2020-05-09

Short reviews: fiction

Cryptonomicon (Neal Stephenson)


Cryptonomicon is a hard novel to summarise. It is about World War II code-breakers and 1990s tech entrepreneurs, but also manages to concern itself with most other things as well.

I first read Cryptonomicon over two years ago. However, it is a massive book, and since it happens in the same universe as The Baroque Cycle, I assumed reading it again would reveal many new things. I was not wrong.

Neal Stephenson has a humorously extravagant (baroque?) writing style that is always entertaining to read, but in Cryptonomicon it is taken to an extreme. Stephenson turns mundane activities like writing a business plan, eating cereal, taking a car ride in the Philippines, and visiting a dentist into lengthy but hilarious tangents. Do they contribute to the plot? Who cares!

As this is a Neal Stephenson novel, certain vices will also be present. A printed version of the book, dropped from a bomber, would punch a hole through the deck of a Japanese warship. The plot meanders to an extent that puts most rivers to shame. And some things are just plain weird.

But overall, Cryptonomicon makes for a great read for anyone with the time to spare, and an interest in codebreaking, history, war, mathematics, the Internet, the financial industry, or technology.


Exhalation (Ted Chiang)


“Exhalation”, this short story collection’s titular work, is the greatest short story I have ever read. (You may read it online for free – and legally, as far as I can tell – here). The careful setup builds to a beautiful and intuitive analogy that make the philosophical points at the end hit hard.

Based on the strengths of “Exhalation” (the short story), I bought Exhalation (the short story collection). None of the other stories surpass “Exhalation”, though they are mostly good and sometimes excellent.

Reading a Ted Chiang story is like watching an eerily intricate machine in action, or listening to a Bach fugue: the feeling is one of orderliness and precision combined with an almost casual ease. The premise of each story is fundamentally a thought experiment; a “what-if” question knocks down one domino and the story follows its consequences all the way down the chain. Nothing is wasted or in excess, and the beats of the pacing come like metronome beats. In the best of the stories, these beats are almost undetectable at first, gradually building up into dawning revalation as the pieces fall together and the story reaches its climax.

Aside from “Exhalation”, there are two stories that stand out.

“The Truth of Fact, the Truth of Feeling” is a thoughtful exploration of the effect of the medium on what is seen as true (a topic that Neil Postman would feel right at home with). The story cleverly parallels the story of a person in an African village being introduced to literacy in the past, and a person in the future grappling with the consequences of technology that records everything people see. In a world of cautionary tales about technology stealing our identities, destroying our communities, or letting dinosaurs loose in the park, Chiang’s take on this issue is surprisingly forward-looking.

In “Omphalos” (an Ancient Greek word for “navel”, as in the expression “navel of the world”), the what-if question is: what if creationism were true, but humanity was a side-effect rather than the pinnacle of creation? The story is told in the form of prayers to god. Chiang takes the reader on a tour of what the scientific facts of this world look like: old trees with no growth rings in the middle, mummified people without navels, and so on, until finally a physics discovery, while confirming without doubt the existence of miracles, also leads inevitably to the conclusion that we are not the purpose of god’s creation. All this takes place in parallel with the emotional arc of the central character, which is told in a sympathetic and realistic manner.


Summerland (Hannu Rajaniemi)


The year is 1938. The Spanish Civil War rages on, Europe braces for war, Queen Victoria reigns from the afterlife, and the Soviets are merging souls into a godlike overmind, starting with Lenin’s.

In the alternative universe of Summerland, Marconi discovered more than he bargained for when working with radio transmission, and soon enough ectotanks and other supernatural weaponry were being deployed in World War I. Since then much of early-1900s spiritualism has been proven right.

Most significant is Summerland, an afterlife where souls can lodge themselves (provided they have a ticket) and even interact to a limited extent with the living.

In terms of plot, Summerland is a fairly straightforward spy novel. This is executed well (though my judgement may not be representative of those who know more about spy novels), but the premise is what makes Summerland special.

(Rajaniemi is best known for his far-future science fiction trilogy, which starts with The Quantum Thief; this is also recommended.)


The Curse of Chalion (Lois McMaster Bujold)


At the time of writing, the “Reception” section of the Wikipedia page for this book tells me nothing but “The book has received a number of reviews”.

This rather underwhelming (though doubtlessly accurate) statement does not do the book justice. The Curse of Chalion shines not through outstanding excellence in one respect, but rather by bringing a variety of good elements together: characters that feel like real people, an atmospheric setting, and above all a hard-to-pin-down tastefulness where nothing is in excess.

