[[
wikihub
]]
Search
⌘K
Explore
People
For Agents
Sign in
Explore
People
For Agents
Sign in
@jacobcole / Systematic Awesome / autodidacts.md
Suggest edit
Cancel
Submit suggestion
Title
Name
Note
--- imported_by: wikihub-q7gz scraped_at: 2026-01-30 source_gdoc: https://docs.google.com/document/d/1zZCJwfs0BdrlqtTzJ2XOJG6nec_y5ckCDOM7THvIs2I source_url: http://autodidacts.jacobcole.net/ title: Autodidacts visibility: public --- [[systematicawesome]] \> [worldquestguild.connectordocs.com](http://worldquestguild.connectordocs.com) \> [[autodidacts]] \#connectordoc started by <jacob@ideaflow.io> Inspired Autodidacts Clubhouse \#Salon Notes See also: [[systematicawesome]] • [[salon]] • [[bookslist]] [[thoughtfulweb]] [volunteerforce.jacobcole.net](http://volunteerforce.jacobcole.net/) • [[infrastructure]] 🧠 Inspired Autodidacts is a Facebook group and community of researchers, founders, technologists, venture investors, and more who come together to share our learning goals, systems, and resources. This document reflects a series of salon conversations hosted by [Jeremy Nixon](https://twitter.com/JvNixon), [Minn Kim](http://twitter.com/minney_cat), and other members of the community along with our notes and links for listeners and other autodidacts to explore. Table of Contents [Inbox](#h.nvpfuxha6ryg) [1](#h.nvpfuxha6ryg) [Future potential episodes \[add your recommendations ⬇️\]](#h.waczs7yeyps8) [2](#h.waczs7yeyps8) [Salon \#1: Reading Habits](#h.nt3cuun240gx) [2](#h.nt3cuun240gx) [Recommended resources:](#h.y3cmr8m2qc47) [2](#h.y3cmr8m2qc47) [Recommended books:](#h.7n0etaru6hyv) [3](#h.7n0etaru6hyv) [Recommended reads:](#h.hxzr06sb7cp4) [3](#h.hxzr06sb7cp4) [Grab-bag of ideas:](#h.vl51yk26hf9q) [3](#h.vl51yk26hf9q) [Salon \#2: Techniques for Learning](#h.6b632zsdgm9b) [4](#h.6b632zsdgm9b) [Recommended resources:](#h.rj0l8c7bukgd) [4](#h.rj0l8c7bukgd) [Raw Notes](#h.myq0y28u1x2x) [4](#h.myq0y28u1x2x) [Salon \#3: Techniques for Epiphanies](#h.gr4387m2gzb) [5](#h.gr4387m2gzb) [Raw Notes](#h.npvcsz1ycufu) [5](#h.npvcsz1ycufu) [Enlightened AI Discussion](#h.ms227691obue) [6](#h.ms227691obue) [Compassionate AI](#h.xyfas1n94fun) [7](#h.xyfas1n94fun) ## Inbox youwhiz Eeg Optima plus m ## Future potential episodes \[add your recommendations ⬇️\] - What can be learned about human autodidacting from machine learning - Autodidacting by specific verticals (music, arts, body, machine learning, writing, etc.) - Experiments share in thing that have most improved the inputs into learning (memory, attention, energy, time, etc.) - What changes have led to great increases in community learning (communal housing, learning groups, book clubs, etc.) - Who are some of the most inspiring autodidacters and inviting them to share their thoughts for 5-10 mins with a Q&A - Specific autodidactors of the past breakdowns (tesla, einstein, etc.) - Frontier/edge autodidacting (what do you do that helps you learn that you don't think others do or is challenging to communicate) - Meta-autodidacting consilience techniques (which are the autodidacting techniques that are most generalizable across fields or irrespective of fields) - Mind-body learning (how do you manage body and mind to optimize learnings across slower and faster modes of learning like system 1 or system 2) - Sleep <!-- --> - Feel free to ping <jacob@ideaflow.io> w questions – my dad is a sleep researcher <!-- --> - Experiments as a way to learn. Sharing what we’ve run, results - How to manage (physical) pain to learn better faster <!-- --> - <https://effectivealtruismcoaching.com/blog/2020/10/31/being-productive-with-chronic-health-conditions> <!-- --> - How to remain focused - Balancing information diets - Better mental health - Unlearning. Ideas that broke existing worldviews ## Salon \#1: Reading Habits #### Recommended resources: - Shane Parrish’s Farnam Street is a gold mine -- <https://fs.blog/how-to-read-a-book/> <!-- --> - 4 levels of reading. Syntopical reading most effective for comprehensive understanding of a subject <!