by Peter Seibel
I like the premise of this book – find some of the best programmers out there and talk to them about their craft (or art, or science, if you prefer). First thing to say though, is that the list of interviewees tells you something about the state of the computing industry today: out of fifteen people interviewed, only one is a woman. Of the other fourteen, all of them are white men. You can argue that most of these people are pretty senior and are mostly more a product of the culture of the 1960s, 1970s and 1980s than now, but still. I’ll come back to this in a minute, because it turns out that the interview with the sole woman, Fran Allen, was one of the most interesting.
Of the other interviewees, many felt like natural choices and had a lot of interesting things to say. How could you ever write a book like this without interviewing Donald Knuth, for instance? Seibel asked all of the other interviewees if they’d read The Art of Computer Programming. Knuth, of course, wrote it…
Other high points for me included Joe Armstrong, Simon Peyton Jones and Guy Steele. Perhaps this is because they’re more on the functional programming side (Erlang, Haskell and Lisp respectively), but they seem to have a much more intellectually engaged approach to programming than some of the other interviewees.
Some of the other choices of interviewees might be a little arguable, but I did find myself nodding along to Jamie Zawinski’s description of some of his coding practices, and it was quite interesting to read a little more about the history of Emacs.
The real highlight for me though was the interview with Fran Allen. She was the first woman to win the Turing Prize (after 40 years of male winners) and spent the whole of her career working on optimising compiler technology at IBM. The way she described life at IBM makes it pretty clear that the parallel technical/managerial paths they used to have (don’t know if they still do) can work really well. She also made some very interesting observations about the way the industry has changed over the years, particularly with reference to the current gender disparity.
In one of the earlier compiler projects she worked on, three of the four technical leads on the project were women, and this wasn’t considered particularly unusual for IBM (this was in the late ’50s!). They just hired whoever they could get who was qualified, although they also had explicit diversity policies as well (going back to 1899 apparently).
Seibel asked Allen if she had any thoughts on when things in the industry shifted, and she was pretty clear that things started to change around the time that university engineering and mathematics departments started offering degrees in computer science. The “credentialising” of the profession meant that all of the historical barriers to women in engineering and technical fields were also imposed on the newer discipline of computing. It seems as though there was a period when the computing industry was new when those barriers just weren’t there. It’s kind of depressing that we’ve ended up where we have.