18 ms·

Effective Spaced Repetition

• roxgib 4y ago
  I've used Anki to great effect on a number of different fields. Using it well it definitely skill you need to learn, but there's nothing quite like it. Strongly recommend.
• topologie 4y ago
  Just a side note: It's easier to remember two connected things/facts than a single thing/fact in isolation...
• maayank 4y ago
  So… anyone used Chatgpt to generate cards for Anki?
  • dantheman0207 4y ago
    It’s certainly an intriguing idea worth investigating. But for me, the process of creating the cards and being forced to distill what I read was critical to the success of using SRS.
  • totetsu 4y ago
    I used it to write a python script to convert a large json file full of a corpus of japanese business phrases and their english translations into anki card format.
  • mjadobson 4y ago
    Yes, I have trialled this. I've found it works best with this workflow: 1. Summarise some prose (from a textbook) into bullet points 2. Convert these to flashcards 3. Make these flashcards more concise I then verify they are correct/relevant, only keeping those that are worthwhile.
  • keiferski 4y ago
    I have been using it to do exactly that. However, as GPT is wrong on certain facts, I use it more for generating Anki-formatted text from learning sources that I provide myself. For example, I’ll give it a paragraph from Wikipedia, then say, “generate 5 questions and answers from this text. Format the answers to be inside brackets like this: {{c1::Answer text}}.
  • SpaceManNabs 4y ago
    Making your own cards is a very important step
    • comfypotato 4y ago
      Science says otherwise. Saw a study with no significant difference between the people who made decks and people who used premade decks. Too lazy to get source, sorry.
      • rjh29 4y ago
        My deck has 10,000 hand-made cards and I know for sure they stick in my brain easier than cards from a premade deck that I have no connection with. I suspect it depends how you make the deck. I intentionally use sentences and images that I care about. It only matters for the first few reviews though, even a premade card can be forced into my mind with enough effort.
        • firsttimebigboy 4y ago
          probably depends on the type of material too. if it's purely memorization (e.g., I'm learning spanish as an english speaker) then there's probably no difference. If it involves any sort of understanding then I'd be very surprised if there was no difference, at least in speed of learning
    • atahanacar 4y ago
      No. As a med student who uses Anki for basically everything, this is plain wrong. The time spent on making cards is better spent on studying. Premade cards (Anking in this case) have relevant pages from multiple sources, relevant histology/gross photos and additional notes for every card, and all cards are tagged with their corresponding chapters in multiple textbooks. If I spent time trying to do something similar (and even without the tags) I'd have no time left to actually study. The only use case I can think of where making cards yourself would be better is sentence mining while learning a language, because these cards depend on context and creation doesn't take more than a few seconds even when adding images.
  • dowakin 4y ago
    Yes, I'm using for language learning. Usually via tool `https://github.com/sigoden/aichat https://github.com/sigoden/aichat` with different roles that I've created. For example for grammar ChatGPT generates bunch of examples for specific grammar rules, which I memorize.
  • knubie 4y ago
    Not Anki, but you can use the AI field in Mochi [0] to generate content for cards [1]. [0] https://mochi.cards/ https://mochi.cards/ [1] https://twitter.com/MochiCardsApp/status/1635289548229062657?s=20 https://twitter.com/MochiCardsApp/status/1635289548229062657...
  • Kelamir 4y ago
    Yes, you can also prompt it with the 20 rules of memorization https://www.supermemo.com/en/blog/twenty-rules-of-formulating-knowledge https://www.supermemo.com/en/blog/twenty-rules-of-formulatin... and you've got yourself an agent which create flashcards for you.
• fwlr 4y ago
  ”Individual cards should be extremely brief, but your deck as a whole can be as repetitive as you want.” Huh! I struggled with using Anki for pretty much this exact reason, always spent too much mental energy figuring out the “correct” number of cards for a given topic. But the author makes a good point here, if there’s too many cards on a certain topic you’ll just hit “I remember” on the repetitive cards and the algorithm will make them disappear for months - so there’s basically no cost to having “too many” flashcards!
  • rjh29 4y ago
    I'm a big fan of making lots of cards, like you say you can just hit Easy and send them into next year, or conversely if you keep forgetting them and they're using up lots of time, just suspend or delete them. When I couldn't remember a particular kanji in Japanese I used to make lots of cards that featured that kanji at the same time, and usually I'd review them all the same time, it's sort of cheating but it always seemed to help me remember the kanji in the end.
    • samus 4y ago
      It's actually the correct way to make that work. Some Hanzi and Kanji are really abstract and are best learned by associating it with multiple related concepts. Each can have dozens of very distinctive meanings.
      • rjh29 4y ago
        Yeah I learned pretty much every kanji with jukugo (did Remembering the Kanji at first, but it wasn't that helpful). A lot of failed cards at first since I was trying to learn recognition of the kanji, the kana reading and the word's meaning - but it got easier over time.
  • spidersouris 4y ago
    The only disadvantage I may see to this is that it skews your counter of daily cards. So you may feel discouraged if you see that you have a large amount of cards to review and that it will probably take you a lot of time to go through them all, while in fact, you will only spend a few seconds on those that you already remember. Unfortunately, there are times when I cannot for the life of me do my flashcards because there are too many of them to review, and I either skip that day's session because a) I tell myself I don't have enough time (which is a lie most of the time!) or b) I don't have the mental energy to go through them. The only (partial) solution I have found to that is doing them first thing in the morning by trying to plan my schedule accordingly and wake up a tad earlier.
