Karu’s blog

Charts the course of the Chandler project. Provides an amazingly encyclopedic view of software development. Scott Rosengberg gets through the Open Source software movement, intricacies of software development, good overview of Fred Brook’s mythical man month, and lots more. James Fallows calls this the successor to “Soul of the New Machine”. I don’t quite agree, but a worthy software and modern-day version of Soul of the New Machine. If Mythical-man month is too technical, this is a great replacement. I liked reading the book and it does a good job of explaining software engineering and the fundamentals behind writing good software.

After reading the book though I was left wondering what hope(if any) there is for the new constructs for parallel programming triggered by multi-core! If programming sequential programs is already so hard and is resulting in missed deadlines is there any hope in the real world for the added complexity of concurrent programs, notwithstanding the so-called atomicity and isolation benefits of transcations.

For people in software engineering this may be blase, but I found this observation (quoted in book) from Richard Gabriel, distinguished engineer at Sun, interesting.

“My view is that we should train developers the way we train creative people like poets and artists…what do people do when they’re being trained, for example, to get a master of fine arts in poetry? They study great works. Do we do that in software engineering?….No. You don’t look at the source code for great pieces of software…So you don’t study the literature of the thing you’re trying to build.”

This is a pretty bad book. It extends this hypothesis that the 60s counter-culture revolution shaped the modern computer industry. Creates all kinds of convoluted connection between LSD in the 60s, Vietnam war protests, ARPA, Engelbart etc. Seems like a concocted web more than anything else. The one insightful thing I found in the book was the observation that, all the intellectual heavyweights in the 60s and 70s were firmly behind a network centric computing paradigm, but for some reason when personal computing actually took off in the 80s PCs became stand-alone units primarily. The book didn’t explain why this occured – yet another weakness of the book.