They barely mention why they're even doing this. I see flashcard as a means to something (e.g. learning a language, preparing for a test...). The measured outcome should be the success in task, rather than the number of reviewed cards.
Reminds me of when I was looking for good workflows for note taking tools like Obsidian that would be relevant to me, but so many of the big articles on how to do various types of note organizations strategies all used theirs to organize notes on videos and conferences they had watched or gone to about note taking.
Reading the article I did not ask at all the question for why they're doing this.<p>I see no reason why that is important.
That's what the business model of Doulingo is based on
Yeah, can anyone else that does this explain what the point is? Genuinely curious.
1. I have always loved learning things, both big and small.
2. I enjoy trivia competitions.
3. I'm interested in human memory more generally.
4. I think that spaced repetition software could be a lot better, and I'm trying to make such software. So when I study, I'm also getting the value of using and improving my own software.
If you use Anki and want to analyze your review history, you can export the review history into a CSV, and then use pandas to analyze it.<p><a href="https://www.encona.com/posts/custom-statistics-for-anki-flashcard-reviews" rel="nofollow">https://www.encona.com/posts/custom-statistics-for-anki-flas...</a>
As someone who did about 51k in 2025, oh my goodness how would you possibly have time for 300k?
1. I have a long commute, and I have about 45 minutes of walking per day, during which I like to do my reviews.
2. I've put serious effort into reducing the friction in my software in order to reduce time per review.
3. You've got to do <i>something</i> when you're on the toilet, right?
I did 10k reviews. Admittedly, very streakily - so spurts intense study for 2-3 months than several months off - I think about 2.5 spurts total. I felt pretty wrung out after. 51k is a ton, congrats!
Interesting to see the stats here. My total active library size is about the same as the author's (~50k cards), yet I performed less than 100k reviews this past year. That said, my overall retention is a good bit lower (~83%). Wouldn't have expected a 6% difference to make for a 3x higher review load!
What are your cards about? The author seems to be learning for the sake of learning.
1. My algorithm is probably inefficient, and a big Q1 2026 goal is to figure out where the inefficiencies are and (better) to get a better system for addressing and remediating them in an automated way.<p>2. A lot of my cards were also <i>made</i> in 2025 (and 2024), so I'm probably much farther to the left of you on the learning curve, on average.
I would have liked to see statistics on study session length eg. average duration. Also how long you typically spend creating per card and how many you created this year.<p>That information would help us all better assess whether the time spent on a spaced repetition flashcard system is justified
From the article:<p><i>> For example, I only record the correctness of a response, not its subjective difficulty, and I mix in random cards with my study sessions to make it harder for me to guess the answer on the basis of when I'm seeing the card.</i><p>Sounds a bit like the Leitner system [1] with respect to recording only Correct/Incorrect responses. One of the reasons I avoided Anki for a <i>long time</i> was that I wanted to be able to answer cards quickly without actually looking at my phone. I ended up using a combination of automatic TTS, bluetooth headphones, and swipe-up/down gestures to indicate my response.<p>Made it much easier to go through cards while driving or during daily runs with my husky.<p>[1] - <a href="https://subjectguides.york.ac.uk/study-revision/leitner-system" rel="nofollow">https://subjectguides.york.ac.uk/study-revision/leitner-syst...</a>
If you have the software for that in a repo somewhere, I'd be interested in seeing how you made that work as well.
Would love to know more -- what software do you use?
I would love to hear more about your setup!
I sympathise. I did ~200k reviews in Anki and no idea how many in Renshuu in 2025 for language learning: German, Spanish and Japanese... and I think I got into Anki hell.
For a long time, Anki was really useful for me, it pushed my Spanish and German forward, but now I plan to decrease the number of reviews significantly. I hope to spend no more than 30 minutes per day of flashcards, and the rest of time on immersion.
There are some days when I feel I don't know this community at all. :-)<p>Do this many people use flashcards? Maybe I'm way too old. Probably.
First of all, big kudos for not missing a single day. When I used flashcards in the past, missing even a couple of days led to an avalanche of cards to review.<p>Since you’ve been so consistent and are using your own software, have you experimented with different resurfacing rates? Did you notice a material difference in recall?
Thanks to everyone for all the useful notes and questions here. I've compiled a follow-up post here:<p><a href="https://www.natemeyvis.com/22-reasons-i-did-301432-flashcard-reviews-in-2025/" rel="nofollow">https://www.natemeyvis.com/22-reasons-i-did-301432-flashcard...</a>
How do you decide what gets a card? I see you like trivia and such, but what triggers you to turn a trivia sounding fact into a flash card?
At this point there's just a reflex I have that says "ah, I'd like to remember that." It's the same feeling whether I'm learning something for trivia, for work, or for personal reasons.
301k reviews/year is serious commitment. Curious — do you ever prune low-value cards, or is the goal to never delete anything?
I prune some, but less than I probably should, and less than most other serious SR people do. I'm more interested in techniques for (i) raising the quality of even my lower-value cards and (ii) figuring out how to actually learn the stuff that people think of as "leeches."
Does doing this have utility? What problem does it solve?<p>Years ago, I memorized 1034 digits of pi just to see what it felt like (reciting pi from memory felt like walking through a forest at night without bumping into any trees). So, there was some value in that experience.<p>I wonder what this guy gets out of it?
His homepage says that he likes memorizing things. There is utility in doing something you enjoy for the sake of it.