• ranting-moth 4y ago
  > Spaced repetition is, by far, the most effective cognitive hack I’ve used. I totally second that. Well, sleep and exercise do amazing things for you too, but if you need to memorize things, I don't know of a quicker method than spaced repetition. Check out Leitner system: https://en.m.wikipedia.org/wiki/Leitner_system https://en.m.wikipedia.org/wiki/Leitner_system Flashcards Deluxe supports Leitner, I don't know about Anki? https://play.google.com/store/apps/details?id=com.orangeorapple.flashcards https://play.google.com/store/apps/details?id=com.orangeorap... (I have no relation to that app, I just find it incredibly useful).
  • shen23 4y ago
    Huge fan of SRS here. Anki never really gelled with me personally. I still use it but really prefer this app called StickyStudy. It’s mostly for learning Chinese / Japanese but there was a generic version which has sadly been abandoned by the dev. It has very straightforward SRS / leitner system and a bit more simplified ui. Not my app but just wanted to share and maybe inspire someone to make a successor ;-)
  • fnl 4y ago
    Thanks for the pointer to the Leitner system (my last name coincides…) But why would you want to use that simplified Leitner system if Anki can track each card individually? Disclaimer: I don’t know if Anki supports a simplified system that applies spacing to a whole group of cards instead of each card individually, but why you would want that?
    • alwayslikethis 4y ago
      You don't, and it is not desirable. The Leitner system was devised before powerful enough computers were widely available.
  • roxgib 4y ago
    The Leitner system is designed with the assumption that you have lots of physical cards you need to organise without going insane, but you can do much better if you're using a program like Anki. Anki's algorithm is based on an older version of the SuperMemo algorithm, with some evolution since then. The consensus seems to be that while you can fine tune it a bit with the settings, the returns are diminishing at this point.
• bulldog13 4y ago
  I use SuperMemo https://super-memo.com/ https://super-memo.com/ for memorizing things. It uses spaced repetition. It has a bit of a learning curve, but once you figure it out, very effective. I feel like I have seen an article on HN about it before.
  • pleasejustdont 4y ago
    Probably this one : Augmenting Long-term Memory http://augmentingcognition.com/ltm.html http://augmentingcognition.com/ltm.html
• oslac 4y ago
  Tending to a digital garden of your own making has been the most effective for me by far. No other organization beyond that, at most hitting a random note inside a subfolder for 10-15 minutes a day.
• ekkeke 4y ago
  Big fan of spaced repitition, especially for language learning. Unfortunately I feel like it fares worse for topics that require more application instead of memorisation, like mathematics or electrical engineering. Would love know if there was some super effective way to learn these similar to spaced repitiion. So far, the only thing that really works for me is solving lots of problems until I have the technique mastered, but even then after a while I'm prone to forget how to solve them. Perhaps there some way to combine the problem solving with the spaced repition? It seems like it would be far harder to make a deck for this and I don't think most flashcard software handles it very well.
  • nomadpenguin 4y ago
    I've been experimenting with "spaced free recall". So first, I'll read a section of a textbook. Then, I write down everything I can remember about it in a blank text file, organizing things in a way that makes sense to me. Next, look back at the section and compare to my recalled notes, filling in missing information and committing extra attention to missed spots. Repeat the process with increasing intervals between reviews. From what I understand of the literature, free recall produces better learning compared to cued recall like flash cards. Part of the reason is that it forces you to organize information and associate it with existing knowledge. Anecdotally, it's much easier to learn conceptual knowledge, and I don't really feel like my recall of specific facts has suffered compared to traditional SRS.
  • schneems 4y ago
    > require more application instead of memorisation, like mathematics or electrical engineering I’ve dreamed of having some app that mixes in bite sized learning lessons with otherwise “fun” internet (social media, news, etc.) I could imagine it could give you a little tutorial and then ask you a quiz (to force application). If you get it wrong it keeps you at the same concept and explains it a different way next time, maybe asks if you want to revisit prereqs. Even if you can’t memorize the answers, you can change your understanding and intuition.
    • krsrhe 4y ago
      [dead]
• endisneigh 4y ago
  spaced repetition is dying for some amazing UX. it doesn't even really make sense for you to make flash cards. ideally they could be contextually created based on what you're viewing, e.g. you read a wikipedia article, it infers what you're reading from scroll position, takes content, makes cards, presents later, etc.
  • Veen 4y ago
    You may be interested in Andy Matuschak and Michael Nielsen's work on mnemonic media. - https://andymatuschak.org/ https://andymatuschak.org/ - https://withorbit.com/ https://withorbit.com/ - https://numinous.productions/ttft/ https://numinous.productions/ttft/
  • _dain_ 4y ago
    >it doesn't even really make sense for you to make flash cards. ideally they could be contextually created based on what you're viewing It does matter. You shouldn't train on a flashcard until you have learned the idea. The computer doesn't know if you actually learned what you read. Making the flashcard based on your own understanding is an important part of encoding the memory -- it's an active process, rather than passive. It's well established in communities that use Anki a lot (like language learners) that someone else's pre-made decks aren't as effective as making your own. The exceptions are either small and simple (e.g. NATO phonetic alphabet), or had a lot of thought put into them with community feedback, like the ones medical students use.
    • endisneigh 4y ago
      you're talking about training on flashcards. I'm talking about making the flash cards. of course using other people's cards isn't going to work (as well) - you don't have any context around them.
    • Al-Khwarizmi 4y ago
      It's well established in communities that use Anki a lot (like language learners) that someone else's pre-made decks aren't as effective as making your own. This is indeed a usual claim. But given a reasonably good pre-made deck, I doubt it's universally true if you factor in the time it takes to make the cards. I'm learning Chinese vocabulary with a pre-made deck. I devote 20-30 minutes per day (have been doing so for a couple of years), introducing 8 new cards per day. My success rate is typically worse than for most people who make their own decks (floating around ~80% for mature cards) but I see clear and steady progress. If I wanted to create my own cards, I would probably need to slow the pace to around 2-3 cards per day tops (as deciding which cards to create and then creating them would take time which I wouldn't be able to spend studying), and they would have no audio, which the pre-made deck does and makes a difference.
