to modify the spaced repetition algorithm
I feel that the algorithm is too optimistic compared to my learning capacities. There's a flash card i reviewed recently : Got it wrong the first try because i didn't remember very well, than got it right the second try... Now the next review will be in 50 days !!! Almost two months !!! So my suggestion for the algorithm simply is to "punish" harder when you're wrong. Cut the next review delay in half for example when you failed putting the correct answer once. That's a nice tool anyway. Good job !
-
Steven commented
Being able to tweak the algorithm is important, because I have the opposite problem. I want the cards to be more lax to free me up to read more. SuperMemo has a concept called the forgetting index, where you can modify how likely you will forget an item when it comes up for repetition. The default is 5%. That means that it spaces the testing out far enough that there is a 5% chance that you won't remember the answer when it comes up. I prefer to have my forgetting index to 15% which for me finds a healthy balance of reviewing previously known material and getting new input.
Also, it would be nice to know more about the algorithm to inform my decision on whether to keep more words in Readlang or move them into SuperMemo.