From checking logs, it doesn’t look like many people have actually tried supermemo to any significant degree/talked to high level users.
I’m not an expert but I’ve used SuperMemo for ~2 years (1.5 of those years with incremental reading) and I’d be happy to teach/answer questions from anyone interested in it.
I’ll briefly summarize incremental reading in words but I recommend watching this video I made (SuperMemo for Wizards) instead to get a better idea of how it actually works: https://www.youtube.com/watch?v=NURjZuHvDaU&t=672s
tl;dr: spaced repetition is systematized memorization. It doesn’t do anything about learning though (card making process also sucks). Incremental reading is systematized learning. You can throw all the material you want at SuperMemo and it’ll manage review/prioritization for you (I have somewhere between ~1,000-3,000 articles imported and I can meaningfully do reps every day and learn and memorize the parts I care about)
I never studied much for school (in large part due to no one ever teaching me how to + adhd brain not liking to sit down and do things). SuperMemo made a huge difference in making it really easy for me to learn whatever I want since I can completely offload any concerns of when or what to read with its priority system.
I highly recommend anyone interested in long-term learning give it a try. You can get a trial of SuperMemo 18 (sm15 is free but I don’t recommend it) at supermemo.wiki/learn. I’ll also gladly teach anyone interested in learning incremental reading Supermemo (as long as you promise to try it at least 3 days before session, at basic level/go through built in tutorial).
[for anyone interested in my motivations, it’s entirely altruistic. Took me like 5 months to even start incremental reading. Of the probably 15-30 people I’ve taught, the majority within a week or two get proficient enough that they definitely enjoy it. Takes me little time for significant results]
Most people here use “Anki” instead of SuperMemo! I think,Anki has almost all the important features that SuperMemo has including Incremental Reading addon! Besides,Anki is free for Android…And it is Open Source!
Something to also note is that SuperMemo is not some innovative new option. Anki is actually based on SuperMemo. All incremental reading is based in some way on Piotr Wozniak’s (creator of SuperMemo) writings. It has 30 years of work behind it that people mainly don’t take advantage of because it’s quite hard to start with. Other than that, I would say for high level users it definitely clicks well.
Definitely true on the free + open source part. Alongside it only working on Windows. If those are things you care about a lot SuperMemo is not for you.
I do disagree though about the incremental reading add on: SuperMemo is significantly better at it (it does have like 20 years of work on IR vs. much less on the anki plugin). If you ask on the anki discord server even, anyone that’s ever tried SuperMemo long enough to actually understand it would agree (even though there might be other constraints such as windows only preventing them from using it regularly).
In my opinion (I am fairly confident in this since I know lots of people that switched from anki → supermemo but very few in the opposite direction. If you want to confirm just ask on any of the SuperMemo discord servers or even on the anki ones) SuperMemo is at least say 1.2 (very very conservatively) times as good as anki, long-term, if you are proficient. IR on SuperMemo is just a completely different paradigm.
For some reason I can’t link posts but I’d strongly recommend looking up an article from Master How to Learn title “Why I Switched to SuperMemo After Using Anki for 5 Years, With Over 50k Cards and 420k Total Reviews”.
I think he can explain the value of SuperMemo much better than I can since he’s used both SuperMemo and Anki thoroughly.
I’m not an Anki fan,btw!!! I personally don’t like Anki,especially because its cards are not very editable. And it has several other issues that I could not overcome. I have been using Anki since 2015! And I am slowly switching to other apps,apps that are more flexible!
Before Anki, I used SuperMemo for a couple of months. And I hated it because of the way its interface worked. It was needlessly complicated! And the creator of SuperMemo seemed to have wanted the user to do things the way he wants! His way is the ‘only’ way!
I prefer open source tools like Anki to the closed one like SuperMemo,simply because I have full control over the Database and internal structures of the opensource tool. I aim to store Data for a very very long time(15+ years or lifetime). So,I want to make sure that my Data is in my hand and I can move the data anytime I want,in any format I want. Anki’s SQLITE DB lets me do that! Tools like “Evernote,OneNote,SuperMemory” won’t let you convert your data the way you want. This is a serious issue,a hassle in the long run! I take “Data portability” very seriously now a days!
