The original idea was that it'd be available for all users, and that's why we used localstorage to track its state. We could start by enabling it for logged in users by default, as that's maybe a quicker (config) change, but I think enabling for everyone would be good. I'm not sure what the condition of cxserver.wikimedia.org is, but I presume it's more stable than it was last year when we first turned it on.

Thanks for the background! I agree enabling for all would be ideal.

From T386889 it sounds like work was done to improve stability of CXServer. Still, I guess let's ping @KartikMistry @Nikerabbit just to be sure?

Is there a way to do this gradually and closely monitor the load with @KartikMistry? We also have T386371: Request capacity increase in preparation for MinT for wiki Readers experiment and want to avoid causing problems for the experiment due to capacity issues.

Glad I pinged! Yes, we can do it more gradually.

Knowing there are capacity issues, I want to first improve performance on our end. We could do better on deciding what should be sent to CXServer. For example the "Vote count" column in the wish-index table doesn't need to translated. More specifically, I was thinking we should restrict use of MinT to user-submitted content only?

Looks like some capacity problems are still being worked on based on T411058: Can't deploy machinetranslation due to exceeding resource quotas. I guess we should wait for that to be resolved, then we can re-ping LPL to see if it's a good time to switch on machine translation by default.

Knowing there are capacity issues, I want to first improve performance on our end.

Noting that I think we've now got our implementation as efficient as it can be (r1204089). Any given wish typically makes up to 4 translation requests at most.

In T408160#11676356, @MusikAnimal wrote:

Looks like some capacity problems are still being worked on based on T411058: Can't deploy machinetranslation due to exceeding resource quotas. I guess we should wait for that to be resolved, then we can re-ping LPL to see if it's a good time to switch on machine translation by default.

Knowing there are capacity issues, I want to first improve performance on our end.

Noting that I think we've now got our implementation as efficient as it can be (r1204089). Any given wish typically makes up to 4 translation requests at most.

Yes. Things didn't go very well after we upgraded the capacity for MinT. I'll ping once we have any updates.