Books I read

Books I’ve read, a short summary, mostly for my own benefit re: I have the worst memory and often forget even books I’ve spent hours and hours of my life reading and waxing lyrical about.

Artificial Unintelligence, by Meredith Broussard

TL;DR - Robots are not as smart as we think they are (or can be), there are limits to artificial intelligence, and we still need human brains. We saw this in Star Wars, and we’ll see it again, and again, and again despite the best efforts of futurists and technologists to convince us of the imminent arrival of singularity. For what can be a very technical topic, Dr. Broussard is a very accessible writer. Even though she provides some in-depth examples of how to use machine learning on specific datasets, with the corresponding code, even that was understandable and even relatable. I was trying to decide between reading this book and another similarly titled book, but like that this book is written by a relatively young female professor, who is an academic and a data journalist and a software developer, and that’s cool. I’m glad I went with this criteria, because there are not that many non-fiction books that can keep my interest to go cover to cover. Usually by the time I hit the third chapter, the author has made their core point; and every subsequent chapter is a slightly different way of rehashing that point #cough-originals.

There is a recurring point in this book as well, of course, and in the author’s words, it’s that “computers are good at some things and very bad at others, and social problems arise from situations in which people misjudge how suitable a computer is for performing the task.”

For instance, machine learning on big data sets is useful for providing clarity on events or phenomenon that have happened in the past; but far less effective at using that knowledge to adequately predict what will happen in the future. This is again, because human systems are complex and sometimes illogical, and the data will only tell some, but not all, of the story. Data doesn’t know what human or environmental factors were present that could affect why and how historic events happened the way they did, and it doesn’t know when those factors have been unaccounted for. This is the stuff of tacit and learned knowledge (not data).

Don’t worry, human-cyborg relations are unlikely to upend anytime soon. The machines Dr. Broussard talks about - autonomous vehicles, drones, computers etc. - are still lightyears away from being sentient beings. But, data in the hands of the wrong people, corporations and government - entities that we can’t reprogram, or simply unplug from a data source to keep them from doing harm? That’s an entirely different story and much more likely source of a dystopian future (my opinion, not the author’s). Stay in school, stay vigilant, stay safe, all. ~Jan2020

The Dutch House, by Ann Patchett

A tale of two children who grow up to be adults who are scarred from being children. They have a very hard time letting go of their past, much of which revolves around stalking an alluring house that they grew up in called the Dutch House, and from which they were unceremoniously ejected by their stepmother after the death of their father. It's an interesting reflection on how our past shapes and has an indelible hold on us - while being completely open to the idea that maybe the past is a warped figment of our imagination. I like this book a lot. ~Feb2020

Daisy Jones and the Six, by Taylor Jenkins Reid

This book is like an eminently drinkable table wine. Goes down really easy but it's on the cheaper side. A multi-perspective novel about a free spirit in the 1970s (Daisy) who happens to be a looker, drug addict and singer songwriter who joins forces with a rock band (the Six) and goes on tour. The main theme seems to be around how everyone goes through their own set of trials and tribulations but can still make redemptive choices in the end. It flirts with the question of the difficult genius, and whether people who are really brilliant at what they do are justified in being kind of horrible self-centered people. Which is not irrelevant in the era of #metoo. ~Feb2020