Skip to content

Latest commit

 

History

History
43 lines (29 loc) · 4.15 KB

antifragile.md

File metadata and controls

43 lines (29 loc) · 4.15 KB

Antifragile: Things That Gain from Disorder

by Nassim Nicholas Taleb

If you like the notes, go ahead and buy the book!

Overview

  • The book is about systems that gain from unknown risks and are getting stronger over time. In short, how to apply "learning from mistakes" in various life domains (finance, health).
  • It is a part of "Incerto" (many ideas overlap with his other books).

What he is criticizing?

  • Taleb is against naive interventionism (not all of it). Often interventionism focuses on less (or none) important aspects (because it is easier) and hides real problems behind the smaller ones/

  • Iatrogenies. He points out that medicine doctors 1-2 centuries ago were too vain to admit mistakes and therefore they were causing much more pain and problems then actually help people. Today that problem is still there. It might produce less mortality and smaller but more hidden health issues. Solution: avoid doctors for small issues, find one in case of a real emergency. This rule, known as the "barbells technique" can be applied to other domains.

  • Economics, especially when ignoring very rare events and their consequences.

What he defends?

  • Taleb verbalizes some of our internal feelings and hunches as heuristics, which can be explained in the context of antifragile systems. He criticizes people who underestimate that kind of approach because this was working for milennia before scientific revolution and shouldn't be ignored (example - religion)
  • Complex systems (organism, health, social systems, nations, politics) are so complex that it's better to admit it and allow/leverage some level of randomness because it's safer in the long term.
    • The whole point is to know how much randomness is good. It's hard but that doesn't mean that we should avoid randomness at all.
    • A lot of Taleb ideas might be actually close to The Long Now project. When examining them in our current context, they might feel very bad and not centered around particular people. I think when approaching this book, we should constantly remind ourselves that most of these ideas apply when you're thinking about systems in a long-term perspective. Personally I think this is exactly the reason why people kind of ignore him (or that kind of approach) and why it so hard to plan in such a way.
  • Intervention by subtraction (via negativa), removing, not adding knowledge
  • That people who present given idea should live up to it (practice it), not only in theory. This can be linked to eating your own dog food.

Other ideas

  • Avoiding decision is also some kind of decision (see Fabian Society)
  • It is important to be open to options, not make decisions to soon
  • Avoid situations, which has a very small probability, but enormous impact when they happen
  • Problems with science - in particular with confirmation bias and taking correlation as implication too often.
  • He thinks that most of the innovations don't come from universities / theory, but from engineers, people who try various things and make a lot of mistakes.
  • Randomness and options could be in some cases replacement strategy for knowledge.
  • Redundancy is sometimes very important

What was confusing for me?

  • Taleb seems not to see that what he assumes as ignorance about antifragile systems can be in fact antifragile from the level above. People right now tend to favor stable (but short-term) solutions, however looking at this from distance, the whole society can learn from their mistakes later, at some point. Current "black swans" (financial crisis, pandemic) might be catastrophic, but we might also learn larger patterns to avoid even greater failures in the future.
  • The whole tone of this book could be more humble. That would help him find more readers and spread really important ideas even further. On the other hand, his arrogant tone might be actually the best strategy for spreading this information.