But like I said,I am slowly leaving Anki…I don’t think,Anki or SuperMemo’s algorithm is as scientifically valid as the users assume…They are good for storing knowledge,but there are other ‘better’ tools for both storing ‘knowledge based stuffs’ and setup a ‘spaced repetition systems’.
re: interface, completely agree. Drove me crazy the first few months when nothing made sense. I strongly encourage you to to give it another try though, if I teach you the basics you’ll have a much better, less confusing time of it. In 1-2 weeks I think you’d be able to decide again for yourself if it’s worth it or not. If someone had taught me at the start would’ve saved me probably half a year too.
For me personally, I don’t think I could ever use anki long-term. Card loads are too high (get more into that later) and the lack of learning alongside memorizing makes it horribly boring. Agree on the editing too, I think it’s pretty important to do (I edit my cards a lot while doing reps).
re: database control: you can export SuperMemo’s db whenever you want. The program is proprietary but you can do whatever you want with the files. There are even scripts out there for converting from SM to Anki. You just need to go to file: export: xml.
re: storing knowledge: could you elaborate on what you mean by knowledge based stuff? Do you mean material that you’d want to learn or notes? I think SM does have exactly this with ‘topics’. I have lots of notes in SM. Other apps might have other benefits but being able to have material/notes scheduled to be reviewed is quite important I think (otherwise they just end up sitting there, never used again).
re: algorithm: is there something that makes you think they don’t work? It doesn’t seem very hard to test, you’d just have to do random repetitions and see if you can actually remember as much as it predicts you can.
Thanks for responding by the way. Appreciate the discussion.
With great respect to the creator,Piotr Woźniak, of the algorithm of SuperMemo,which Anki copied,the algorithm has not been actually scientifically proven to be very effective(yet)!
I have great admiration to Piotr Woźniak. His theory/algorithm has dramatically changed my life. But his algorithm is based partly on his subjective experience. And here lies the main problem with the algorithm of Anki,SuperMemory,etc…They are partly subjective!
Spaced Repetition surely is a scientifically proven method to learn something. But science still could not discover the perfect scheduling for SRS that will universally and effectively work for everyone in the same way. This is because we don’t learn something at the same pace,speed,etc. What may take one hour for me to learn may take ten minutes for you to learn. What may need five spaced repetition for me in one week to store the data in my long term memory may need only two repetition for you in one week…
So,the algorithm is subjective to a great extent…There is no ‘optimized’ scheduling in all these tools…Anki,Memrise,SuperMemo,etc!
But like I said before,I admire Piotr Woźniak. His tool,SuperMemo, should have gotten more exposure and recognition than Anki. Not many people know his contribution to the world of SRS…
There actually is a (very old) academic paper from Woz title “Optimization of Learning” (I unfortunately can’t link).
I think if you take a look at the SuperMemo algorithm you’ll find exactly what you mentioned. Later algorithms are designed to adjust to people’s memory. Scientific approval/disapproval (in my opinion) doesn’t matter when you can practically test out the results for yourself with random subset review (or just vast amounts of anecdotal experience).
I think it’s a bit too easy to just dismiss the last 30 years of his work just because there’s no modern scientific paper on it. I’d say: let practical usage be the judge.
As an aside, I’m curious what you’d consider a more effective means of doing large scale reading and retaining that reading over long time frames. If you were to read, say a book a week, how would you want to learn and retain that?
That will not prove anything,as any type of ‘review scheduling’ would work well enough for SRS! For example,any variation of “leitner system” can be used for reviewing! You can use one week gap,one month gap,six months gap depending on your circumstances or even choice and subject matters! And it will work,too!
My point is,it looks like there is actually no scientifically proven optimum scheduling system out there for SRS(yet)! So,everything ‘works’ if you use them regularly!
