Antifragility
Antifragility
Antifragility is a term coined by Nassim Taleb in his book Antifragile: Things That Gain from Disorder
Antifragility -
- the ability to not only withstand stressors and shocks but to improve and thrive under them.
- the category of things that not only gain from chaos but need chaos or use (leverage) chaos in order to survive and flourish.
- It is the phenomena that describe things that are the opposite of fragile. We don’t currently have a word to describe this in English.
It is not exactly resilience/robustness. It is better.
Antifragility is beyond resilience or robustness. The resilient resists shocks and stays the same; the antifragile gets better.
The opposite of fragility is not resilience/robustness in the sense of mere resistance or indifference to randomness and time. Words like “robust” or “resilient” describe systems that don’t break under stress, but what about things that benefits from stress?
- Fragile is everything that loses from chance and time.
- There are things for which randomness and time are negative risks.
- For example, a flower-vase can only lose when confronted with randomness.
- Robust is everything that doesn’t care about chance and time.
- Antifragile is everything that benefits from chance and time.
How to make a system antifragile?
Limiting the possible costs while keeping the possible upsides is one way to make a system antifragile. You protect the system against catastrophic events, but let it fully benefit from anastrophes (opposite of anastrophes).
Antifragile idea generation
by John Carmack
References:
- https://www.youtube.com/watch?v=dSCBCk4xVa0
- https://amasad.me/carmack
- https://gregorygundersen.com/blog/2020/08/05/antifragile-ideas/
Here is what Carmack thinks an antifragile system might look like
- You are working on a problem and you get an idea and with it the initial idea high
- You should instantly try to defeat your idea – think of all the ways it could not work, test it out, put it under stress
- If the idea survive the brutal scrutiny then it has legs for further investigations or implementation
- If the idea is implemented and it works then that’s great
- If the idea fails the scrutiny or implementation you can quickly move on to the next idea without feeling the lows because you haven’t obsessed or talked about it i.e. it’s not your pet idea.
Carmack describes how this becomes like a game – as soon as you get an idea you try to defeat it. You’ll be able to generate more ideas because you freed up mental space. Furthermore, your existing ideas will even be stronger because they survived heavy scrutiny.
Take aways
- Ideas are overrated and execution is everything.
- Ideas need to go throught testable throughput.
- We can estimate how good an idea might be by how testable it is. If you get an idea about something but you cannot test it (because you don’t work in that domain or for any other reason), the idea is no good. You cannot test it, scrutinize it.
- When you get an idea, try to think about it and prove why it won’t work.
- A good idea is an idea is that thrives under scrutiny; so to have good ideas, have many ideas and test them quickly.
- Good ideas do not just survive scrutiny, which is what Taleb calls “robust”; good ideas are better after scrutiny.
- Don’t overcommit (to ideas - big or small); test your ideas as quickly as possible; test alternative lines of attack.
- If you have a bad idea, it is no big deal. You only spent a few minutes thinking through the implications. But if you get lucky, if the random thought does in fact solve the important problem, you have a big potential upside: you’ve solve a hard problem and are now labeled a genius.
- If you can’t do that, start prototyping as soon as possible. The longer you delay it, the higher the risk of getting attached to it
- Falsifying an idea (either by thinking about it or prototyping) is often the way to get better at having ideas or at understanding the problem you’re working on.
- Don’t be too attached to your ideas, because most ideas aren’t actually that valuable
- Once you get attached to an idea, it’s difficult to maintain objectivity.
Examples of antifragile systems
One can argue that large open source software as antifragile. The more stress people put the software under the better it becomes. The more people use it in unanticipated ways and the more code path combinations are exercised then the more bugs are found and fixed. In contrast, proprietary software is usually used in controlled environments all the while building up fragility for a major catastrophic event waiting to happen.
Benefits of practicing this approach
This approach leads to better idea generation as we keep iterating.
This approach converts the abstract notion of “having ideas” into a process. Processes are good because they change our focus from goals, which are often long-term and open-ended, to daily habits. Whenever we are stuck while working on something, if we write down as many testable ideas as I can, the process would be freeing because ideas do not have to be good, since “good” is a property of antifragility that can only arise out of testing.
We end up prototyping a lot more and going through many iterations of ideas. Many ideas that sound great in theory end up tanking in practice. So to make progress we need to have a lot of ideas and try many of them.