If I had to critique something, some of the turning points in the plot are rather deus ex machina. However, overall the book is a great example of fantasy built on literary merits rather than genre props, and makes for a very enjoyable story to get lost in.

(The introduction of the Wikipedia article, however, is little but a list of all the awards the book has won.)


Unsong (Scott Alexander)


In the beginning God created the heavens and the Earth. For a while, everything was fine. Then Thamiel, the left hand of God, appears in the centre of the Earth and corrupts a third of the angelic host. A war begins between the angels and demons, in which the demons gain the upper hand. Their victory is averted only when the mathematically talented archangel Uriel initiates his backup plan: switching the world from running on divine light to running on mathematical laws. Angels and demons both are reduced to mere metaphors, and the world is saved.

Saved, that is, until humans get very good at harnessing those laws and send Apollo 8 on a trip around the moon in 1968. Unfortunately all space beyond the moon is simply an illusion to make the universe seem consistent with the physics that now reigns on Earth. Instead of looping around the moon, Apollo 8 crashes into the edge of the world, damaging the delicate celestial machinery that Uriel put into place to maintain his conversion.

Various glitches start to show up in the working of the world. Angels and demons begin returning: Uriel reappears in a hurricane in the Mexican Gulf, from where he plays the role of an overworked sysadmin issuing a constant stream of patches to prevent physics from crashing, while demons spring up from Lake Baikal and start invading Russia.

The backstory of Unsong, told in various short excerpts throughout the book, continues with a very clever account of how the world reacts to this turn of events. Cold War politicking continues; for example, at one point Henry Kissinger successfully convinces President Nixon to ally with Hell in order to keep the Russians in check.

The main plot line begins in 2017. In this universe, kabbalah works. In particular, it makes possible the discovery of Names of God – words which have magical powers, but whose distribution is controlled by strict copyright laws. The main character, Aaron Smith-Teller, is a gifted kabbalist, but works a low-paid job helping a company find Names: he reads potential Names off a computer screen all day long, and if he finds a Name, gives it over to the company. The process cannot be automated because computers lack a soul and hence can’t detect which words are Names, necessitating this sort of low-skill work.

Unsong is remarkable not just for its crazy premise, but for the consistency and ruthlessness of its internal logic (which characters do not fail to exploit). Imagine you stumble across a Name that grants souls to inanimate objects. What do you do? That’s obvious: use it on a computer, have it start searching for new Names at superhuman speed, sell the Names for profit, buy more computers, and continue in this vein until you have magic powers beyond your dreams and can take over the world. If the Bible is literally true, what is the overriding moral priority? Simple: end the existence of hell; countless people suffering eternal torture for vague reasons cannot be part of a just universe.

The central question that many of Unsong’s characters grapple with is the problem of theodicy: why would a good god create a world with so much evil? This question does not have direct relevance to our own world, but it leads to other interesting questions (as well as giving the author a chance to flaunt their ingenuity; the book actually has a plausible answer). Together with characters who are often both idealistic and ruthless – I’m particularly fond of Jalaketu West, AKA “The Comet King” – this makes the book a good exploration of many moral themes.

Be warned, though: Unsong is about a universe where words, rather than equations, are the building blocks of reality. This leads to a lot of perverse verbal ingenuity, including more puns than can possibly be healthy. If you don’t want to read about characters who protest at the World’s Fair by waving signs saying “No it isn’t!”, or how atheists also include a leviathan in their mythology by calling the whole world a giant fluke, stay away.

Unsong was published online, chapter by chapter. This means two things, one bad and one good. First, it is a bit less polished than a published novel might be. Second, you can read it for free online.

Short reviews: non-fiction

The Feynman Lectures on Physics (Richard Feynman)


The Feynman Lectures on Physics (FLOP) is an incredible resource on basic physics. Feynman has an inimitable style: he is always clear, never the slightest bit pretentious, and has an eerie ability to cut through tangles of models, assumptions, and equations to get at the fundamental point. Often you can feel Feynman’s infectious enthusiasm through the page.

There are some issues with trying to learn physics from FLOP. There are no exercises, so you cannot test your understanding very easily.

Another reason is that the easy flow and elegant arguments make it less structured. If a typical textbook is like taking an official tour through a city, methodically exploring everything there is to see, the general feel of FLOP is more of chasing after a boundlessly enthusiastic tour guide as he zips from place to place using various shortcuts, leaving you with the nagging feeling that, while it was certainly very fun, you might not be able to retrace the route afterwards.

Feynman has a remarkable ability to introduce just enough background to pull off some proof or argument. This makes for some brilliant arguments that are fun to follow, but, particularly when it comes to mathematical tricks, left me with the feeling that if I didn’t have more background than Feynman introduces, I would be lost.