• robertbob 4y ago
  Does anyone have any insight into deciding what information should be memorised, and what information it is sufficient to simply store in a searchable digital knowledge base for rapid retrieval when needed? Takes a lot of effort to commit my notes from a book into my head, but a tiny amount of resources to store them on my computer.
  • rjh29 4y ago
    Language is really the ideal use case, you cannot stop and look up words every few seconds when talking to someone. And doing it while reading or watching TV spoils the enjoyment somewhat. You need to frontload a ton of data into your mind and flashcards are the best way to do it. Another good use case is country flags, because you can't easily look those up (other than pulling up an image of ALL the country flags)
    • bluGill 4y ago
      Only at the beginner levels. Once you get beyond beginner you realize that many words have 20 different definitions, and each definition maps to a different word in the other language - and the problem happens when you reverse as well.
      • rjh29 4y ago
        I'm not a beginner (12k cards, 10 years). What works for me personally: 1) test word -> meaning, not word -> word in other language 2) I don't test reverse most of the time, unless I really want to remember how to say something - output tends to emerge by itself given enough input 3) When a word has multiple meanings I add context (usually a related word or colocation as a hint - whole sentences can cause you to associate the sentence with the word but not recall it in other situations). Words with lots of meanings (in Japanese I guess that would be stuff like 引く or in English 'run'?) tend to be very common words so you don't need to use flash cards so much anyway.
  • outlace 4y ago
    Ask yourself if the cost of having to look it up at a potentially inconvenient time (e.g. in the middle of a busy work day) is greater than the cost of memorizing it during scheduled less busy times (e.g. doing flashcards while eating dinner or during a bus commute).
• yanis_t 4y ago
  Spaced repetition looks very promising to me. I've been a long-time Anki [1] user, and it allowed me to learn Czech much faster. Recently I launched a website [2] where I try to blend a markdown-based knowledge manager with spaced repetition. It's not an easy task and there's a long way to go, but after adding and maintaining > 400 cards, I already [1] https://apps.ankiweb.net/ https://apps.ankiweb.net/ [2] https://retaind.io/ https://retaind.io/
  • mburee 4y ago
    Basically trying to do exactly the same thing, do you have any recommendations on decks or are you maintaining your own?
  • wanderingbit 4y ago
    Upon reading the OP's blog, I had the same idea! It's cool to see others are hungry for this as well. My basic approach to implementing it in Notion (my preferred knowledge manager) is: 1. build some block template in notion that lets me structure the block such that it’s easy for Mochi to ingest it 2. reads new pages from Notion 3. uses Mochi to generate cards for them 4. build a habit of reviewing Mochi cards, then I will actually learn from the things I write in Notion It looks like you've implemented the core functionality in retaind. How come you decided to go with a separate knowledge base, rather than try to use one that already has strong network effects, like Notion?
    • austincassidy 4y ago
      I've been trying to find a solution to this exact problem. I use Notion as my PKMS and I've been struggling to find an integration to make flashcards with my content in Notion. Going to try out your system - thanks for sharing! As an aside, Notion would completely disrupt the flashcard app space (Quizlet, etc) if it had a feature to create flashcards from a Notion database with spaced repetition. I really don't think it would be that difficult for them to implement that feature. Hopefully they will soon.
  • AlexErrant 4y ago
    This looks incorrect to me https://i.imgur.com/pECgx2j.png https://i.imgur.com/pECgx2j.png
• bulbosaur123 4y ago
  Combine spaced repetition with major mnemonic system and method of Loki and you have a superpower.
  • all2 4y ago
    Can you mention some major mnemonic systems? I've read about this stuff before and I remember a few minor ones for numbers, or for chaining ideas, but that's about it.
    • bulbosaur123 4y ago
      Choose one of these and stick with it: https://en.wikipedia.org/wiki/Mnemonic_major_system https://en.wikipedia.org/wiki/Mnemonic_major_system https://en.wikipedia.org/wiki/Dominic_system https://en.wikipedia.org/wiki/Dominic_system Both are good. You can train mnemonics via Anki deck: https://ankiweb.net/shared/info/2032755489 https://ankiweb.net/shared/info/2032755489
• cdelsolar 4y ago
  I am a tournament Scrabble player and the state of the art for studying words is spaced repetition. You quiz on "alphagrams", like ABEISTT, and after a few times you just see BATISTE BISTATE. There are at least 80K words between 2 and 8 letters long though, so it does take many hundreds or thousands of hours to learn them all well. I have very poor studying discipline so I have my own methods of studying that don't use spaced repetition, instead I just study all the words periodically and focus on the harder ones. But I don't know them as well, and for the words I did spaced repetition on more than 15 years ago I can still recall those immediately.
  • LVB 4y ago
    This is interesting to me. Do you make many permutations and use those as cards?
    • cdelsolar 4y ago
      No - it's more efficient to just study a single permutation (we call it an alphagram -- the alphabetized letters on your rack). Then, when you are playing, just sort your rack alphabetically and watch spaced repetition at work.
• zetalyrae 4y ago
  I wrote this. Ask me anything!
  • hendry 4y ago
    I find writing cards / decks the challenge. IIRC the ones you download from Ankiweb are all such low quality! Could you link to some good quality decks for inspiration please?
    • zetalyrae 4y ago
      I write essentially all of my cards. I find that it helps a bit with recall. Also, people differ in the "density" of cards they need around a given topic. I might need a lot of cards to cover a particular area of a topic in many different ways, while someone who already knows that area needs fewer. I agree that public decks tend to be terrible. So I don't really know any good ones. For the examples in the post, I tried to keep the wording as close as possible to the flashcards in my own Mochi decks, without adding extraneous detail.
