This book is very much in my wheel house--computer science and interdisciplinary analysis, and I really enjoyed it. I think it makes for a lot of interesting conversation. This may become 5 stars if it sticks with me.

I really liked the ideas about how our seemingly irrational behaviors are often due the fact that the problems are intractable and thus are rational, how the fact that it takes longer to remember things might not be degeneration, but a natural consequence of searching massive amounts of data, that piles are not a disgraceful form of organization, overfitting (I am personally angered by how this happens in our education system), computational kindness, and how exponential backoff "offers a way to have finite patience and infinite mercy."

My one critique was the section about pecking orders and dominance hierarchies, while yes there are observations in the natural world, and that can be related to sorting, making arguments that hierarchies are "scientifically" better or more peaceful is strongly akin to past scientific rationalization that have been used for centuries to justify racism, sexism, etc. It is better to just not go down that road. ( )

One of those books which take concepts from some field and show how they are (or could be) used in daily life (just like [b:Think Like a Rocket Scientist: Simple Strategies You Can Use to Make Giant Leaps in Work and Life|51720411|Think Like a Rocket Scientist Simple Strategies You Can Use to Make Giant Leaps in Work and Life|Ozan Varol|https://i.gr-assets.com/images/S/compressed.photo.goodreads.com/books/1570447685l/51720411._SX50_SY75_.jpg|72907223]).

The narrative is great and entertaining. However, working with computers most of my life, I didn't find it very interesting as the book would try to explain very well known ideas. It can be very different for someone without an IT background though.

What I liked about the book is that I realised that I'm using some of the concepts from computer science as mental models in my daily life and it is such a common sense I didn't really think about it before. ( )

"This is a sunk cost..." ???
Sort vs search trade-off
Explore vs exploit
Protect your priors: turn off the news
Idolatry of data
ACK - acknowledgement of receipt. In speech too! ( )

Started a bit slowly for me and had to return and recheck as well, but wow, what a book! So many useful nuggets and frameworks:
Stack of papers on my desk is great organization! LRU model places most recently used (and most likely next needed) item on top.
Prediction depends on priors. Language distorts priors, especially when magnified by Internet/press. Carefully evaluate priors. (Plane crash deaths since 2000 is small number. Car crash deaths greater than population of Wyoming.) ( )

This is more of an introduction to problems in computer science, than "ten algorithms that will change your life".

I think the thesis of the book is that "there are analog equivalent of computational algorithms and architecture, and they can inform life decision-making". And it opens strongly to support this claim. The first chapter is on optimal stopping, for which the rule of 37% is indeed a powerful solution. It gives its user 37% chance of selecting the best candidate in real-life scenarios such as interviewing candidate and house hunting, where opportunities present itself one-by-one, and there is no chance to backtrack to an earlier rejected one.

However, as one reads on, the analog equivalent of algorithms get less and less well-defined, and the rewards seem less and less concrete and exciting. Partly it is because sometimes our intuitive response to problem is already optimized. For example, the big pile of documents on your desk actually represents a decent implementation of caching (using LRU, or least-recently-used strategy). On the other hand, the book also introduces a whole class of problems for which there is no optimal strategy, or in computational terminology, untraceable. The book becomes meta in this aspect - it goes from "how to solve problem" to "how to attempt a problem if it is unsolvable." For that several strategies were offered, such as sampling, constraint relaxation and simplification. Sometimes knowing when to stop optimization is the most wise thing one can do. It is not the author's fault that the book becomes weaker as it gets to this territory, the problems are simply too specialized to be discussed meaningfully in a popular science book.

The lesson is it is always worthwhile to figure out how hard the problem is before attempting a solution:

- First, try to figure out whether an optimal strategy already exists. If yes, what is the big O notation of this solution? Does it scale well?
- If there is no, try to solve a variation of it. Can the problem constraint be reduced or baked into scoring system? Can the rules change? Is there a three-way trade-off between space, time and accuracy?
- Walk away if marginal return for further optimization is low.

Also some rules to live by:

- Constantly provide feedback in communication so people know their last message went through
- Be clear and honest, because guessing another person's mind is computationally untraceable
- Prefer simplicity because extra parameters make models brittle
- The goal optimization is *less* computation

Overall a fruitful read and I would recommend to anyone with an interest in computer science and mathematics.

Bonus: There are quite a lot of intellectual heroes in this book. I think I will check out Von Neumann (mentioned in multiple chapters; this guy is a genius) and William Vickery (who is also the hero of [b:Radical Markets|36515770|Radical Markets Uprooting Capitalism and Democracy for a Just Society|Eric A. Posner|https://i.gr-assets.com/images/S/compressed.photo.goodreads.com/books/1519379027l/36515770._SY75_.jpg|58236821]) ( )