Book Review
Algorithms To Live By
3/5 stars
"Gimmicky. Moderately interesting. Of Limited Practical Utility."

Of the book:

-256 pages of prose over 11 chapters plus an intro. (23.7/per)
-≈475 sources. (1.9/page). 

This book is difficult to describe. It has things that are either/ both interesting-in-the-conceptual-sense/ picayune-in-the-practical-sense.

And I do know that authors need a delivery vehicle to transport their message and I do understand that computer programmers need to explicitly consider every aspect of logical operations--hence the choice of Computer Science.

But: human beings are not computers, and so when these authors take a bunch of random concepts from Computer Science and staple them together to talk about ways that human beings can behave more successfully.......

What could go wrong?

For example:

1. If you have to arrange your bookshelf, how often are you really going to sit down and think about what method you're going to use to sort it? It takes more time to go through those methods than it does to just do it. And there might not even be that much benefit to doing it at all, since you have a rough idea where things are.

2. Caching. Moving any number of things to the place where you are most likely to use them. So, his example person puts vacuum bags behind the sofa because that's where they are likely to be when needed.

So, am I going to do that with every item in the house? *Can* I do that, with a whole bunch of kids running around moving things every minute of the day?

3. A lot of these problems are intractable, and the author admits as much. (p.117): "... The status of about 7% of problems is still unknown.... Of the 93% of the problems that we do understand... Only about 9% can be solved efficiently and the other 84% have proven intractable." (But then, what is the point of the exposition, then?)

4. A lot of these things are trivial: of course I'm not going to organize my bookshelf because the time investment to organize something so fluctuant is more than the benefit.  

*******
It seems like other of the topics alternate between:

1. Things that we already know from other contexts (people who have read Nassim Nicholas Taleb's books are likely to be familiar with things such as the Lindy effect /misapplication of the Gaussian distribution)

2. Things that are likely to be helpful in very limited circumstances (when you want to choose your life partner, it's probably better if you reject about 37% of what you're able to get and then choose the best right after that).

Interesting sundry tidbits:

1. Three types of distributions. Power law. Normal. Erlang. Prediction rules for each: multiplicative, average, additive.

2. Copernican principle: your best guess for how long something will last into the future is exactly as long as it has lasted already. (This is sometimes called the Lindy effect.)

Verdict: Weak Recommendation. I understand the gimmick of packaging these ideas around Computer Science, but there may have been a better way.
*******

Vocabulary:

-thrashing (where a system grinds to a halt because it is preoccupied with metawork)
-timeboxing
-pomodoros
-Dutch auction (vs. English)
Good quotes:

(p.123): "This is thrashing: a system running full tilt and accomplishing nothing at all.....computer scientists now use the term "thrashing" to refer to much pretty much any situation where the system grinds to a halt because it's entirely preoccupied with metawork. A thrashing computer's performance doesn't bog down gradually. It falls off a cliff. Real work has dropped to effectively zero, which also means it's going to be nearly impossible to get out.

"Thrashing is a very recognizable human state..... When merely remembering everything we need to be doing occupies our full attention--or prioritizing every task consumes all the time we had to do them--or our train of thought is continually interrupted before those thoughts can translate into action, it feels like panic, like paralysis by way of hyperactivity.... It's thrashing, and computers know it well.
24