  • untech 4y ago
    Thanks for the article! I noticed a small issue: the “Powers of two” subsection didn’t render properly, probably due to a Markdown syntax error.
    • zetalyrae 4y ago
      Thanks! Fixed this.
  • bluechair 4y ago
    Thanks for the write up. Could you share with us how you’re applying this knowledge in your work?
    • zetalyrae 4y ago
      I mostly use SR so I can study intensely whatever strikes me as interesting, then turn to the next thing, while still retaining the knowledge and being able to make progress. Like a few months back I had the sudden inexplicable autistic urge to learn geology. So, I picked up a textbook and went through it like a novel and just wrote the flashcards as I read. I think of it as checkpointing for learning: https://borretti.me/article/spaced-repetition-checkpointing-learning https://borretti.me/article/spaced-repetition-checkpointing-... Currently going through Jaynes' Probability Theory: The Logic of Science.
• dan-g 4y ago
  Andy Matuschak’s How to Write Good Prompts is another good resource about this: https://andymatuschak.org/prompts/ https://andymatuschak.org/prompts/
  • AlexErrant 4y ago
    Going to add Supermemo's classic "Twenty rules of formulating knowledge", which is referenced at the end of the article. https://www.supermemo.com/en/blog/twenty-rules-of-formulating-knowledge https://www.supermemo.com/en/blog/twenty-rules-of-formulatin...
    • kqr 4y ago
      I wish these comments were higher up in the discussion. These things are essential reading for new users who want to succeed with spaced repetition.
    • sn9 4y ago
      Michael Nielsen has also written some influential stuff on SRS [0][1]. [0] http://augmentingcognition.com/ltm.html http://augmentingcognition.com/ltm.html [1] https://cognitivemedium.com/srs-mathematics https://cognitivemedium.com/srs-mathematics
• matt_LLVW 4y ago
  I just learned hiragana and katakana in a week using spaced repetition. It's crazy effective.
• mattgreenrocks 4y ago
  I need to memorize the shapes of seven modes on the guitar. (A mode is kind of like a scale.) The image here shows them starting from F: https://www.anyonecanplayguitar.co.uk/three-note-per-string-scales/ https://www.anyonecanplayguitar.co.uk/three-note-per-string-... Is this a good fit for this technique? I'm not sure how to decompose it further than "G Dorian." I'm also not sure if the time would be better spent just playing them more.
  • submeta 4y ago
    Definitely! You can use the image occlusion addon to hide parts of the images and try to recall it. Edit: Here is a link to the mentioned add on: https://ankiweb.net/shared/info/1374772155 https://ankiweb.net/shared/info/1374772155 Create screenshots of your seven shapes, then redact parts of it.
  • damontal 4y ago
    I can't learn guitar looking at fretboard diagrams like that. They completely confuse me. The only way I learn shapes like that is by repeatedly playing them.
  • pohl 4y ago
    One could, but I think there are better ways that involve doing the thing. In other words, practice patterns on top of the modes (up 2 tones, down one, [repeat], etc.) Also, there are more simple ways to conceptualize these patterns that make them much easier to digest: https://www.youtube.com/watch?v=CGmj2kuHojQ https://www.youtube.com/watch?v=CGmj2kuHojQ
  • dbalatero 4y ago
    You can use SRS to prompt the review, but do the actual "review" physically on the guitar. But yeah I think you need to build kinetic memory here, not just visual memory.
  • fsociety 4y ago
    Play them more, your fingers need to learn the shapes not your brain.
  • 4y ago
• albert_e 4y ago
  I read so many good things about spaced-repetition but havent had the discipline to stick with it and make it work for me ... I wil give it one more shot with flashcards on a exam prep I am about to embark on. Is there any gentle kid-friendly introduction to this topic with a fun exercise that I can introduce my K-12 kids to so they might grow up with better tools than me?
  • Kelamir 4y ago
    https://ncase.me/remember/ https://ncase.me/remember/ could help a bit.
• TheRealPomax 4y ago
  This could do with some non-serif fontage. Effective learning includes having a typeface that doesn't get in the way of reading at smaller pointsize for a large segment of the population.
  • n8henrie 4y ago
    TIL -- I thought serif was generally supposed to be easier to read. How is this kind of thing studied, and what kind of outcomes suggest "easier to read"? Speed? Recall? Subjective? What proportion of the population are we talking about, roughly? Is it even enough that one should consider providing both a serif and a sans version of their content? Serious question.
    • TheRealPomax 4y ago
      Worldwide, at least 10% of the audience (but depending on your specific demographic, much higher) which is a crazy significant number (for contrast: your users are three times more likely to have some form of dyslexia than they are likely to be using Firefox). For anyone with (partial) dyslexia the serifs are information-murder. https://www.bdadyslexia.org.uk/advice/employers/creating-a-dyslexia-friendly-workplace/dyslexia-friendly-style-guide https://www.bdadyslexia.org.uk/advice/employers/creating-a-d... But also note that font size and spacing matters a lot: at 16px with 24px line spacing, this is super hard to read. Change that to font-size 20 with the same 24px spacing, and this read perfectly fine.
      • n8henrie 4y ago
        Thanks for the response and link! I'll see if I can find some primary literature on the topic -- I'm interested in how such a thing is studied! Accessibility is certainly one of my many weak points.
  • lbotos 4y ago
    Is your beef the serifs, or browser's justification doing a number on the word spacing? I find it easier to read with the justification removed.
• submeta 4y ago
  A question to the experienced Anki users: The recommendation is to learn before you memorize. Many times I hack the infos (from an article) right into Anki. Now if it‘d want to review the infos in the Anki app, that totally destroys my stats. What are some of you doing? Extract the info into some other tool, review the infos there and then quiz yourself in Aki?
