Script writers: is it possible to extract your learning data for items -- number of times you got it wrong/right, review interval -- and save it somewhere?

You could then use this to make a list of items you’re having difficulty with, which would be nice.

At least on the web, yes, it’s possible, because that data is available to the client.

1 Like

Okay… Then also maybe it is also possible to extract all that data at once per course?

Any ideas on how to do that?

I’m thinking this does something very similar. It gives you a forecast of the number of items that are due for review in the future, for that it needs to know the reviewing intervals of your items :

Yes, it looks like that extension does something very similar. If you contact the author, s/he may be willing to adapt it for you.

In principle, it’s not a difficult task to do. The complication is that the Memrise API is a moving target, and not officially documented, so the script writer has no choice but to reverse engineer the web app, and go from there. Given Memrise’s total lack of communication about current and planned future changes, you are certain to experience unexpected failures and undesirable interactions when you install any user script for Memrise.

Thus, it’s a very thankless task, and not many people will want to release a script publicly, and have to deal with problems that the users experience, because Memrise suddenly changed something.

We just have to look at what happened with Eunoia’s excellent Uprise application, to see how even the most generous people with the best of intentions eventually give up in frustration in dealing with the reality of Memrise, and abandon their apps altogether.