Re: gnumeric: Note to translators

[Dennis Bjorklund <db zigo dhs org>]

On 12 Feb 2001, Karl Eichwalder wrote:

> > 1. The old word-by-word translation scheme.  The translations will
> >    never become good.  "Mean" could become "heisst", "Durchschnitt",
> >    "gemein", and probably a zillion other things -- you need context
> >    to decice.
>
> Put a comment at every occurence?  Yes, I know the will clutter the code
> with "crap".

I think is the better solution. Maybe also some new comment that can put
strings together in a group and put next to each other in the .po file.

There is a lot of context that is missing when one translates, and I don't
understand what's wrong with having the programmer annotate the strings
with this context.

I've translated one or two applications and it's almost never any comments
in the po fil helping the translator. And in many cases it does not work
to use emacs to jump to the correct place in the code to at least try to
get the context. I've been forced to search through sourcecode with grep
to find the strange strings and then to read the code to understand the
context.

Maybe one needs another macro that takes the string to translate and a
group string used to put things together.

_#("Mean", "Math")

_#("Tan", "Math")

But it's probably just as good and easier to just use the comments that
work today

/* Math */
_("Mean")

To just put strings together like in gnumeric is just making it harder to
translate.

As an programmer, every time you introduce a string you need to annotate
it with some context. Just simple things like annotating menu options with
the comment "menu option" would be good.

Other things I've had problems with is programs that use the same string
for menu labels as for toolbutton labels. If the english string is short
it works for them, but if the translation is long then the toolbutton
becomes big. If you translate to a abbrivation, then the toolbutton is
nice but the menu is ugly.

/Dennis