  • atahanacar 4y ago
    >that totally destroys my stats I don't understand what you mean. There are only 2 stats that are mostly significant. The first is, when first learning the card, if you hit again too many times the card becomes a leech - as in it leeches your time, at which point you should temporarily suspend that card and to the others first. The other one is the recall rate: after you learn a card (on default settings, when you hit the 1 day mark for that card) how often you don't fail to recall the information. This rate is the single most important metric in your stats, and a value that is constantly below 90% means you should probably tweak your settings to show cards more often. This can be easily done by reducing the interval modifier. A value close to 100% might also be slightly problematic depending on what you want, because you might be wasting time with unnecessary reviews, which can be solved by slightly increasing the said interval modifier. As for your main question, you can do it however you see fit. Some people might prefer to do everything in Anki, while others might prefer to read a textbook, watch lectures etc. first and then move on to Anki for that specific topic. Don't use Anki as a way to quiz yourself, it is not a testing tool. If you want to test yourself, use question banks. Use Anki as a tool that will help you memorize things. For this, you have to do your reviews on time, or the algorithm just won't work.
  • Graziano_M 4y ago
    Why would you care about your stats? Do you have the stuff memorized? Nothing else matters.
  • sureglymop 4y ago
    I have a similar issue. I create cards for the subject I need to study, then I study it all at once in a custom study session. I then go to the exams but how do I continue afterwards? Like what are the best settings to retain all the information after that? And how do I make sure that I'm not destroying those settings when I need to study before an exam again?
• MisterPea 4y ago
  Having tried spaced repetition methods for studying for swe interviews, I can concur that it is the most effective way for me learn. It does require an intense amount of discipline though, so wonder how well it will work for me in execution for hobby learning.
  • ngai_aku 4y ago
    What kind of cards did you make for interview prep? Reviewing algorithms?
    • MisterPea 4y ago
      Yep, just LC problems - helps a lot with identifying patterns
      • harshalizee 4y ago
        Would you mind sharing your setup and/or deck?
        • rsanek 4y ago
          For my job search process I created a custom note type specifically for interview problems. My general process was go to LeetCode, find a medium/hard problem, hack on it for 30-60 minutes, then look at the solution if I couldn't get there myself. At the end of the problem, regardless of if I solved it or not, I'd create an Anki card with the following fields: Title Question Additional Criteria Example input/output Insight (1 sentence maximum) Insight explanation (can be longer/bullet-pointed list) Key Data Structure (at most 1 data structure; if there are multiple, use the most important one) Time complexity Space complexity Full answer code (can use syntax highlighter add-on) Source (can provide link to associated question online; can include link(s) to solutions that the insight and/or code come from) There are 4 cards that are generated from this template, which test the same question in slightly different ways. They individually ask for the insight, the key data structure, and the time and space complexities. I found this note type to be critical to my success in the following interviews. In two cases, I was asked literally the same exact question I had already added to Anki; I was able to write out the solution from memory in one go. If you'd like to use my note type directly, I've exported an example here. [0] [0] https://drive.google.com/file/d/12NsYNIBjIPI1Nhq5wE1xPljr9rHUErzi/view https://drive.google.com/file/d/12NsYNIBjIPI1Nhq5wE1xPljr9rH...
• tiagod 4y ago
  >A common failure mode (and I did this more than once, before I got the hang of it) is to use Anki for two weeks, then drop it, and pick it back up six months later only to find you have 600 cards due for review. This is not encouraging, and it defeats the point of spaced repetition, which is to review the cards on the intervals the algorithm chooses. >I don’t have much advice in this area, except that if you have persistent problems with conscientiousness, untreated ADHD etc. you should address that first. This keeps happening to me, and I have somewhat treated (but severe) ADHD. Does anyone have recommendations to make this easier? Either Anki settings, or using another app.
  • sangpal 4y ago
    You could maybe use an alarm/reminder.
  • Scaevolus 4y ago
    Assuming you use a computer daily, it should be easy to script Anki to open itself at startup at some point during the day, letting you effortlessly have a habit of sitting down and doing 5 minutes of review before your other tasks.
  • cafemachiavelli 4y ago
    The only thing that worked for me (also ADHD) is to have fixed study sessions, typically in the morning ("Kanji before breakfast") or during my train commute. Beyond that, I wouldn't worry about backlog too much. I use a custom deck that gives me my maximally doable number of cards per day and completely ignore the rest. It's not like the actual repetition frequency is critical - many people change those settings quite drastically and still manage to memorize their cards fine.
  • samus 4y ago
    If you don't touch it for that long, just throw away the scorefile and start from new. It probably doesn't fit the degree you actually remember any of the things that you learned.
  • examplary_cable 4y ago
    The issue you're having is due to the fact that Anki "Accumulates" cards if you skip one day, which can build up and create such a large amount of scheduled cards for a day that you end up dropping the thing. I'm working on a spaced repetition algorithm that solves this issue by letting you review when you have time and letting you skip the days you can't do the reviews.
