Facts and Fallacies of Software Engineering, by Robert L. Glass

Summary: A collection of facts related to software engineering, from management to maintenance to a variety of other topics, that are either often overlooked it controversial.

Thoughts: While I found this book interesting, I'm not sure if I actually got anything out of it. Many of the facts relate to areas of software I'm not really involved in (management, for instance), and many of the references that I might trawl to find more items for my to read list are instead journal articles.

The Master Algorithm, by Pedro Domingos

Summary: An overview of the different tribes within machine learning, with a focus on trying to unify them.

Comments: First, I have to say that I was looking for Machine Learning, A How-To For Dummies, and instead got Machine Learning, A Brief History. Which is really my fault for not paying more attention to what I was actually getting. But anyway. This book somehow manages to both be fairly technical and not require much, if any, background knowledge. So if that's something that interests you, I'd definitely recommend this book. On the other hand, I could definitely see most people who are interested in the gritty details also wanting some how-to info, which you're not going to find.

The Pragmatic Programmer, by Andrew Hunt and David Thomas

Summary: This book is basically a collection of tips and other best practices in writing software (or preparing to write software or judging the software once parts of it are written).

Comments: This isn't really a book to read straight through. It does seem like it might be a good reference though. Assuming I remember that I have it.

Code Complete, by Steve McConnell

Summary: This book is a series of tips about how to write good software. It starts with the design process, moves on to coding, and then debugging. The author manages to fit quite a few tidbits into about 1000 pages.

Comments: I think I learned quite a few things, but I would say most of the value of the book lies in being able to reference it later. And in spite of being long (and taking me 6 months to read), I found it interesting. I kept stopping to write down things that I found useful or amusing or that caused me to reflect on something. I did find a couple (probable?) inaccuracies, but neither of them had anything to do with code; they were both in examples. One related to the highways in western Washington (a reference to an exit that doesn't exist) and the other related to car speedometers (a claim that they read 0 when you're backing up because they can't go negative). I'm not sure what percentage of my comments related to code and what percentage were things like squee over the penguin that was on a picture of an iceburg or noticing random pictures from somewhere in western Washington (the author is from somewhere not far from Seattle, which is probably why places I'm familiar with kept popping up). My notes also included a long list of books and articles that might be worth reading. There were a number of references to studies that people had done, and I'm curious how they came up with their numbers and if they have any speculation as to cause. I would say it's worth reading for any programmers out there, but probably over the head of most other people.