Reviews for Coders at Work

The average rating for Coders at Work based on 2 reviews is 4.5 stars.

Review # 1 was written on 2009-11-20 00:00:00 Mark Smith One of my many, many areas of deep intellectual insecurity is computer programming. As a kid I remember writing BASIC programs on paper in the backseat of the car during family trips to Florida. I also remember spending hours on my Vic 20 as a kid and even crashing it a few times trying to execute 6502 assembly with POKE and JMP. I whizzed through FORTRAN and Pascal my freshman year at Purdue. Then tragedy struck the next year as I got my first B in college from a microprocessor class based on the Motorola 6809. Later I had to drop a class in C because I could not get my head around pointers. After that I embraced the beautiful equations of Maxwell and Laplace and disdained the impure applied science of computers. A few years later I was in Silicon Valley begging for a job working with computers after seeing the bleak career of an RF engineer as a coop student. It was a smart move, but even after working in high tech for 20 years I still feel sometimes like I'm on the outside looking in. So I was eager to read these interviews with 15 luminaries in computer science and maybe get the answers to questions I was always too embarrassed to ask. It felt very satisfying to read, and I was not disappointed at all. Some of the interviews were dry and scholarly, but most were very interesting and thoughtful. I eventually had to put post-it notes on the inside cover to write down all the names and concepts that I need to follow up on now that I'm done with this book. It's a shame that none of the coders is as famous as the founders interviewed in the very similar book Founders At Work. Ken Thompson and Peter Norvig were the only two I had heard of before reading this book. Some of these interviews, notably with Brad Fitzpatrick and Ken Thompson, made me want to put the book down and do a Wayne's World-esque "I'm not worthy!" genuflection. The biggest surprise in this book is that almost everyone mentioned using print statements as their main method of debugging programs. I always felt like I was cheating when I did this. But hey if it's good enough for them... Other surprises: A lot of these people work for Google and are surprisingly open about the processes used there. Norvig's comments on the hiring process got some news coverage when the book came out. I guess the Google PR people don't harass the big shots as much as the rest of their people. Another surprise was hearing some of these people complaining about C and C++. I dutifully plowed my way through K&R (and may do so again), but I never thought to voice my complaints out loud. Fran Allen in particular has harsh things to say about C. ("We have seriously regressed, since C developed. C has destroyed our ability to advance the start of the art...") Someone else (forget who) said that he never wanted a degree in CS because it was like getting a degree in Microsoft. Equating computer science with trade school was my snobby rationalization for not pursuing computer science, but most of the people interviewed described what they did as a craft. I think that is more accurate. Highly recommended. Now I want to find a copy of Programmers at Work, released more than 20 years before this book, interviewing people like Bill Gates and John Warnock. I wonder what the people in Coders At Work will be doing in 20 years.

Review # 2 was written on 2011-08-12 00:00:00 Larry Shoemaker Coders at Work is one long read into the lives of several fantastic computer scientists, the software-writing variety. Peter Seibel interviews sixteen "programmers", among them Joe Armstrong (Erlang), Brad Fitzpatrick (OpenID, memcached), Simon Peyton Jones (Haskell), THE Donald Knuth, Peter Norvig (AI), and Ken Thompson (UNIX). A few of the missing topics: high-performance computing, social networking, peer-to-peer file-sharing, more Internet. Each interview goes over a number of standard questions, ranging from biographical to technical to philosophical: How did you learn to program? What is the role of documentation? What is the role of testing? What is the worst bug you have faced in your career? What is the best way to debug? What is the role of computer science courses, in particular assembly, in your career? What is the role of math in your career? How big were the teams in which you worked? What is your best achievement? Is your work science, engineering, or art? How does your schedule affect your family and social life? Seibel also tailors questions for each interviewee, knowing well their biography and confronting them about milestones, contacting former professors and current friends, creating ties between different interviewees, etc. The book is very long, but this is because each interview tries to capture not only the answers, but also the essence of the interlocutor. We get to appreciate the kindness and open spirit of Simon Peyton Jones, the non-nonsense attitude of Donald Knuth, the die-hard approach of Joe Armstrong, etc. I found this immensely appealing and well-worth the wordiness. A few of the lessons: - Most interviewees have started programming early, and among their early programs were games; - Most interviewees had a middle-class to rich family and did not have to struggle much in life; - There is no programming language that can do all, but C++ is widely regarded as an abomination; - Modern programming languages include C (performance), Python (prototyping), Erlang (distributed, fault-tolerant); - Assembly is regarded as either still important and relevant as part of computer science curriculum, or completely outdated--in the latter case, it still teaches you how to think about low-level problems; - Math is not seen as important as a whole, but computer scientists should learn parts depending on their branch of work (cryptography, graphics, etc. do require math); - The most difficult to debug programs are related to concurrency, parallelism, multi-users; - The most successful pieces of software are written alone or in small teams; once the team enlarges past, say, ten people, the ability to deliver becomes rare; - Working in this field does not necessarily lead to long hours, but it drains physically and psychically, and affects negatively family and social life; - Often, the best achievement is just shipping. All in all, an amazing book about computer scientists, and an absolute must read for anyone with aspirations for this industry.