• n8henrie 4y ago
  Used SRS via Mnemosyne (and later Repetitions.app) heavily in studying for all of my US medical board exams. The effort to payoff ratio seemed very satisfactory. For pre-clinical rotations, a few nerdy peers and I collaborated on a shared deck of a couple thousand slides -- many of them pathology images -- synced via Dropbox. For the Step exams, I mostly used practice test questions. Any question I missed prompted me to read up on the topic to determine what piece of knowledge would have helped me come to the right answer, and then figure out how to make a decent card for that principle. Every morning I would start by reviewing all my SRS cards, then do a few hours of practice tests. It was really nice being able to be able to take a core component m of my study material on the road by just bringing my phone! A few of the practice question apps had protections in place to prevent copying text (copy and paste saved a fair bit of time, even if there was also a lot of rewriting) -- but figuring out workarounds like running in a VM or screenshotting from the iOS app was never too hard, and I would just queue up the screenshots to batch process toward the end of the day. I used a similar technique to pass the tech, general, and extra exams for amateur radio with near perfect scores (the verbatim questions and answers are freely available -- I did try to learn the concepts behind most questions, though a few were admittedly memorization without understanding). Unfortunately my small town is too rural to have a local club, and I have too many hobbies to shell out $1k for a HF rig, so I have yet to make a single QSO. I'll eventually find time to put together the QRP CW kit I got for my birthday :)
• Avijit_Thawani 4y ago
  Other software which have nailed spaced repetition (and a much better UI than Anki): - https://www.duolingo.com/ https://www.duolingo.com/ (language learninig) - https://www.chessable.com/ https://www.chessable.com/ (chess) - https://readwise.io/ https://readwise.io/ (book highlights) - https://examarly.com/ https://examarly.com/ (test prep) - https://app.bestudious.io/login https://app.bestudious.io/login (certifications - CFA) - https://magoosh.com/ https://magoosh.com/ (vocabulary)
  • edylemond 4y ago
    If you're interested in a more visual approach you can try https://traverse.link/ https://traverse.link/ - it's an app I created which has spaced repetition, but really its goal is to cover the whole learning process, so it also has mind mapping and note-taking so you get a big picture view of what you're learning, why reinforcing bottom-up with spaced repetition
  • ChadNauseam 4y ago
    I’d like to add Mochi to the list https://mochi.cards/ https://mochi.cards/ It’s just a standard flashcard tool like Anki, but with a much better user interface and a simpler (IMO superior) SRS algorithm.
    • Avijit_Thawani 4y ago
      totally love it as an alternative to Anki, didn't mention it since it has already been praised by the original post
    • greymalik 4y ago
      And a subscription required for key functions like multi-device sync.
  • jayro 4y ago
    At Math Academy (https://mathacademy.com https://mathacademy.com), we implement spaced repetition in combination with a knowledge graph consisting of several thousand math topics and tens of thousands of connections (and growing). We're working on a post that explains how this all works technically. We have a Linear Algebra course (https://mathacademy.com/courses/linear-algebra https://mathacademy.com/courses/linear-algebra) that some of you might find interesting given how often that topic shows up on HN, and we just finished our Math for Machine Learning course (https://mathacademy.com/courses/mathematics-for-machine-learning https://mathacademy.com/courses/mathematics-for-machine-lear...) for anyone who might be interested in giving that a look. I'm the founder if anyone has any questions. Sorry about the shameless plug.;)
• kashunstva 4y ago
  A piece of advice that is repeated here and elsewhere is to create atomic cards. I’ve long wondered whether that is good advice or not, or whether it depends on the subject domain. The better I know a subject, the less atomic it feels. I don’t think and talk about the topic in discrete facts but in interconnected knowledge. This is why the complete independence between cards seems more like a weakness than a strength. The argument from the spaced repetition community is that purposeful linkages between cards builds in a contextual dependency. And that dependency makes the resulting memory contingent. But in some knowledge domains that seems more of a feature than a bug.
  • wordpad25 4y ago
    @kashunstva that's really interesting, how would you link cards together, though? Have you seen this done anywhere?
  • Dr_Birdbrain 4y ago
    I have been a regular Anki user for many years. My approach to making cards is actually to take long-form notes on the topic at hand, then apply close deletions to key parts, like equations, key ideas, and definitions. This makes it so that the piece I need to actively recall is small(ish), but I still have the context
• somsak2 4y ago
  >A common failure mode (and I did this more than once, before I got the hang of it) is to use Anki for two weeks, then drop it, and pick it back up six months later only to find you have 600 cards due for review. This is not encouraging, and it defeats the point of spaced repetition, which is to review the cards on the intervals the algorithm chooses. For successful long-term use of a spaced repetition program, I believe that this model of thinking is unsustainable for most. I used to have this relationship but had to grow past it as my life circumstances changed and prevented me from having that consistent amount of time every single day. Now, I look at Anki as a way to prioritize my learning time. When I get to it, I have it present me the things that are the most-overdue first. This has meant that I've gone from a typical backlog of ~0 reviews at the end of the day to flexing between ~500-2500 backlogged reviews. Just because I'm not reviewing the piece of information at the exact right time doesn't mean that I'm "defeating the point" of the piece of software. Spaced repetition, even if done imperfectly, is still many times more efficient than traditional study methods.
  • reitanqild 4y ago
    > Just because I'm not reviewing the piece of information at the exact right time doesn't mean that I'm "defeating the point" Some of us, thanks to personality and previous experience (diagnosis and/or childhood and/or early work experiences etc) needs to hear this again and again it seems.
• rendall 4y ago
  What papers or data out there support spaced repetition? Has its support weathered the replication crisis?
  • rodonn 4y ago
    Spaced repetition has a very long history with hundreds of papers published on the topic. Here's a good survey of some of the published research https://gwern.net/spaced-repetition#background-testing-works https://gwern.net/spaced-repetition#background-testing-works
• all2 4y ago
  I've started working on my own code specific flashcard program. The algorithm, app, and cards are one part of the app, and the other part of the app -- the actual cards -- is whatever application you want. Basically, I wanted a flashcard app mashed together with repl.it. At first I thought I'd just have the python interpreter built in, but I slowly realized that correctly implemented flashcard is a programming template and a correct answer, and that your choice of interpreter should produce the correct answer given your input. --- So now I've come to the general formulation of the app: Question -> Some application specific file. For Python it might be the boilerplate into which you insert a nested list comprehension. For C# it could be a whole project boilerplate into which you insert your LINQ statement. Answer -> The expected output when your program is done executing. Probably on STDIO? Maybe something else. There are lots of IO options in an Linux OS. Card -> Question, Answer, Executable (we could probably infer this from the "question" card) --- Everything after this would be pretty much like Anki as it is now. You could have a script that listens to you play a song and then scores your performance. You could take touch/writing input and score penmanship. And so on. The possibilities here are limited by your imagination. Unfortunately, I got derailed trying to sort out the review algorithm. If anyone has any ideas on this front, I'm all ears. I know Anki's algorithm punishes skipping review days, and I'm one who skips review days frequently. Which leads to me quitting altogether because the review load becomes boring and unbearable.
