SuperMemo

2021-11-27 links: Intelligence Learning reference: https://supermemo.guru/wiki/School_dropouts https://supermemo.guru/wiki/War_of_the_networks https://supermemo.guru/wiki/Knowledge_valuation_network https://supermemo.guru/wiki/Planning_a_perfect_productive_day_without_stress

• YT - Experimental Learning

SuperMemo #

• 2,054 golden pages written by a Win Wenger kind of guy, except it’s way more rigorous and just interesting in general.
• https://www.masterhowtolearn.com/

Incremental Reading #

• Basically you copy+paste articles into supermemo. Highlight + add creates a sub-file to the article, and you can make a cloze card for SRS. It’s quite similar to LingQ.

  • Cards can also be hierarchical; you can go down hyperlinks and the most downstream card will be reviewed/learned first. This is powerful and I don’t think there’s anything like it in Anki! R See here: 2.1 - Formulation - Fowl’s Guide to Supermemo - this guy has a serious workflow going.

• Does it make you autistic? Are the geniuses of history fighting against the drive for stasis and ‘knowledge darwinianism’, or are they skilled at retaining information? Is this counter to creativity? Complementary?

  • Perhaps supermemo-style learning is simply a method to reach maximal time efficiency, and without it, the emerging knowledgebase is heavily pruned, yes, but perhaps it’s no more pruned than a supermemo SRS database, and is simply smaller due to time wasted. Is this ‘wasting’ of time ever beneficial?
    • Perhaps massed repetition style reading generates/requires a heuristic algorithm that contextualizes things in a certain personal way; copy+pasting stuff necessarily makes you soak up information ad verbatim, rather than putting it in your own words.
      • But at some point, you have to wonder if that’s even valuable, especially for things like rigorous mathematics. There’s only so many ways to word highly formalized things. It’s probably merely the cominbation building on top of it (jenga tower; having a solid foundation) and being able to abstract outward (creativity?) that creates ‘competency’. Maybe any other claim is a meme.
        • Being interested is also a major factor; having that fire in your eyes! Soaking up information like a robot with SRS may fundamentally be no different, but it’s the enthusiasm to use that information and its presence in your subconscious mindsoup that counts.
        • We treat our knowledge as our own. Maybe the totality is, and maybe the whole is greater than the sum of its parts, but the source of our knowledge is not exactly our own. In the lexical paradigm, there’s only so many configurations of semantically identical bits of information.
  • Skepticisms have been addressed - Harm of inremental reading
    • IR (reading shuffled snippets from various sources) is not suitable for research papers, programming, or material that requires large amounts of working memory.
    • “Texts that are too complex have low learntropy”;
      • Learntropy: The attractiveness of something to the learn drive. The evaluation of learntropy occurs via the knowledge valuation network.
    • Disadvantages of incremental reading
      • The payoffs of incremental reading is most apparent with chaotic trawls of text, rather than very well-written and cohesive books like, say, Feynman’s.
      • Knowledge Darwinianism
        • I took note of this because one would think SRS would artificially reduce this, but it’s actually refers to semantically equivalent concepts found in different contexts, and the selection for which is the best fit in your mind.

Precocity Paradox #

• The longer the brain growth, the better the ultimate outcome. This can manifest as ADHD, delays in speech/numeracy/literacy, delayed peak of cortical thickness even after 12 years.
• *Exclusive focus on semantic learning may determine the brain arhitecture due to non-interference in the conceptualization process.
  • (Asemantic learning is cramming, when there is a weak or inconsistent framework of connections; arbitrary things, like the days of the week or months of the year.)
• In healthy development, cognitive metrics of certain skills may show a sudden exponential explosion.
• Evolution of attitudes towards precocity
  • Prior to 1930 experts believed that precocious children were losers - the (wrong) idea that they weren’t inititally ran against the grain.
• Fast thinking (as in “thinking fast and slow”; stability of established connections) is determined by pruning and myelination and is slow to develop. First pruning, then myelination. This is a metaphor for how to learn, and how school fucks you - it makes you try to solidify things too quickly rather than explore what you want. Dual-process model of white matter development - Development of white matter and reading skills

How to Become a Genius #

• Part of a long article written earlier
• Pleasure of communication/learning/knowing.
• Obvious:
  • No stress
  • Sleep
  • Always be learning
  • Self-discipline

Optimization of behavioral spaces in development #

20 rules of knowledge formulation #

• Apparently adapted from Knowledge structuring and representation in learning based on active recall

  • Part of his 1994 doctoral thesis which is actually insane: Economics of learning - New aspects in designing modern computer aided self-instruction systems which has a whole chapter on the molecular basis going into NMDA, calpain, epigenetics. Nothing insane but yeah.
    • Reverse cards are always recommended
  • Planned redundancy as a way to cross-strengthening synaptic patterns
    • The function of redundancy is here exclusively to promote the establishment of additional synaptic patterns serving as emergency access routes to the remembered knowledge

• In summary, it’s mostly obvious stuff. But I wonder if pictures on the front inhibits recall in everyday circumstances, or if it helps.

• Understand before you learn before you memorize. I mean, easy enough to understand.

  • Technically this is contrary to Bloom’s Taxonomy but they’re about different things; this is all providing the framework for ‘memorization’ at the bottom to even begin the ascent to creation.

• We want answer to be as short as imaginably possible!

• Avoid lists, sets, and enumerations. I want to read about this one because it’s often these type of things we’re tested for - but maybe it’s the fact that it’s useless knowledge, as opposed to imply knowing the things enumerated themselves and personally relating them together yourself, making an abstraction around them.

  • Like “what countries belong to the EU? A: Austria, Belgium, etc.” - instead have 10 cards like “What country joined the EEC in 1981? A. Greece”

• Keep memories simple; 1-1 connections. Save complexity for something outside of SRS. Not sure how much I agree with this one for the sake of building understanding - but the justification is obvious if the main priortity is just memorizing facts.

17. Redundancy does not contradict minimum information principle; may even be welcome.
18. Provide sources! Is this applicable to anki necessarily?
    • Sources should accompany your items but should not be part of the learned knowledge (unless it is critical for you to be able to recall the source whenever asked). - this has been my intuition thus far also.

I would never send my kids to school (2017) #

• Also Problem of schooling, in the modular/non-printable format A lot of the zettelkasten trawl of SuperMemo is composed of chapters of this page.

Sleep #

Sleep