Next Project: Anki Replacement

I love Anki. I know a lot of people hate it. And I know how easy it would be to hate it. It has some "flaws" which make it that way. I also think that they are pretty easy to solve.

Just wrapped up my last project which was to work on my first version of an API Gateway project and integrate with my brother's application. It was the first step for this project actually. I hate writing and rewriting basic functionality which could and should be handled by a API Gateway. Such as auth and routing.

But, enough of that. The next step is to push my Gateway project by expanding what it needs to handle. And for that it is time to work on the first project which might actually go live at some point. Which as you might have guessed is my Anki replacement.

Back to those flaws:

1. Anki-lanche
2. Fixed Calculation

The first is why most people end up hating Anki. When you first start you want to start out with a small daily influx of new cards. But, this pace often ends up being boring so most people jack it up higher. But, then when past cards come up again for review you start getting buried. Eventually you get into a groove. And then... you miss a day or two. I've been pretty good, and I rarely miss a day. And my decks are mostly mature now. But that took years. So now an "Anki-lanche" for me is smaller than my daily reviews were back in the midst of it. For most people though, they fell off the bandwagon in the middle and rage quit.

And the problem is partly due to that fixed calculation. When you answer a card it gets assigned a due date based on a fixed calculation. And when that due date comes you get that card in your review list + your daily review amount.

And all of this runs counter to what you need to do to be successful with Anki: regularity.

We may be able to dedicate a certain amount of time. But, the amount of time in Anki isn't the variable you control. You control the rate of new cards. And as you answer cards and run back into old reviews you initially start with a workload that gradually grows and then gradually decreases again. But, it can spike MASSIVELY if you miss a day (or do horribly bad).

It would be much easier to simply set a daily target in terms of either time or number of reviews.

Also, the fixed calculation allows your brain to cheat a little. You can start noticing patterns in what reviews show up together. For instance, a lot of my Japanese decks had a mix of words and phrases. And they were often introduced together. So, if I got one right, when the other came up, I had technically already reviewed the target word.

The rigidity of how it calculates makes it easy and boring at times. There needs to be some amount of randomness involved as well.

So, that is my plan. Write an Anki-like application which handles calculations quite a bit differently. We'll see how it goes.