubuntu) would show an "st" ligature, without
> me ever knowing it existed; so, ligatures that the font file provides for
> are probably supported out of the box already.

I have tried to select the font manually, but no application (kate, Kopete or 
so) showed the ligature. The only ones who did it were XeTeX and Firefox 3.

> For 2) it would be very helpful to know whether it is possible to convert
> from new to old spelling with some rules ("All sequences of three
> consonants become two"), or whether it is on a case ­-by-case basis ("dass
> becomes "daß"). Anyway, if your only issue is with "ſs", you could just as
> well use the current spelling, substitute the appropriate "s" with "Å¿" and
> run a 'sed -e "s/ſs/ß/g" text.po' on the file: this would be way easier to
> do.

The ss->ß thing could be done automatically, I guess. I am not quite sure with 
the other rules of the spelling reform, but I think they mainly just cover 
capital spelling and if words are written combined or apart.

But the long s-issue would still need manual coverage. But this is not big 
problem, i tried this, and it would take about 2 hours for kdebase.

Also, all English words would need to be displayed in Antiqua. I don't know 
how far this could be automatically done, maybe via a dictionary or so. But 
even though I don't think this would be soo much work. 

> For 3), it looks like a problem that could be sent all the way upstream to
> the Qt libraries. The current HTML subset supported in rich text is this:
> http://doc.trolltech.com/4.3/richtext-html-subset.html
> I suppose you could use as a workaround "<span style="font-family:
> fantasy"></span>" or "<span style="font-family: cursive"></span>" to
> enclose text that cannot be rendered in Fraktur, and configure Antiqua for
> that kind of style. The text itself should be somewhat easy to find with a
> regexp, if the only cases are all-caps words. However, I do not think that
> every text on screen supports rich text. Not sure it can be made to either.

Ah, ok, thanks, I will take a look.

> You may want to look to the Serbian team for advice, since they already
> maintain two parallel versions of their translations (latin and cyrillic,
> not sure which is the manual one and which is the machine-generated). You
> might generate a de@Latf locale, pretty much the way there is a sr@Latn
> locale.

Hm, but I think the Serbian at Latin is a bit different, because there are 
different codepoints and they would not need to change the latin font in a 
sentence. But I take a look

> > So, what do you think of it? After all, I think Fraktur writing is really
> > a beautiful writing, although some people think of it as negatively
> > connotated, which is not the case.
>
> Some people associate Fraktur with Nazi Germany, but in fact the Nazi
> regime abolished the use of Fraktur, calling it "Jewish script", and Hitler
> himself had a personal dislike for it.
> http://en.wikipedia.org/wiki/Antiqua-Fraktur_dispute

Yes, indeed. It could only pose some problem, because a Fraktur font is rich 
in detail, so it could be hard to read on a computer monitor. But I guess 
there are some Fraktur fonts which could be read properly on the monitor.

Gerrit

[prev in list] [next in list] [prev in thread] [next in thread]