p-christ

Do you agree with the assumptions of the Anki scheduler? That each time you review a card you'll remember it for incrementally longer and longer?

If so then a fixed amount of review time per day will let you remember things for longer & longer.

thanks a lot, your comments make a lot of sense and are really helpful!

Do you agree with just this part?

> Effectiveness of spaced repetition scales exponentially

If you've forgotten something then you don't have that knowledge anymore. So remembering things for longer means you are effectively gaining knowledge.

It's like someone giving you $10 and then trying to argue that they haven't actually given you $10 because just a few minutes ago you had a $10 bill in your pocket but you spent it on something else.

> I don’t need to remember the exact way something was written by the author.

I agree. Who ever said you should try to remember that? Remember the high-level important and transferable information, don't waste time trying to remember information that won't help you elsewhere.

The simplifying assumption around units of information and the graphs are just to visualise the main point.

Effectiveness of spaced repetition scales really fast / exponentially whereas other learning methods don't scale like that - do you agree?

If so then over the long-run spaced repetition is always going to be extremely efficient relative to other learning methods.

The article links to a paper which estimates how much more effective you get at reading the more knowledge you have. It shows that for most ages you don't improve in this way, but for older ages you do.

Then it says for the "best case" that "adults" takes years to get twice as effective at reading.

If you don't believe this then it means you think it takes adults less than a year to get twice as effective? If effectiveness doubles each year then that would mean by the time a 25 year old becomes 65 that they're 1 million times as effective at reading than when they were 25. Does that sound reasonable to you?

Save All does that. You can create a deck and then share it so a whole team can use it

https://saveall.ai/

You misunderstand how important memory is for learning. Ofsted, the UK school inspection board, actually defines learning itself as a "change in long-term memory". You can't learn anything without a change to your memory, they aren't separate things.

So, sure, you should try to apply your knowledge and layer concepts on top of each other. But if you do that AND also remember a lot more of such experiences you'll learn a lot faster.

Thanks for the feedback, we are going to add an export feature sometime soon

Think about it in terms of how many days of knowledge each rep gets you.

The first rep gets you 1 extra day of knowledge. The next rep gets you 2 extra days. The next one gets you 5 extra days.

So each rep is getting you more and more knowledge even though the time required to execute the rep remains the same. This is exponential growth.

If you only have 1 card then for most days you won't have any reviews to do so it doesn't work. But if you add cards regularly then over the long-run every day you'll be getting more and more "extra days of knowledge" by doing the same amount of reviews each day.

Summary of article:

The effectiveness of spaced repetition scales exponentially and much faster than other learning methods. So use spaced repetition and you’ll learn a lot faster in the long-run.

I use it for both types yeah. I basically try not to ever forget anything I hear that’s useful.

On Save All you can create cards that are just statements, no need to turn them into Q&A. So if I hear an interesting fact I usually just dump it in quickly verbatim.

Haha, where would you rank this one in terms of quality?

Agree, I shouldn't have used the word "proved". But there is a lot of robust evidence for it going back to the 1800s. Many of the studies are linked to on the wikipedia page

https://en.wikipedia.org/wiki/Spacing_effect

Yeah. There is software that organises the reminders for you aswell e.g. my company Save All does this https://saveall.ai/

Very interesting, how would i learn about something like that? Is there a resource you'd recommend?

thanks will check that out

what's that?

thanks lol

This seems like what a spaced repetition app does, is there a difference?

For example lots of people use Save All for this exact reason https://saveall.ai/

> It only takes days to get twice as effective at reading with spaced repetition? Or at learning? If it's the latter, I don't think that's what the earlier study measured?

It only takes days for spaced repetition to get twice as effective, not reading. It takes you years to get twice as effective at reading.

> I think about other ways to build your understanding, like working through problems and applying the knowledge, that are key to learning. Much of learning physics is getting your hands dirty in the equations, and there's a big difference between knowing a formula and really understanding it in action.

I agree, you should definitely work through problems and apply your knowledge. The post is arguing that you should do that AND spend a little bit of time doing spaced repetition, not to only do spaced repetition and nothing else.

How do you get videos to 3-4x speed? I find most video players only allow up to 2x speed?

What bit do you doubt?

The core concept that spaced repetition increases rapidly in effectiveness over time is called the Spacing Effect. There are many many studies that have investigated and proved it

For sure it's also good to practice things aswell, it doesn't have to be only one or the other. My understanding and experience is that if you do 95% practice & 5% spaced repetition you will be significantly better over the long-term than doing 100% practice.

Sorry, you're right. I was maybe too focused on being efficient rather than polite. I'll edit my comment

I understand what you mean but I think you underestimate how critical memory is for all forms of learning (even creative work/learning).

Our working memories have a capacity of 4. This means that we basically can't understand something if it requires more than 4 pieces of New knowledge to understand. To understand more complicated things we need to move some of the knowledge into our long-term memories.

We wrote an article on this topic here that i'd love to know what you think of

https://saveall.ai/blog/learning-is-remembering

So there is a theroetical point at which your brain will get "full" but it would take so much knowledge to do that that it's basically impossible.

Yep! Our long-term memories can store almost the entire internet as it was in 2016 inside it!

See here for a reference for that: https://www.livescience.com/53751-brain-could-store-internet...

Or this article we wrote also talks about it: https://saveall.ai/blog/learning-is-remembering

(1) We are going to incentivise high quality decks by allowing people to sell access to their decks

(2) The main problem we see with Anki public decks is that the user themselves decides whether they got the question "right" or not when reviewing the cards. This lack of a "teacher" means that it is very very difficult to learn using someone else's cards.

You basically end up kind of getting something wrong but then saying it was right anyway. Do that a few times and your trust & investment in the process goes and you'll eventually you lose motivation to carry on with the deck.

Save All decks are different. WE decide whether you got it right or not, not you. This makes it much easier for you to learn using someone else's decks