Hi,
Not sure if this is interesting to anybody, but I thought I'd share how my methods for language learning have evolved over the years.
As you can image, I spent quite a lot of time doing flashcards, amassing more than 30k of them over more than a decade. However, when I started hitting 300 daily reps, I realised this was not sustainable anymore. So, that's why two years ago, in Mnemosyne 2.8, I added a feature to stop showing cards once they had a certain number of successive successful reviews.
That helped getting my workload under control, but after a while I started to realise that if you're doing a lot of flashcards, you're getting really good at... doing flashcards... I didn't really feel like the flashcards improved my actual language abilities a lot. This was even when using sentence cards, because I would often remember what a sentence meant simply by reading the first few words. So, reading the rest of the sentence had no more benefit.
Rather then using fixed sentences, I then started experimenting with having different sentences for a word each time. I initially thought of doing this inside Mnemosyne, but the interface was not a good fit for this, and grading became kind of meaningless anyway with this approach. So now I have a bunch of scripts which pull words from a list, using a finite and fixed sequence of intervals, and then collect sentences from the web (mostly from Reverso Context, but I even experimented with using ChatGPT for this). I originally generated an epub ebook from them, but now they are collected in a webpage, so that I still have the benefit of using browser-based dictionaries, sound files, shuffling word lists, etc.
Anyway, over the last year or so, I felt that spending 15 minutes doing this generated far greater dividends than doing flashcards for 15 minutes. In hindsight, this should have been obvious: if you want to get good at something, you should practice exactly that, and not something that's tangentially related to this...
Cheers,
Peter