Bjoern Voigt spake unto us the following wisdom:
> Yes of course we have many such problematic strings in Gaim. My samples=
> only show some of them.
The following is possibly a useful thing to consider; changing these
particular strings was not.
> The point is that our information flow is not optimal in i18n issues:
> 1. A single translator finds a problematic string in his .po file.
> 2. He tries to get an explanation in #gaim, Google or in this list.
> 3. Someone gives him an explanation and he translates the string.
> 4. Nobody changes the source code (no string changes, no comments).
This, it would seem, is where we could make the most effective
changes. When a string or phrase is unclear and requires
clarification, if changing the string is not the best solution (as it
is not for the strings you have brought up), a clarifying comment
should be added to the code which the next translator will see. In
several of your second set of examples, the strings which seem to have
given you trouble contain technical terms which should probably remain
(e.g., TOC and HMAC), but may need clarification for the translator.
Both of those strings are things that users either a) should not see
(a busted HMAC is a serious problem that shouldn't happen under normal
circumstances), or b) shoudl not mess with (users have no business
messing with their TOC server).
> 5. Weeks or month later another translator finds the same problematic
> 6. He also tries to get an explanation in #gaim, Google or in this
> list (may be he finds old mails in his archive about the
> problematic word).
> 7. Weeks later the next translator finds the problematic string.
> 8. He lets the string untranslated because he has no time to talk in
> #gaim etc.
> 9. ...
It would be interesting to know how often this actually happens. I
suspect that for the *most* part it's another hypothetical problem --
the questions I see are often about different strings, and often stem
=66rom a translator translating a portion of Gaim they do not themselves
use. This is a problem, I think, in the ideal world ... it would be
best if translators found the string in the program and translated it
intelligently based on context. I realize that this is not always
feasible, but lack of familiarity seems to lead to most of the
problems I have seen.
> For me it's the best solution to have an=20
> internationalization/localization manager in Gaim. His main tasks could b=
> * Helping translators with problematic strings.
> * Commenting problematic strings in the source.
> * Changing "bad" strings in the source. He always should have user's
> view, human interface guidelines etc. in mind.
As well as the flavor of our culture, and a healty sense of "this is
> * Dealing with other internationalization problems like date/time
> formats, right-to-left languages, far-east language specials etc.
> * Collecting translations from the translation tracker and commit them.
> Of course this manager should have write access to the repository. But=20
> there should be a verbal agreement with the developers so that the=20
> internationalization/localization manager should only concentrate on=20
> internationalization/localization issues.
> I suggested this several times more or less directly in discussions with=
> the developers. Developers always declined this.
I have never heard this. The only objection I would have to pinning a
particular person down to this is finding someone with the time to do
it right. If someone has the time, more power to them. Right now,
Luke handles a lot of this and seems (from my viewpoint) to handle it
well. If there is something specific that he (or we) could be doing
differently, please let us know.
Frankly, I think you, specifically, have made a few suggestions which
have not been accepted because community consensus (including that of
other translators) has been that the changes are not always for the
better; this experience may be coloring your opinion of the general
state of affairs. (I remember at least one set of plurality changes
which you wanted to make to make German easier, but which wouldn't fly
because of, e.g., Slavic languages having three plural forms, for
example.) In general, I feel that we are relatively responsive to
*actual* problems in strings. "This string is offensive when I
naively translate it to my native language without respect to the
flavor of the original language's nuances" is not a string problem
which we should be solving by changing code. When we find strings
that are, say, not marked as requiring a plural form because English
does not, but some other language does, the change is made immediately
without resistance. If this is not the case, then I apologize.
The laws that forbid the carrying of arms are laws [that have no remedy
for evils]. They disarm only those who are neither inclined nor
determined to commit crimes.
-- Cesare Beccaria, "On Crimes and Punishments", 1764