Random example: In my last job I would point out JS language features that would have made my coworker's code more concise and canonical.<p>I hadn't done any JS coding beyond a few examples from a tutorial.<p>There's no way I would have retained that knowledge otherwise.<p>There are <i>many</i> such examples. In general it's extremely useful for retaining things you are <i>not</i> going to develop muscle memory for.<p>"Memorizing by doing" is great if you're doing often. What if you're not?
1. I enjoy it.<p>2. I like trivia competitions.<p>3. I like making and using my own software.<p>4. Memorizing facts is an underrated way to become a better software engineer. Not the best way or even close to the best way, but an underrated way!<p>5. It enriches my experience of the world (I plan to write more about this soon).
I think they just enjoy memorizing things. Roughly the equivalent of meeting someone who runs 10 miles a day. They enjoy it and it has some benefit to their life as well, even though they are probably far past the point of diminishing returns.
Very cool, and have you actually used forest as memory palace ? this and chunking + imagery mapped to digits, is what got suggested to me just now by chatbot as technique that memory athletes use; got curious myself but never tried something like this.
> The prompt of my most-missed card (39 misses in 2025) is: Merrily We Roll Along (the musical) is based on a 1934 play by what two people<p>> [...] But, ChatGPT and Gemini both tell me that these effects are not statistically significant, despite having pretty large sample sizes.<p>Imagine he instead learned how to calculate staristical significance instead so he didn't have to believe AI guesswork.
I read the author's attempt to explain why memorization is important, and found myself unconvinced. Of all the things we consider to be "intelligence", memorization of facts seems like one of the least valuable in the Internet-era. That said, I am open to hear some counter-arguments (pro-memorization).<p>Of course, if you simply enjoy the process of memorizing facts, then no explanation is needed - it is entertainment for you, and comes with a benefit, like enjoying exercising. Otherwise, it does not seem like a remotely optimally productive way to achieve mastery in any field I am aware of, other than being a student who will be tested on fact memorization.
Memorization focuses on the set of things you want to recall, but don't use often enough to naturally remember.<p>This is most peritent for language learning because you need to 'bootstrap' a large set of words and grammar, and you can't use all of them often enough to put them in long-term memory (at first).<p>Aside from foreign languages, I also use flashcards for English - more difficult words that show up rarely enough that I can't remember their definitions - and country flags.<p>For general learning too, if you need to keep looking something up over and over but can't seem to remember it, flashcards will bootstrap that into your brain and make future learning smoother. Obviously Internet/AI can help - but LLMs can't explain 100% of a topic in their reply, they always assume some level of abstraction, and the higher-level it is the faster you can absorb a topic.
See my other comments here for some of my motivations, but also:<p>Even in the Internet age, getting the latency from "fast" to "effectively zero" has a lot of value for staying in flow, synethesizing information, etc. Your memory is the ultra-low-latency fact retrieval system you always have. No, you definitely don't want to use it for everything, but it definitely does <i>complement</i> modern tools in important ways.
> My correct-answer rate is approximately 89%<p>That sounds like incredibly boring way to spend time. I'd aim for something like 20% at most. What's the fun in being asked things you already know?
The point of spaced repetition is to have you recall something right when you're just about to forget it. If you recall after it's been forgotten, that means you're forgetting a lot more, which means the system isn't working.
This is a really good question!<p>1. As others have said, the idea is to study something <i>before</i> you forget.<p>2. It's hard to predict when you're going to forget something, so you do wind up studying a bunch of stuff before you really have to. It's a limitation of prediction (and also of the technology as developed so far).<p>3. It really is pleasant to work to recall things even when you succeed at it. It does "freshen them up" in your memory. And sometimes just the experience of seeing a fact can be pleasant. (A lot of us review familiar things for the joy of it in other domains--movies, etc.)
I guess I can see how someone might be enjoying this. Especially folks who are into excercise of any kind. But it's not for me. I think there's value in forgetting and re-learning as needed.<p>I learned A* at least 5 times already. And each time I learn it, I feel like I'm having better appreciation and understanding of how it works. Each time I'm falling in different traps, make different mistakes, that teach me more. I wouldn't have that insight if I just memorized how to correctly implement it the first time and recalled it whenever I needed a new implementation. Also I'm perfectly happy not knowing how to implement A* in between times that I need it.<p>I would like to be reminded that things exist, after I forgot them, so 20% sounds way more fun for me.<p>Anyways, thanks for sharing your fun. Don't let me be a buzzkill. ;-)
That sounds nearly perfect for FSRS [1], the default spaced repetition algorithm used by Anki, which aims at estimating the time it takes for memory stability to decline from 100% to 90%. At the estimated 90% stability point, FSRS would require a review, so naturally a mature deck of flashcards would hover between 90-100% stability.<p>1. <a href="https://expertium.github.io/Algorithm.html" rel="nofollow">https://expertium.github.io/Algorithm.html</a>
would be really funny if this was actually someone preparing for who wants to be a millionaire
Are these physical flashcards or virtual, some sort of flashcard software? Definitely writing out 301,432 physical flashcards would be a lot of work, not to mention [kinda] a waste of paper, however many people still like the physical over the virtual; this is me when it comes to reading books.
Bullets 2,3, and 4<p>* I'm using my own software<p>* I did 301,432 flashcard reviews in 2025.<p>* Those reviews covered 52,764 distinct cards.