• throwaway675309 4y ago
  "This is the most important thing. By far the worst failure mode is to put too much in a flashcard." He's wrong on this one. The problem with making extremely atomic flashcards is that many times it's impossible to prevent slight amounts of overlap between the cards. What this means is that answering card A might contain information that allows you to answer some subsequent card B. As a result you end up effectively getting a hint for some dependent cards, as opposed to if you had to recall all that information from scratch.
  • swinglyj 4y ago
    Overlap isn't actually an issue. Overtime Anki will spread the cards out more so you don't see them together.
    • jbstack 4y ago
      You can also make the related cards as siblings and use the bury feature so that you can only ever see one on a given day. I frequently use this to learn lists e.g.: Card 1 front: Countries in the UK: 1. [...] 2. 3. 4. Card 1 back: Countries in the UK: 1. England 2. Scotland 3. Wales 4. Northern Ireland Card 2 front: Countries in the UK: 1. England 2. [...] 3. 4. Card 2 back: Countries in the UK: 1. England 2. Scotland 3. Wales 4. Northern Ireland Card 3 front: Countries in the UK: 1. 2. Scotland 3. [...] 4. Card 3 back: Countries in the UK: 1. England 2. Scotland 3. Wales 4. Northern Ireland Card 4 front: Countries in the UK: 1. 2. 3. Wales 4. [...] Card 4 back: Countries in the UK: 1. England 2. Scotland 3. Wales 4. Northern Ireland Because the cards are siblings, I'll only ever be tested on one in a given day. Each item in the list (aside from the first) is prompted by the preceding item, so that your brain learns to generate the entire list from memory sequentially. The same technique can be used to memorise a long quote by breaking it down into a list of chunks. I also have a note type that consists of multiple cloze fields so that I can capture related information as siblings.
• quickthrower2 4y ago
  Thanks for this resource. It is timely for me, because I am finding memory is probably the limiting factor for me learning about machine learning. So many concepts and names of things to remember. Once you get past 1-2 lectures the short term memory becomes overloaded and it gets muddled. I also worry that chatgpt is going to make my memory worse, because I no longer have to remember stuff, and I can just remember what they asked chat gpt. I think there is value to even remembering APIs and syntax from heart. So this post is counter "just get the computer to do it" which is a good thing. I can imagine that just like people didn't need gyms when they had to hunt, gather and survive, but now we do, we will need mental gyms to make up for what technology has made too easy.
• wisemanwillhear 4y ago
  A few things that I've come to learn using space repetition systems... * You are effectively memorizing. Memorizing something takes a long time compared to other learning techniques like learning concepts and looking up details as needed. I find it works best if I'm selective about what to memorize since memorizing everything limits the breadth of what I can learn in general. * Mnemonic devices, even ones that seem silly, are incredibly valuable for things that are hard to remember. It may take time to find or think up a mnemonic, but it's worth the up-front effort. * If you naturally have unavoidable life interruptions every week/month that require you to skip learning days, you may be happier with more forgiving algorithms. Algorithms like the supermemo algorithm used by Anki try to push you right up to the edge of you ability to remember it. I still maintain old flashcards in those systems, but I'm much happier since I started tracking my learning in a custom spreadsheet where I control the algorithm.
  • ReC757 4y ago
    I use ChatGPT to create Mnemonic devices out of the lines from my favorite movies for my Accounting flashcards :)
• qumpis 4y ago
  Has anyone used spaced repetition in their research/job for a problem they are trying to solve? I sometimes feel like there are many aspects I need to hold in my brain at once and keep having the need to revisit them. If I could chunk these aspects through spaced repetition it seems things could be easier. I know Michael Nielson got insights by using spaces repetition for some math properties, but my usecase is a bit different.
• asdfman123 4y ago
  > You won’t get smarter by… playing the violin Yes, you absolutely will
  • Gatsky 4y ago
    This is reversing the causality I think.
    • asdfman123 4y ago
      Anecdotally, musical instruments are a constant, unending brain game. If they don’t make you smarter, almost nothing will. Of course, there’s the studies, but we’re all capable of looking those up.
      • zetalyrae 4y ago
        >If they don’t make you smarter, almost nothing will. Because nothing makes you smarter. There are no means of intelligence enhancement, other than correcting deficiencies (e.g. iodine) in early development.
        • zamnos 4y ago
          Au contraire! TMS, transcranial magnetic stimulation, and related sort of magnetic-field-on-brain things have early results that say it can increase working memory, improve logic/math/3d visualization skills, and push motivation.
        • asdfman123 4y ago
          So school doesn’t either? If that’s true, then I don’t think we mean the same thing when we say “intelligence.”
          • alwayslikethis 4y ago
            School doesn't make people more intelligent. It makes them more educated, which can mean a variety of things depending on the cultural context, such as an understanding of social mores, skills such as reading and writing or arithmetic, and/or ideological indoctrination.
• chrismarlow9 4y ago
  Actively building a memory palace while I read or listen to things was a game changer for me. It's very draining though so I can only do it for about 30 minutes at a time but generally that's enough. I'm sure spaced repetition works well for most but it felt like work and I didn't want to take joy out of learning. Memory palace was a way to do that.
