The return of the 'Book Club'...

One of the best things about the past nine months is that I've been getting back into reading for pleasure, as opposed to reading for work [1]. I've been reading so much, in fact, that I splurged a bit and got myself a Kindle Paperwhite to replace my aging 3G. I can't stress enough how useful an illuminated e-reader is when you're trying to soothe a baby. 

But I digress. During my Ph.D, I maintained a blog - sadly, now defunct - in which I used to write regular 'book club' posts, discussing books that I thought would be of interest to folks other than myself. I think I'm going to start that up again here with the book I'm currently reading. But for now, I'd like to give a brief shout out to a couple of interesting things that I've read during the past few months.

The Signal and The Noise, Nate Silver, 2012.

Nate Silver's fivethirtyeight blog applies quantitative analysis to topics that other news organizations typically treat qualitatively, especially politics. His book focuses on forecasting: more specifically trying to identify the factors that lead to good predictions (empiricism, Bayesian inference, rigorous self-criticism and identification of distorting incentives, etc.) versus those that lead us astray (personal biases, poor/overly complex models, hidden variables, etc.).

Of note, Silver is highly critical of the lack of proper probabilistic presentation of predictions, such as how the multiple predictions of the severity of climate change's effects are implicitly presented as equally probable scenarios. As he discusses, the most catastrophic outcomes are on the tail end of a probability distribution under any reasonable model. Focusing on 'worst-case' scenarios simply make the models incongruous with the observed trend of temperature increase (I should note that Silver is not a climate change skeptic).

Finally, Silver is one of many people who point out that simply looking at the success rate of any talking head's prediction history should be enough to stop every news organization from taking their claims at face value every single time [2].

It's an interesting read, though frequentists may get upset.

Boss Fight Books, Various, 2014-

Anyone who knows me knows that I love videogames. However, I quit reading gaming magazines and websites several years ago. For some reason, in comparison to writing about other media (books, movies, etc.), games-writing is uniquely intolerable. I don't think that it's simply the result of a young medium finding its legs. Rather my frustration comes down to two factors: First, a lack of history wherein young gamers don't play old games and interpret everything through the lens of their own sollipsism, and second, a pervasive culture of hyperbole in which nuance is discouraged and everything is the best thing 'evar!!!1!11!1!!!!'    

Boss Fight Books launched as a Kickstarter with the goal of writing a series of long-form essays (~100 pages) each centered around a single classic game. Each book's approach is unique, with some written in standard historical chronicle style (à la Masters of Doom), while others range from discussions of the game's effect on the author's personal life, to in-depth dissections of a strategy game's mechanics and underlying code.

While the writing in a few of the books is overly experimental and off-putting - Galaga is particularly weird [3] - the majority are fascinating little reads. At first I thought that nostalgia would make me prefer the books about games I'd played as a kid, but I've been surprised to find that my favorites, Jagged Alliance, Bible Adventures, and ZZT, are all about games that I missed out on... but would love to revisit if and when I get more free time. 

If you enjoy videogames, I'd recommend trying out a few of the books. They're still publishing going strong, and I'd like to support them simply because, unlike every other classic game discussion on the internet, they're not focused solely on console games from Japan.  

[1] Of course, the two are not mutually exclusive. For instance, some of the books that I've read have been about coding and data analysis. Others have been about the inner workings of the American healthcare system, which classifies as both 'horror' and 'mystery'. On the other hand, I have read a bunch of pulp-fantasy, which is unlikely to help me at work unless we are attacked by orcs or trolls. Thinking about it, I'm not sure how what I've read would help in this hypothetical situation...

[2] Of course, this assumes that the primary purpose of the 'news' is information and not simply entertainment.

[3] The Galaga book is written as 255 mostly independent paragraph-long tidbits weaving advice on playing the game and memories of classic arcades with stories about the author's sexual abuse at the hands of his father. It's a courageous effort to be so open about such a difficult part of his life, but its lack of flow is jarring to the reader. I suppose that as an art piece, that may have been its objective.