I wanted to like WaniKani. I spent nearly a year trying to use it but my memory appears to be incompatible with their way of doing spaced repetition, which is e

I wanted to like WaniKani. I spent nearly a year trying to use it but my memory appears to be incompatible with their way of doing spaced repetition, which is extremely simplistic. There's no attempt to set intervals based on the user's settings, the user's previous performance or the vast accumulation of data they must have regarding which ones tend to be easier or harder, let alone some sort of ML approach. Anki handles the first two of those fine, WK should be in a position to push that aspect further but they haven't bothered. The overall feeling is actually somewhat adversarial - as if it's a challenge, rather than a way to help. I ended up spending 2+ hours a day just doing reviews to stay in one place - not even attempting to study any new material. I even wrote a couple of tools to let me review recent failures, but in the end it was clear it just wasn't working. The subscription pricing approach is also wrong, in my opinion. Yes it's technically a web service but there are very rarely any updates and although they heavily advertise the potential to learn all the kanji in a couple of years, in reality it'll take a lot longer. So the spend is going to be $300+ even if it works for you, and the incentive to get through it quickly to save $$ is misaligned with a better goal of learning in a thorough and sustainable way. Anki is better in most ways, really.