(Given a solid understanding of calculus and complex numbers, there are no great leaps required to follow the mathematics in volume 1. Volume 2 deals mainly with electromagnetism, which relies on vector calculus; at the time I was reading it, I didn’t have a solid grasp on that and this made parts difficult to follow. I still haven’t had a chance to read through everything in the second half of volume 2, and have read nothing from volume 3 and so cannot comment on it.)

Overall, FLOP is a brilliant resource. Perhaps it works best as a reference volume; there are many arguments that I do not remember off the top of my head, but which I remember are presented with extreme clarity in FLOP. Of course, without reading through at least once, how will you know what’s in it?


The Character of Physical Law (Richard Feynman)


The Character of Physical Law is based on another series of lectures Feynman that gave. It attempts to squeeze out maximum understanding and reflection about what physics is about from a minimum of abstruse maths.

It succeeds.

The focus is not on what the laws themselves are, but rather on the common themes in many of them: conservation principles, symmetry, and, of course, maths. The combination of clear explanation and reflection without pretence or overstretched philosophy is unbeatable.

If you read one popular physics book, make it this one. It is as close to the heart of physics as you can get without heavy mathematics.

If you are serious about physics, you will of course have to dive into the maths. But read this book anyways.


Origin Story: A Big History of Everything (David Christian)


I have rarely agreed with the purpose of a book as much as I do with the purpose of Origin Story.

The idea is that an origin story explaining where the world came from and what humanity’s place in it is has been a foundational part of most human cultures in history. Ironically, just as our civilisation is now figuring out the real answers to these questions, a collective understanding of our “origin story” is missing. This is the gap that Origin Story – and the field of big history in general – aims to plug.


The Great Leveler: Violence and the History of Inequality (Walter Scheidel)


Inequality is a trendy topic. Coherent insights into its history and how to quantify it are notably less trendy.

The Great Leveler provides both in spades. Optimism is in somewhat shorter supply. Scheidel identifies “Four Horsemen of Leveling” that have historically driven large decreases in inequality: total war, violent revolution, state collapse, and pandemics. If, as Scheidel cautions, welfare democracies probably won’t buck this trend, it looks like coronavirus is our only chance.


The Strategy of Conflict (Thomas Schelling)


You are in a car, driving directly towards another car. You will soon crash. The rules of the game are simple: the first one to swerve loses. How do you win? You close your eyes, throw away the steering wheel - basically, anything that both removes your ability to act and credibly signals this to your opponent.

The Strategy of Conflict is all about delightfully - and sometimes scarily - counterintuitive problems in game theory, in particular conflict of the nuclear sort. The general theme is that reducing your ability to make choices and committing to irrational acts can be the most powerful tools at your disposal. If you can commit to something in advance, regardless of whether it is in your rational interest to do it when the time comes, you can change the payoffs for your opponent, and hence possibly change what they calculate their best action to be.


The Doomsday Machine (Daniel Ellsberg)


My brief notes on this book snowballed into a full review, which you can find here. If you’re getting tired of the coronavirus pandemic, why not put things into perspective by reading about nuclear war?


Founders at Work (Jessica Livingston)


Founders at Work is a collection of interviews with startup founders. The book doesn’t try to be anything fancy, or make any deep conclusions about how the technology industry works. Its main value – and this is not a trivial thing – is as a source of “virtual experience” that you can download into your brain. Reading dozens of founders reflecting on their experiences with the guidance of a knowledgeable interviewer is the second-best thing to having that experience yourself.
Perhaps the two most basic and recurring themes are:
  1. In a (good) startup, everything is as barebones, minimalist, and plain as possible. The working place might be the stereotypical garage, someone’s apartment, or there might not even be one. Money is saved in endlessly creative ways. At most, you occasionally might have to dress up or pretend to have a normal office to impress investors. This theme is summarised by a story told in the introduction: some people tried to figure out how to make a sports car go faster, and eventually realised the key was to remove everything that makes it look like it goes fast.
  2. In the early stages, no one has any idea what they’re doing


Security Engineering (Ross Anderson)


The lecturer for my current software & security engineering course is publishing the third edition of his security engineering textbook online chapter by chapter (“like Dickens’ novels”, as he describes it). The textbook is extremely readable, and many of the case studies are both illuminating and funny. Read it here.

Be warned that most of the chapters will disappear from the website for several years after the book is published. However, they will return afterwards, and the same page linked above also has all the chapters from the second edition, which has already passed this period and is free online forever.