Page MenuHomePhabricator

Explore usefulness of "hardest" translations data
Closed, ResolvedPublic

Description

Migrated from: https://wikimedia.mingle.thoughtworks.com/projects/language_engineering/cards/2895

Since TUX, we're keeping track of what messages are skipped the most. However, little if anything has been done with such data. Data should be inspected for meaningfulness and possibly used for something.

Event Timeline

Amire80 raised the priority of this task from to Low.
Amire80 subscribed.
Nemo_bis renamed this task from Add filter for the "hardest" translations to Explore usefulness of "hardest" translations data.Feb 22 2015, 6:10 PM
Nemo_bis updated the task description. (Show Details)
Nemo_bis set Security to None.

I am afraid that the data we have collected is not useful. We have 33615 hard tags in translatewiki.net corresponding to 33615 different messages. For comparison, translatewiki.net only has 58937 messages in total.

I propose this feature to be removed completely for now. It can be done again properly later if there is time.

That was fast! I don't remember the backend, is there no (easy) way to make the criteria stricter like a threshold somewhere?

I guess the idea was to keep counts so that we could determine and filter by threshold, but the way it is currently implemented is that it stores only one vote per page.