Bonjour,
After discussions with the French localizationlab coordinator, I proposed a feature request to weblate to have the ability to lock a single string.
It sounds like an effective way to converge to a definitive translation for a given language.
Thoughts ?
There is a bounty open to implement this feature in weblate. Currently funded at 100 USD but I’m pretty sure that’s not enough and additional funds would be welcome !
@dachary I will definitely help with the bounty and support the feature add, at least as an option for projects to enable. I can see where there might be some resistance on the side of Weblate and some Weblate users however. I have talked to translator-users (primarily from a strong free software background) who fundamentally disagree with the idea of locking strings because they believe that exactly this inhibits quality control by limiting contributions. I’m not sure exactly where Weblate stands on the issue ideologically.
This is great news ! Thank you 
I understand the rationale and in this case another review process would be better. Some weblate users already have the right to lock an entire translation for a given language, via the weblate interface. The proposal does not introduce a new level of control that did not exist before, therefore I don’t think it changes anything philosophically. It proposes to make it finer grain (i.e. apply to one string instead of a whole translation). And if some people using weblate do not like the concept of locking altogether, this can be left alone and not used at all. It will not interfere with the current workflow in any way.
Does that answer your question or am I missing something ?
A short note about this feature about to be released in weblate 2.18. It can be tested in the weblate demo site when login as review / review (to see and try the Approve radio button) which is only available in Czech or demo / demo to be a regular translator with no reviewer privileges.
Unless there is a major bug there, it is going to be released as is. If something is inconvenient it is however worth filing an issue referencing the issue tracking the original request.
Thanks a thousand times