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

List:       kde-i18n-doc
Subject:    Re: Fuzzy Clock
From:       Nicolas Goutte <nicolasg () snafu ! de>
Date:       2004-09-26 11:47:32
Message-ID: 200409261347.32193.nicolasg () snafu ! de
[Download RAW message or body]

Well, I do not think that you have much choice about doing it otherwise.

However what I do not know is if you need also separated strings for morning 
and afternoon or if it does not matter. (Perhaps "noon" and "midnight" would 
be enough differentiation and the other hours would not need to be 
differentiated.)

Have a nice day!

On Saturday 25 September 2004 23:55, Frerich Raabe wrote:
> Moin,
>
> I hope this is the right place for reaching the people who translated
> the KDE GUI into different languages. If not, please forgive me. :-)
>
> I've recently gotten a few complaints about how the fuzzy clock (an
> operation mode of the clock applet) is virtually impossible to translate
> properly into many languages. In particular, I noticed a few bug reports
> about this:
>
> http://lists.debian.org/debian-qa-packages/2002/08/msg00010.html
> http://bugs.kde.org/show_bug.cgi?id=87287
> http://bugs.kde.org/show_bug.cgi?id=79410
>
> This is just german, french and norwegian: I bet there are even worse
> problems in other languages like russian. In fact, the fuzzy clock
> code already contains a hack for spanish, so that it appears properly
> for spanish users.
>
> The root cause of all this is the way how the messages which the fuzzy
> clock displays are generated. Instead of having hardcoded strings like
> "Ten past one" or "Quarter to seven", the fuzzy clock assembles the
> strings out of e.g. "Ten", "past", "One".
>
> The advantage is tht one needs fewer strings for this: one list of strings
> for the numbers from 1 to 12, another list of strings for terms like
> "quarter to", "half past" etc..
> The disadvantage is that this does not work.
>
> Hence, for the sake of correctness, I'm tempted to change the fuzzy clock
> code to just use one big list of hardcoded strings. I suppose this should
> work a lot better, since translators get more context to look at.
> Unfortunately, this will increase the size of the applet (which is really
> no problem at all) and it will increase the number of messages the
> translators have to translate. Also, it will be quite boring to translaate,
> because the messages will be like
>
>
> "Five past one", "Ten past one", "Quarter past one"..."Ten to one", "Five
> to one", "Five past two", "Ten past two", "Quarter past two"..."Ten to
> two", "Five to two"
>
> You get the idea.
>
> What I'd like to know now is, whether this change would be okay with you
> people, or whether you know better ways to fix this problem.
>
> - Frerich
>
> P.S.: Please CC me on replies to this message since I'm currently not
> subscribed to this mailinglist. Thank you!

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