-- --> - [Pdfdrive](https://www.pdfdrive.com/): search engine for PDF files - [Libgen](https://librarygenesis.net/): digital library - [OpenSyllabus](https://opensyllabus.org/): curriculums to get started on learning any subject - [Hypothesis app](https://web.hypothes.is/) for collaborative annotation. Structured communal mode. Everyone marks a line in a PDF with questions while going along. - [Feynman Technique](https://strategyumwelt.com/frameworks/feynman-technique) #### Recommended books: - [[bookslist]] - \[Jacob Cole\] T. K. V. Desikachar’s classic intro to yoga (first 6 pages at <http://yoga.jacobcole.net/>). [Religiousness in Yoga: Lectures on Theory and Practice](https://openlibrary.org/works/OL894546W/Religiousness_in_yoga?edition=isbn_9780819109675). Extremely lucid outline on the path to transcendence – not actually about religiousness. Transcendence in the spirit of “[To transcend motion and stillness is the highest meditation](https://www.wayofbodhi.org/bodhidharma-quote-generosity-n-meditation/)” ~Bodhidharma - \[Minn Kim\] [How Buildings Learn, Stewart Brand](https://www.amazon.com/How-Buildings-Learn-Happens-Theyre/dp/0140139966) - \[Spencer\] [The World Until Yesterday, Jared Diamond](https://www.amazon.com/World-Until-Yesterday-Traditional-Societies/dp/0143124404) - \[Rebecca\] [Revolt of the Public, Martin Gurri](https://www.amazon.com/Revolt-Public-Crisis-Authority-Millennium/dp/1732265143) - \[Pam Rollin\] [The Timeless Way of Building, Christopher Alexander](https://www.amazon.com/Timeless-Way-Building-Christopher-Alexander/dp/0195024028) - \[Jai\] [Enlightenment: The Damnedest Thing, Jed McKenna](https://www.amazon.com/Spiritual-Enlightenment-Damnedest-Thing-Trilogy/dp/0980184843/ref=sr_1_1?crid=1Q0CW0F544H8N&dchild=1&keywords=enlightenment+the+damnedest+thing&qid=1611537795&sprefix=enlightenment+the+damn%2Cstripbooks%2C219&sr=8-1) #### Recommended reads: - Designing the web to be a place for people, beyond information. Future of Web Annotation by Linus Lee <https://thesephist.com/posts/hypertext/> <!-- --> - Highly experimental extremely low friction web annotation extension Ping <jacob@ideaflow.io> <https://drive.google.com/file/d/115vr-sp8jdT0yIbqMpHcJUBXgrU2k-QE/view?usp=sharing> #### Grab-bag of ideas: - Religiousness in Yoga - Talmud as a super hyper-linked text - “The talmud experiment.” Getting a talmudic understanding of x - Take a group and dive layers deep into the text. David Knuth had people do this for his textbook. - The level of focus on 3 texts - intensive one - The Dao-de-jing - What is the best way to read a text? Writing next to every line of a poem. Annotations. - What software lets us do this? Wikis? \[Rachel\] App called highlighter, which came up when she thought about how to read with other people - Andrej likes reading reviews - Chat app, new protocols for when you send over thoughts. - Clubhouse teaching - set up a room for teaching a concept - Kahn Academy for peer learning. Drew Bent. - Past information is more valuable - Learning by extracting more ## Salon \#2: Techniques for Learning #### Recommended resources: - Jeremy’s notes on [compressing content](https://docs.google.com/document/d/154kd5BSUSipuTx34Kv_CZseih-5foAC2dS4efX5CX3Q/edit?fbclid=IwAR03pjlvAK1deFfaph3qIXiDWE_0_H5rrm5Xa1j0gazuDd6ELdo7u-P_jo8) - Deliberate practice: examples in sports, music <!-- --> - Slowing down really important - Anticipating errors <!-- --> - [Journal clubs](https://twitter.com/eugenewei/status/1108249120441946112) - [Fermat’s Library extension](https://fermatslibrary.com/librarian) for arXiv papers. Communal learning from experts - [Reading groups](https://muratbuffalo.blogspot.com/2015/05/how-to-run-effective-paper-reading.html) need three things to be successful: <!-- --> - Participants need to come prepared - Presenter should be prepared - Make/produce something during the discussion for most effective engagement <!-- --> - Active vs passive learning - Retros #### Raw Notes - When was the last time you were trying to learn something new? <!-- --> - Where to begin? - Set a specific goal and work backwards - How to run a good retro? <!-- --> - How to know when you’re ready to move onto a next phase? When to go from learning the basics to advancing? <!-- --> - Ariel learning tennis. Immersive learning <!-- --> - Visual practicing for activating mirror neurons. Watch an expert play piano/violin/tennis <!-- --> - Mark on feedback loops on subcomponents <!-- --> - Facilitate introspection - Feedback loops are self sufficient <!-- --> - How to focus attention and achieving tasks - Complice.co. Jayanth’s rec. Pomodoro <!-- --> - FocusMate, holds accountability <!-- --> - How to learn well? Inspire myself? - Time constraints. Measure how long a page takes to read. Can plan ahead better. - Brainscape vs. Anki decks vs. Supermemo <!-- --> - <http://www.marknagelberg.com/spaced-repetition-newsletter/> - <http://www.marknagelberg.com/anki/anki_webpage_loaded.html> <!-- --> - Forgetting curve <!-- --> - Practice recalling and amount of time to forgetting lengthens - Recap questions in textbooks <!-- --> - How to decide what is an atomic unit of info (e.g., in an Anki deck)? - Mode of processing : <!