Accelerated learning
Objective
In this project I look to develop a learning framework to be able to get the most out of my one month projects.
Steps
Define why and how much time I'll allocate to the project.
Define the facts, concepts and/or procedures that I need to learn:
Reduce, interview, reverse and translate them.
Apply Pareto law to them.
Sequence them to avoid failure points.
Create one-pagers.
Method
Metalearning
Every learning project should begin with research. The goal is to gather most (if not all) the material you'll need to succeed.
First of all, I should answer why I'm trying to learn the subject/skill/whatever, which would give me an perspective to take (for instance, if I want to learn a programming language to get a new job, I don't need to learn everything about the language, but only what's most used in that job. Conversely if I want to learn it just for the sake of it, the sky's pretty much the limit). Also, I should try to define my the timeline. This would be dependent on a few factors:
The average time it takes people to get to the level I want to get at (I should aim for something lower, but still realistic)
How much time I'm willing to spend on it (related to the previous one, my goal should be in line with this)
Once this has been defined, I should divide the time I'm allocating for the project in 8, and map it in the following image, as to get an idea of when my milestones and probable setbacks will happen.
After that, the next step should be to answer what I'm going to learn. This is where most of the research takes place.
In general, skills will be compromised of the following three categories, leaning more into one or two of them:
Facts: which are simply things that need to be memorized (vocabulary when learning a language, for example)
Concepts: things that need to be understood (math and programming)
Procedures: things that need to be practiced (which is where drilling would come into practice).
According to which one of them is the predominant one in the project should define which methods should be used (spaced repetetion for fact-leaning projects, etc).
This will probably yield more things to learn that needed, so a weeding-out stage is needed. To do this I need to:
Reduce: find underlying principles in what needs to be learned, and focus on that
Interview: ask people who already achieved what you want to achieve how they did so. Read biographies/listen interviews if no people are available.
Reverse: sometimes, the best way to do something is the opposite of what's normally done. For instance, when learning chess it's better to start by learning endgame instead of openings.
Translate: find what Tim Ferris calls "auxiliary verbs". Basically they're things that allow you to "do a lot with not much". For instance, in jazz, if you learn that scale-off-of-chords thing, you can improvise without knowing everything you'd need normally.
Of course a good way to end this stage is to apply the Pareto law, identifying the 20% that would yield the 80% of the results.
Once this list has been built, it's necessary to sequence them. This is basically coming up with an order that diminishes failure points. Think about gymnastics progressions here, that's what you're after (walking before running and all that)
Reversal
Sometimes, you'd need to apply direct-then-drill, which would imply running to see what you're missing when walking.
Approaches
Directness
When learning, practice should be as direct as possible. By direct, Young means that it should be as close to the real thing as possible. There are 4 ways to achieve this:
Project-based learning: try to organize learning around producing something (if trying to learn a new programming language center that around creating a program in that language, if trying to learn a new musical concept center that around creating a song using that concept)
Immersive learning: surround yourself with the target environment in which the skill is practiced (an OSS project in programming, a place where that language is spoken if trying to learn a natural language, jazz jams if trying to learn jazz)
The Flight Simulator method: if you can't practice the thing you want to learn as-is, create a simulation of said environment (example?)
The Overkill method: (what's up with this weird names) Basically you do something way more difficult than what you want to learn (for poetry it could be writing poems to send to a publication, for linux learning taking TLF's exams, etc.)
There's a certain disparity between this principle and the Drill one. This is remedied by Young with his Direct-then-Drill approach: you practice the skill as you intend to execute it once mastered, look for flaws (focus specially on rate-determining flaws, i.e. things that would yield the biggest ROI if mastered), and drill those.
Drilling
I guess this is where most of my hours should go. Here is where finding those rate-determining steps come into play.
Some methods:
Time slicing: break the skill into 'moments' and practice those (for instance, when learning a musical piece, slice it into bars and practice one each time, when practicing a kata slice it into parts and practice each one separately)
Cognitive components: it's pretty much the same as the one above, but instead of slicing time, you slice concepts (like breaking down gymnastics skills into progressions?)
Copycat: You find a piece, or something already made, and copy everything BUT what you're expecting to drill (for instance, if practicing color theory, you could grab a lineart and paint that, not having to focus on composing, drawing, etc; if learning front end dev copy a page already implemented so you don't have to design, choose colors, etc)
Magnifying glass: spend a disproportionate amount of time on what you want to drill than on the rest of the task (if playing the piano, spend more time trying to make time perfect, than in digitation/analysis/etc)
Prerequisite chaining: You start trying to do something you don't have the skills for, fail, take a step back, learn one of the prerequisites, then try again, and so on.
Feedback
Skills
Focus
Retrieval
Temporary notes
References
Tim Ferris, The 4 Hour Chef: The Simple Path to Cooking Like a Pro, Learning Anything, and Living the Good Life, 2012. Up to The Domestic chapter
Scott H. Young, Ultralearning: Master Hard Skills, Outsmart the Competition, and Accelerate Your Career, 2019.
Last updated
