To learn we must make be willing to make many mistakes. To learn faster we must strive to not make the same mistake twice (particularly in a row).
A few weeks ago I posed a provocative question to a young Vice President at the company I work for. You have this vision for yourself and what you hope to bring to this world. How would you react if you crossed paths with somebody who was head and shoulders beyond you? Someone was doing what you’d hoped to do, who had a greater chance of being successful in the endeavor than you, and who was doing it in a better way than you’d ever imagined. What would you do? There’s a lot to unpack in this question, so I’ll illustrate it with a short example to help communicate the big picture of this idea.
Suppose that it was the early 2000s and your mission was to break our dependence on fossil fuels by developing and popularizing electric cars. You had a vision that you would be the person to accomplish this world changing mission, you spent the last ten years singularly focused on actualizing this dream, and you felt that if anyone could do it, it would be you. Then one day you cross paths with Elon Musk. You hear about everything he’s doing, everything he’s capable of, and his plans that go beyond what you ever imagined to be possible, including sending humans to Mars, and you leave with more confidence in his mission than in your own.
Today I ran into one of those guys.
Max Deutsch
On November 1, 2016 Max Deutsch posted the following learning goals for his Month to Master project.
Max’s project goals
I stumbled across a YouTube video produced by the Wall Street Journal where Max challenges Magnus Carlson to a chess match, which Magnus accepts because he is intrigued by Max’s novel technique (trying to imitate a self written chess AI). Max describes himself as “an extreme learner”, and based on his results which range from performing a set of 40 pull ups to developing perfect pitch in a month. I’d have to agree.
In my final semester of college I had an idea. The idea was to create a place where someone could go to learn anything, and they would be assured success because I would have found a way to learn faster and better through research and experimentation. The idea turned into a vision and the vision was intoxicating. It started to believe it would be my contribution to the world, as well as something I felt I needed to do to reach my fullest potential. I’ve made some progress across the years, but I still don’t see myself as being close or ready to becoming the person my vision.
While digging deeper into Max I found his start-up Openmind, which is an online platform that provides, “mentoring from world-class mentors to help you learn and leverage the skills you need to succeed.” The platform looks impressive. The mentors include start-up CEOs, first rate software developers, and the case studies range from apps with 20k+ downloads to $150,000/yr side hustles. The platform educates its members by connecting them with mentors who have mastered the skills they’re looking to learn. This was precisely my plan for my “learning gym”, which is now only a 3 person club with some demonstration of results, although we don’t leverage as much of a mentor system.
So, ironically, weeks after asking my friend the question, I find myself called to answer it. What will I do when facing someone who appears to be executing my vision better than I am. I’ll use it as a learning opportunity.
What can I learn from Max?
- Extreme learning challenges should be tackled once as a time.
- In my effort to differentiate myself and to quench my polymathic thirst I’ve been pursuing multiple learning goals concurrently for a while. This approach has been both unpleasant and ineffective. I haven’t been able to successfully reach the goals I purse in my target time frames. I’ve been ignoring the shared wisdom of my favorite learning experts, from Cal Newport’s pleas for students not to double major (sorry Cal), through Derek Sivers’ don’t be a donkey advice, to Gary Kellers’ One Thing. I’ve never listened. I’m afraid that there’s too much I want to do, and if I don’t pursue it all at once I’ll miss out on something. Retrospectively I see that most of my successes occurred when I was able to maintain a single focus.
- Meta-Learning is experimentation and should be treated as a scientific process
- To make improvements we take a system that already has some baseline performance and we change one variable. We then watch the system as we let it evolve under the influence of the experimental conditions. At the conclusion of the experiment we reflect on the impact the change made, analyze it, and convert into it wisdom. We use the results of our experiments to guide future experiments and after accumulating a few successes we can start to see the system evolve into something greater or more effective. If we introduce too many changes at once variables become convoluted and we lose our ability to capture wisdom from our experiments.
- Documenting the results success or failure
- As mentioned in the above point, wisdom arises from reflecting on the results of an experiment. If, instead of analyzing and documenting why an experiment failed, we decide to repeat the experiment hoping for different results, we get caught in a destructive anti-learning cycle. On the contrary, when we document the scientific process of our learning experiments we can eventually build up a framework of which approaches are effective and which are not. As this knowledge starts to consolidate we can build a theory from our series of conclusions, something we can capture for ourselves and share with others so the rest of society can benefit from our efforts. This is how the world makes progress. If I had been documenting my experiments, I would have had a system on which to build my “learning gym”. Instead my ideas are so scattered and poorly articulated they might as well not even be there.
- Learning experiments should be short
- As time lines lengthen complexities increase. If you were asked to write a book, you’d probably have a low probability of success, but if I asked you to write a sentence your probability of success would converge to one. The question of how to live a great life is a very daunting one, but if I asked you to have a great hour you’d probably have a better chance of accomplishing that by doing something you know you enjoy doing for an hour. Long time horizons and greater complexities open up systems to more randomness and perturbations. As we mature we are expected to learn how to think successfully over greater time horizons and deal with systems with greater complexity. The way we accomplish this, however, is by taking what we know to be true, what we understand to work in the short term, and repeat it over an extended period. If however, you find yourself not reaching the outcomes you desire, you should shorten your time horizon so you can gain more control and increase your chances of success.
- The purpose of a learning experiment is to learn about learning, and it should serve as a proof of concept. So you heard waking up at 5AM makes you more productive. Maybe you should try waking up at 5AM for a month and observe the impact on your productivity before you commit to becoming an early riser for the rest of your life. We can always extend what works, but short experiments allow us to quickly kill off what doesn’t work and gives us the freedom to perform many experiments instead of committing to a few for long time scales.
- Learning experiments should be fun and badass
- Starting this blog, learning to code, starting a podcast, stepping back on the skateboard, making a website, building an app, and learning how to invest are all examples of awesome things I did in just a month. In a month you can do something awesome that will make that part of your life a little more memorable. A memorable life is a fulfilling life.
What’s next for me? How am I implementing this?
Discovering Max helped me recognize a few of my mistakes and given me something to aspire to. Putting the above insights into action I will be:
- Dropping the parallelization of my learning goals and attempting a discrete focused model.
- Treating my learning experiments as experiments. Setting them up with a scientific demeanor and analyzing and sharing them so I can convert them into wisdom.
- Picking up a fun extreme learning experiment in October that will help me cross off a yearly bucket list item and will inject more enjoyment and badassery into my life.
- Shortening my time horizon and attempting more proof of concept projects.
- Capturing more of my learning insights. Whether I’m going to share them with my club members, the readers of this blog, or my future self, the only way I can develop a novel learning framework is to document individual pieces and add more structure as time goes on.
- Giving Openmind a try. The website is a little buggy right now, but I know exactly what I want to try to learn though his platform. If I were the previously mentioned electric car enthusiast and I met Elon Musk, I know I would want to give the new Roadster a test drive.