• laurieg 4y ago
  When people start using spaced repetition and flashcard software they often end up throwing a big list of "Most common 1000 French words" into it and then promptly being swamped by reviews and giving up. This is a bad idea for two reasons: When it comes to spaced repetition, quality beats quantity. Only put in the perfect nuggets of knowledge. You're walking through an orchard and grabbing a single apple, not a whole trees worth. Spaced repetition is to stop you forgetting, not for learning brand new things. When you study a topic and have an "aha" moment or you make a new connection, that's the thing you should put in. When you make the flashcard it should feel a little too easy. "I could never forget this, it seems so obvious now!". Doing this means your reviews are almost all easy and remind you of your learning experiences. The few things you forget are easily recalled and strengthened. Joyful and fun review sessions are the most important thing. Maximising adherence (and therefore not giving up) is far more important than squeezing out a 1% more efficient SRS algorithm.
  • icoder 4y ago
    This is exactly me, although I've had streaks where I just ploughed through the big backlogs just throwing a lot of time against the problem (having a kid now, I can no longer reside to this option) Question: for language learning: is there anything to say about or help with this initial learning? So before then starting the spaced repetition.
  • Abishek_Muthian 4y ago
    That's a solid advice. > promptly being swamped by reviews and giving up I faced Anki review hell even with small decks, So I had to create an always-on Anki e-Ink review system[1] which sits on my desk and shows me cards all day; Now I can choose to review the cards through out the day during small breaks without overwhelming myself with a single review session. Due to the small real-estate (screen), I have to condense the information to what's really necessary and the act of doing it (on Anki desktop) itself helps me with my learning. What I'm doing might be heretic to Spaced Repetition folks, But it works for me. [1] https://memoryhammer.com/ https://memoryhammer.com/
  • eska 4y ago
    Strong disagree. I learned multiple languages this way. I think you’re better served by turning off the counter for cards remaining, and studying more often. Otherwise you’re not doing SRS at all. One can’t remember 2300 Kanji as an aha-moment each. Drilling all the possible combinations of Hangeul characters for recognition speed is very effective, but most cards won’t be an aha-moment. One can use premade lists of vocabulary and prioritize vocabulary encountered in study material/real life. You’ll have to learn all the other vocabulary eventually anyway. Decks with huge amounts of example sentences and audio simulate real conversation and drill listening comprehension and applied grammar.
• janwillemb 4y ago
  I tried this a while, but then I realized that I'm mostly reading for fun. I don't want to memorize what I'm reading. Almost never. Even whenever I really need to memorize stuff, I ended up using the memory palace technique, with Anki as "backend repository".
• AntonCTO 4y ago
  Overall great tips. Worth trying, but they are very personal. For example, asking in multiple ways will not really help me. I am working on a learning app/system. The reason for that is, that it is very difficult for me to remember staff. I tried some apps and systems like spaced repetition/leitner, but they didn't work well for me, especially when there are a lot of cards. Supermemo is the closest thing that has worked for me. After that I gave up and experimented with plain paper based maps and probably found the best way for me to learn. Then I implemented an improved version of it, mainly because the computer can store more historical information. At its core, the system measures the difficulty of each card and creates appropriate learning batches. I have a working prototype, and so far I have gotten amazing feedback on it. I have introduced a version for people who have even more difficulties to learn. It has a long way to go, but if anyone is interested in trying it out when it goes public next year, please let me know.
• michaelcampbell 4y ago
  > Rule: Understand First Good in theory. Sometimes time crunches don't allow that, and to be honest after having used Anki for many years there are quite a lot of times where I rote-memorized a card and as I learned more about the subject the understanding came later. Usually this is when there are several cards hitting different aspects of a given topic; as I gain understanding in one area, it comes in another. So this is very much a "nice to have" rule of thumb. For me, people vary. But I'd rather have the card and come to understanding than not have it.
  • jbstack 4y ago
    I've found a similar logic applies to language learning. There's plenty of advice out there dismissing book / flashcard learning and saying you just need to get out and practice speaking with natives. In reality, you can't understand someone or form a sentence in reply if you don't know any of the words. Memorising thousands of words / phrases might not make you fluent but it's an important first step. When you're confronted by a real situation you'll now be able, with a lot of struggling, to put together a vaguely correct sentence. Eventually it will come naturally.
• saeranv 4y ago
  Slightly off topic, but am I the only one that dislikes the GUI-based method for adding and reviewing anki cards? Anyone find a decent terminal-based plugin for anki? I've actively looked in the past, and surprisingly there are some partial solutions (can review on terminal but can't add), or hacks, but no complete terminal based SRS...
• kimoz 4y ago
  I also build cards similarly. I also follow the following rules: https://github.com/ninja33/sm-20-rules-cn https://github.com/ninja33/sm-20-rules-cn What I didn't realize is that i can also visualize the cards in a knowledge graph. Thanks for the inspiration.
• dalmo3 4y ago
  Tried Anki for a few months for learning programming concepts and internalizing ideas from books. I created all flashcards based on my own notes, for stuff I understood, did everything by the book. After a while I realised I was visually learning the flashcards, i.e. when reading the prompt side I would see a photographic memory of the answer and get it right, but not necessarily think about the meaning of the contents. After a 15-20 min session I would remember very little of what I had just studied. Then I quit. It reminded me of how trained chess players visualise the game as patterns instead of a collection of pieces. Anyone with a similar experience?
  • rockdiesel 4y ago
    Yes. I had a similar experience. The answer would come to me before I even finished reading the prompt, but I never felt like I was actually learning the info. Just memorizing the pattern. I surmised that it was a result of my deck being too small. It may be different with a larger deck and a greater variety of cards, but I never got my deck to that point.