…. In fact none of the translations is updated because it is not maintainable.

Not true. I have kept the German translations up to date and besides a minor glitch in the UTF-8 version, they also worked. My point of view is that the most important bottle neck is to find a person willing to care about the translation. A technical solution, which lowers the barrier for contributions, would be a progress, but the barrier must be lowered considerably, maybe comparable to a wiki. See below.

> The Lazarus error messages are probably a different story. If the format is unimportant, why is that different please ?

The lazarus error messages are a different case in so far, as there are many more commits per months than in FreePascal. Therefore, the work load is much larger and the ease of plays a role. With the few commits in FreePascal, I have to work on it only every few months and then it is more a question of motivation to start than the actual work, which takes 15 minutes to 1 hour. With Lazarus the work might become 1 hour to 4 hours. In business terms it is fixed costs (motivation, actual start and commit) versus variable costs (doing the translations). My point was that the current status for FreePascal is not too bad and and I support this by keeping the German version up to date, but I do not want to claim this to be the general truth, because it may not apply for the Lazarus error messages. If you can contribute technical work, which lowers the barrier, I would welcome it. At the same time it would need some non-technical activities to actually attract more translators. They do not fall from heaven like rain drops. How about using http://translationproject.org/domain/index.html

Michael.

Hello Michael,

This is not what I meant. There is another passage in my post where I say "almost all languages." And by "not maintainable" I think like you not accessible enough, too disheartening to capture contributors.

Make that in a wiki, do not know if that's possible.

I am part of the French Translation Project Team and I would be very happy to see the POT files for FPC, FPDOC, Lazarus, etc., to this place. I can even handle it. Of course it takes POT files, which involves ressourestrings and ultimately a change in the way of producing the manual for this part.

I'm offered it a few years ago to put Lazarus POT files on Transifex. I received a negative reply. Will this be the same with TP?

I will not deny that I am a beginner in Pascal, but I do not want to charge someone as a work which he does not worry. So I will do my best to contribute a patch, though I know that it is possible that it will be accepted.

Regards,

On Sat, May 9, 2015 12:34, Schindler Karl-Michael wrote: .
.

Indeed - finding somebody who is willing to keep maintaining the translation in the long term is the most difficult part and I'm almost sure that it is not due to the message file format. Why I say this - at certain point in time, we had several translations up to date. If you know the last revision of the English version, it's really easy to focus just the differences (as mentioned above, they are not so many).

Moreover, we also had a tool allowing translation of the messages one by one and showing the English original at the same time (it was part of the former community forum site). None of the translators involved at that time was really using that tool and all of them preferred updating the text files instead.

Note that I do have experience with localisation of various programs personally (different programs using different techniques), so I can make comparisons. Obviously, different people have different preferences.

Tomas