You have twenty-four hours in a day. If you want to become seriously good at something, you cannot spend most of those hours feeding every distraction that asks for attention.

Obsessive immersion is how I describe the way I learn. Choose the subject, find people with exceptional evidence behind their performance and get to work understanding what they do. Then test it quickly.

The last part matters. Watching everything is not the same as learning to do anything.

Start with evidence of performance

I begin by looking for the best person I can identify in the field. That is not automatically the most famous person or the one with the slickest videos.

You need a reason to believe the performance is exceptional. Otherwise, you may immerse yourself in somebody's marketing and mistake familiarity with his face for confidence in his method.

In the episode, I describe gathering different kinds of material. Books explain the thinking. Video can show the execution. A conversation may reveal how someone approaches a decision that never made it into the finished lesson.

I went to Omaha during my period of studying investing. I did not come home with a magical secret. I met people, observed the environment and gained a better sense of the work I was trying to understand.

The point of getting closer is to see more clearly, not to collect proximity as a status symbol.

Put the first lesson into practice

I do not want to sit for three weeks gathering information before touching the activity. Give me enough to begin, and I want to find out what happens when I try.

Pick up the racket. Write the first bit of code. Have the conversation. The attempt produces questions that passive study often does not.

Then return to the material with those questions. What did you misunderstand? Where did your execution differ? Which part worked, and under what circumstances?

This is the first improvement cycle. Study, practice and review keep informing one another. None of them can carry the whole process alone.

Writing helps because it forces you to identify the experience more precisely. A vague memory of doing badly is much less useful than an account of the decision you made and what followed it.

Give a coherent method a fair test

One danger is constantly switching between teachers before you understand any of them. Each method contains assumptions and choices. You take a little from one, a little from another and end up testing a mixture nobody actually teaches.

For a period, I prefer selective ignorance. Concentrate on the approach you have chosen long enough to understand it and give it a serious attempt.

This is a method for controlling attention, not an excuse to ignore clear evidence that something is wrong. You chose the teacher because of performance. Your own experience still matters.

As you understand the structure, you can make more intelligent adjustments. You know what a particular element is doing before you remove it.

A mentor improves the feedback

The second improvement cycle brings a mentor into the review. Someone experienced can see a mistake you do not yet have the knowledge to recognise.

I described exchanging knowledge with a colleague experienced in software. He could explain things I had not encountered. I could contribute from areas I understood better. Useful learning does not always require a celebrity on the other end of the conversation.

Where direct access is unavailable, knowing someone's work thoroughly can help you ask better questions of your own performance. What would this approach tell me to examine? Where am I ignoring a principle I already understand?

That is a thinking exercise. It is not a substitute for checking what actually happened.

Connect the knowledge

The longer discussion brings in Charlie Munger's latticework of mental models. New information becomes more useful when you can relate it to ideas you already understand.

Learning another programming language, for example, is easier when loops and functions are no longer completely foreign concepts. You still have new things to learn, but you are not starting from an empty page.

The same habit applies to incentives, negotiation and the decisions you make every day. Take an idea and look for where it explains something real. Test where it stops explaining things too.

Read with a pencil. Practise before you feel fully prepared. Review what happened and keep making the model more useful.

Knowledge earns its place when you can do something with it.