I've used AnkiApp extensively for the past year, as I've been learning Indonesian out on Bali.

I have come to the conclusion that shared decks don't work, at least not for me. The reason is that actually assembling the deck has turned out to be almost as valuable as practising with it.

I've got well over 1,000 cards in my Indo deck right now. I had to go find every translation and decide whether synonyms should be on the same card. That activity produces its own kind of learning.

I’ve heard this a lot, but my experience is actually very different using Anki to study medicine. While I agree creating cards definitely helps with memory, it’s extremely time consuming. If I need to memorize several chapters of material, I might give up if I had to create flash cards for all that content manually.

For medical students, there’s a fixed number of study resources that create a shared knowledge context for everyone using them. Anki decks are created from these resources and shared. In this case, shared decks are actually useful, high quality, and closely cover the material.

It’s much more efficient to use these premade decks - in the time it would take me to create one card I could have reviewed 5. If you have high quality decks, you can really optimize time spent on active recall by doing flashcards.

I’ve used Anki nearly every day for the past decade or so, and I agree it is not very efficient. It’s too passive. Like you said, assembling your own deck, deliberately going through the material in more natural chunks and actively processing it is more efficient.

But Anki makes up for that in consistency. I went through a deck of about 8000 cards and it was tremendously valuable. I could’ve learned the material faster with other methods, but only if I had the commitment to it. By contrast, opening an all every day on the train is a breeze.