In order not to swamp the maintainers, is there any way of clipping my staged and stashed language strings into packages of no bigger than 30 strings at a time?
I'm currently working on translating H5P to Dutch (still 100 strings remaining) and would like to do things properly.
I don't think there's an easy automatic way. In theory, you should be able to load all the translation into the stage, and unstage all strings one-by-one but those 30. At which state, you would submit them to the maintainers.
However, I don't think it is necessary. How many translated strings are there? If I was that maintainer, I would personally prefer if all H5P module strings were submitted at once. So that I can see and check things in terms of consistency etc. Definitely would work better for me than multiple smaller chunks.
Hi Joost,
I didn't realise it was a problem for you. Just throw them in
The thing with the 30 strings is that when you start reviewing a contribution as a maintainer, it is nice to be able to do the job in one go. It's impossible to estimate how long the review is going to take. That depends on the number of inconsistencies, spelling / capitals / punctuation-mistakes...etc.
Most of the language pack maintainers are volunteers, often squeezing translating/reviewing in between other jobs, while waiting for some process to finish (speaking for myself here ). A 30 string contribution will probably not take longer then 10 minutes of work to review, even if there is some rework to do or if strings are very long. A 100 string contribution can take 30 minutes, so I don't start on it unless I'm sure I have that time without interruptions en no other work that can be done in 30 minutes.
So short contributions will be processed in hours or days, longer contributions can take a lot longer.
Not so much a problem, but in my experience as a translator, I normally deliver full translations without chopping them up into smaller bits. I do understand that this is a challenge for the QA part of translations, but for the translation work itself, it's better to finish a unit instead of chopping it up into smaller pieces. However, I could imagine there will be a feature in AMOS that processes large translations into smaller batches of 30-50 strings at a time.
