For a strong career, you should seek to build career moats. These are hard to replicate skills that keep you in demand. Typically this means being very specialised or having a unique blend of desirable skills.
To keep a reading habit, read multiple books at the same time (light, medium, hard).
Contrast two pieces of advice: You should judge a decision by the process with which it was made, instead of the outcome vs you should judge a decision by its outcome, adjusting your approach until you gain the outcome you desire. They are opposite. Advice reflects the domain a person works in. For example, in poker you shouldn’t judge a decision by its outcome. Because its true you can play the hand right and still lose. But in other domains, this isn’t true. If you make a decision as a manager and then half your team leaves, that is a poor decision. You should judge it as so, based on the outcome. You need to understand if the domain is regular or irregular. Regular domains are sufficiently predictable, with observable cues and have opportunities to learn from cues. Poker fails on the first. Stock picking fails on both. But many other areas pass on both (managing people, coding decision). So before accepting such advice, consider if the domain fits which use case (The Dangers of Treating Ideas from Finance as Generalised Self Help).
On Mental Models
There is a difference between tacit knowledge and explicit knowledge. Explicit knowledge can be readily articulated or codified (like a fact). Tacit knowledge is that that is difficult to pass onto someone else - its a feeling and dependent on personal experience and context (for example, the feeling of a great tennis stroke). The problems with mental models is that some are more tacit than explicit. So even if you learn all of Munger and Buffet’s mental models, you may still lack the experience to make them worthwhile. Hence, both the knowledge and the practice is important. This is true in many disciplines (Mental Model Fallacy).
How can we make mental models useful? How can we practice them? First, we must know how to evaluate tacit knowledge in general. Two main principles here: 1) let reality be your teacher (try it and see - self experiments) and 2) listen to believable people (people who have done it) (Framework for Mental Models). So his recommended process it to practice something in reality -> try to generalise what works into mental models (a la Ray Dalio in Principles). This is different to the Farnam Street approach which is the reverse, learn mental models —> try to apply them in reality. A list of mental models can be useful to put existing explanations or patterns to what you may have experienced in your practice.
Rationality can be defined (loosely) as the effectiveness of ones thinking. Hence someone can have a high IQ (intelligence quotient) but a low RQ (rationality quotient). It also shows why smart people can be assholes and very effective people can have average intelligence. There are generally two types of rationality. Epistemic rationality answers how do you know what you believe is true. For example, I believe Facebook is undervalued. Instrumental rationality answers how do you make better decisions to achieve goals. For example, what should I study, CS or philosophy? So mental models fall into three categories: descriptive (some knowledge about the world), epistemic (finding the truth), and instrumental (deciding better). Epistemic rationality is often about removing biases. But in some fields epistemic (being right) matters more or less than instrumental (deciding A vs B). It’s easier to practice epistemic rationality (learn things) than it is to practice instrumental rationality (how do you know if a decision was good?). If a field has positive convexity (costs of trying are low and potential upside high) then you can trial and error.
A framework for better trial and error may be Baron’s search-inference model. It has three parts. You set out to find possibilities (possible answers to a question). You then have evaluation criteria (goals by which you evaluate the options). Then evidence are the beliefs that help you determine if a possibility will help you achieve a goal. By putting this together, we can make inferences about which is best or worst. There are 3 typical was we can think badly:
- We don’t properly search for something we should have discovered. For example, we don’t give ourself enough options. Or we don’t have all the evaluation criteria in our mind.
- If we make poor inferences, caused by some bias. For example, we just choose something that confirms our existing beliefs.
- Or you think too much. The amount of thought should be sized with the importance of decision.
You can then evaluate mental models by these criteria. For example, Inversion helps with one and two. It can give new options and helps us properly infer some away. Expected utility theory is also a useful approach. Given a set of choices you put a guess of your utility in each outcome, and then you can Some an expected utility. You are rational if you choose the one that is highest. Whilst useful, it’s unlikely you’ll do this given every decision you have to make and the sum is always changing as new information is presented of the world changes.
On Career Moats
A career moat is you ability to build and maintain competitiveness in the job market (Career Moat). You know you’ve achieved this when you don’t worry about finding work (Career Moats 101). It is similar to Cal Newports idea of career capital. This should be the primary goal early in your career as it gives you more freedom thereafter.
To achieve career moats, you need rare and valuable skills. Generally, this comes in two forms. Something really specific. Or a mixture of skills in combination. Three strategies may be:
- Choose a mixture of skills where the path is opaque. Difficult to replicate.
- Choose a skillset that is valuable but not attractive e.g. COBOL cowboys.
- Choose a skillset before it is proven valuable and ride the wave e.g. AR/VR or Bitcoin.
You then need to be on the lookout for changes in the market that could disrupt your moat (e.g. is COBOL going out of fashion) (Career Moats 101). A useful question to ask yourself with each career move, will I learn rare and valuable skills? (Rare & Valuable).
One tactic, is to create synergy between your work and your side hustle. For example, if you are learning Python in your day job, you can create a side blog about learning Python. Each re-enforces the other, deepening the moat (Career Moat Patterns - Tie a Good Thing to a Better Thing).
If you have a portfolio approach to work, make sure they have synergies on the upside. Do they re-enforce each other? Does doing one make the other easier? (Portfolio Thinking).
Career moats help you find your next job. This doesn’t make you recession-proof, but it doesn’t make it easier to get back up. You career moat value changes in relation to the market. If your career moat is growing remote dev teams, now is probably a great time.if it’s designing restaurants, then it’s a bad time. It’s useful to think about complementary skills to your current set. What else would a future employer find useful? Is it rare an valuable? If yes, do that. Perhaps one way to be recession-proof is to attach yourself to the profit centre (Drucker’s parlance) - make sure you’re the money maker in the company. (Career Moats in a Recession)
Categorise books into trees (frameworks), branches (specific topic) and narrative (historical account). Each can be read with a varying degree of effort and style (3 Kinds of Non-fiction).
Read narrative versions of books (Undoing Project) before reading the core idea (Thinking Fast and Slow). This makes them easier to read as you move up the learning arc (Land & Expand of Reading).