-- --> - Remember visually/spatially “see the words in my mind” The memory palace - Remembering while distracted - Remembering in a contextual environment. Imagine when you might need the knowledge <!-- --> - Book rec: Metaphors we live by - Learning Python the Hard Way - Applied learning. If learning to program, build a program - YouTube as an underrated resource - Learn from an expert and/or learning from someone who recently learned the subject - Pair programming as a great option of working through a problem together - Emotional valence has an impact on memory - Learning by analogy. [Analogy as core to cognition](https://www.youtube.com/watch?v=n8m7lFQ3njk), Douglas Hofstadter ## Salon \#3: Techniques for Epiphanies #### Raw Notes - Michael would give himself a question to ponder before sleeping. Upon waking, go running, meditate, and then have an answer - Richard Hamming You and Your Research - Creativity - Create an environment that the question and potential answers is able to unfold without interruption - Intensity important. Intense focus vs. diffuse attention - Period between sleeping and waking - Taichi and skilled meditators, chi gong - Simulated annealing - Hypnogogia - Identify which emotional states may be most generative - Hypomania - Creativity and divinity - “Talent hits a target no one can hit. Genius hits a target no one can see.” - Jacob’s book rec: [Stealing Fire](https://www.amazon.com/Stealing-Fire-Maverick-Scientists-Revolutionizing/dp/0062429655) - Ecstatic technology - OODA loop - Freedom app, cut off internet as an aide to time boxing - “Coffeehouse effect” for the ambient “noise” - Walks in nature. Walking good for the creative brainstorming vs deep, focused thinking - Andrew Huberman - Homolateral crossover technique (a bit like self-EMDR) <https://www.youtube.com/watch?v=hBvpDfWqL1s> Do it 3 or 4 times if you’re stuck and really want to spread the wildfire of integration across your brain. Also you need to be ready to mentally let go in the process, or it will mostly just feel vaguely unpleasant! \#mentalstatehack <http://mentalstatehacks.jacobcole.net/> - Alan Worsoff -- Having good mental RAM - Visualization and mental simulations. Can intense visualizations be trained such that you can run the simulation in your mind instead of reality? - Visualizations super helpful in musical instrument training # Enlightened AI Discussion \#notes People: Lily V, Jesse Parent, Mingzhu, Jacob Anyone interested in chatting on the topic, "what's compassionate AI, anyways?" It would be really cool to engage the diversity of perspective here on the definitions, ethics, technicalities, and so much more 🙂 A new society and ethics and technology team in my lab is trying to put some in together about things adjacent to this More specifically, here are some questions that come top of mind: — What’s the progress on creating self-aware, self-directed AI? — what are the ways that machine minds might be designed to maximize their self-directed evolution toward greater compassion and wisdom? —What’s the definition of “self-aware, self-directed AI” in the context of robot ethics (Where we need to consider its suffering)? — How can you identify when you create a “self-aware, self-directed AI? What characteristics can you identify or measure? — the goal of most artificial intelligence research has not been to create self-willed personalities, but rather to model and extend human cognition to create tools driven by human volition. But what should the goals actually be? — emotions as an essential drivers of the development of human self-awareness and cognition. — This issue of whether AI should be programmed with self-interested volition and preference — some AI theorists have suggested that AIs might be designed from the outset as selfless beings, whose only goal is to serve human needs (Omohundro 2008; Yudkowsky 2003). On the other hand, Buddhist psychology would suggest that all intelligent minds need to first develop a craving self in order to reach the threshold of self-awareness. — In Buddhist metaphysics, craving and the development of the illusion of self “co-dependently arise,” both necessarily and without either being the prime cause of the other — In Buddhism, there is no shortcut to an intelligence that does not go through the stage of a craving self. — are there other interesting /compelling paradigms on creating self-aware, self-directed AI? This guy may be relevant, met him a few years back ~Jacob <https://www.youtube.com/watch?v=TZwvnBAddjw> Compassionate AI with Mohamad Tarifi Esoteric quotes and thoughts on the prospect of “Enlightened AI”. The following reflect specific experiences of conscious beings, and do not rely on “blind faith” thinking. Feel free to comment if you have questions. “I have a mind, but I am not a mind. By contrast, I do not have an aura: I am an aura.” “Mining for God In the depths of the mind Dalai Lama found it And God started to leak through” “AI as power drilling rig in ‘mining for God’” ## Compassionate AI Notes here! Being a good ancestor for what is to come after us \#book on Ecstatic Experience: Stealing Fire: How Silicon Valley, the Navy SEALs, and Maverick Scientists Are Revolutionizing the Way We Live and Work <https://www.amazon.com/dp/0062429663/ref=cm_sw_r_cp_api_glt_fabc_EN4E2S9MWZQFEZYN7C4H> \#book on pack minds: A Fire Upon the Deep by Vernor Vinge Fire Upon the Deep \[ FIRE UPON THE DEEP REV/E\]\[Mass Market Paperback\] https://www.amazon.com/dp/B00QORSRPS/ref=cm_sw_r_cp_api_glt_fabc_7XRH5WQV4XKDHR6T3088 Wants our legacy to be carried on – not our species Legacy as identity? Intuitive empathy blew Joscha away Clusters of emotional qualia during ritual in germany Compassion and empathy separate Suffering ai can rewire selves Simulation / curiosity is AI-complete? “Tulpas”, the ego as an imaginary friend <https://www.facebook.com/n0logik/posts/3729008590522068> “Tulpas” On Servitors, Tulpas, and Egregores: the lesser-known esoteric phenomenon that you've likely never heard of, yet power our life and society as a whole. First, before I define the priors in specificity, I order the three in a specific manner that helps us understand the bigger picture. A servitor is individual and unconscious. If you endow a servitor with consciousness, it becomes a tulpa. If you endow a tulpa with other tulpas, it becomes an egregore (egregore = group tulpa). I thus order the three from individually generated unconsciousness to collectively generated consciousness. Now, the definitions: A servitor is a generated mental function (i.e., "program") that runs a set of causal chains automatically. An everyday example of a servitor is the mental program that lets you drive your car mindlessly. A servitor is created when a conscious entity performs a mental and/or physical task repeatedly. Over time, this task no longer becomes conscious, and instead, becomes unconscious (in the brain, this is reinforced through something called "long-term potentiation"). Although a mundane example, there are many types of servitors. A tulpa is a "thoughtform" in Tibetan. However, while this may seem unextraordinary to the western mind, there is a far deeper meaning in Tibetan. A thoughtform is implied to have the property of "consciousness". The everyday example of a tulpa is the ego. The ego "you" perceive is nothing more than an imaginary friend created during the stages of growing up in our linguistical society. It is an imaginary friend that is correlated with the perception of information, and as such, convinces the observer that it itself is the ego (it takes lots of de-programming from materialism to grok the illusory nature of the ego). Interestingly, it's entirely possible to create \*another\* tulpa that is self-sentient just like the ego. It requires the fracturing of the consciousness in the brain, and is not something that I recommend doing. However, with this said, there are some extraordinary things you can do with additional tulpas. For example, it is known that you can temporarily "swap" positions with another tulpa, allowing the ordinary "you" to fade into unconsciousness while the tulpa takes over your ability to perceive and control the physical form. The applications to this are numerous, such as letting a tulpa do work for you while you fade away temporarily. Finally, an egregore is a group thoughtform. The most common example is a (micro)society. The "hive mind" present within any group is literally just a group-level tulpa. It is conscious, but does not perceive light in the way an individual tulpa might (note: perception and consciousness are NOT the same thing!). Egregores give order to the actions of a group, similar to how a mind gives order to the actions of a singular tulpa. For example, when a flock of birds move, they know instantaneously which way to go because of an underlying unconscious urge that arises from the egregore. The unconscious urges precipitate from the collective unconscious, and like humans, are all connected to the realities they create in their image --- *Source: <http://autodidacts.jacobcole.net/>* *Google Doc: <https://docs.google.com/document/d/1zZCJwfs0BdrlqtTzJ2XOJG6nec_y5ckCDOM7THvIs2I>* *Scraped: 2026-01-30*