Peter Seibel

Thanks for having me, folks. It's been fun.

I think the interviews in Coders support Tim's observations. Almost all of my interviewees talked about reading code, at least early in their career. (Some of them still do it--Donald Knuth was quite moving on the topic.) And they were all having fun. Though as I mentioned in another answer, that latter bit could be due to selection bias--I wanted to interview fun, interesting people.

As I mentioned in another topic, they are (or were) all passionate about programming. And they all worked on interesting projects. But that latter may be an artifact of my selection process; I had heard of them because they worked on interesting software and I chose them because that would be more interesting to read about. So there may be equally great programmers who've been toiling away, unnoticed, on boring software in the bowels of some big company, who you and I have never heard of.

I think the answer to your first question (about learning Java late in the game) is answered only implicitly by my interviewees: many of them have learned many languages over the course of their career. So, at least based on their example, I'd say it's never too late for a language switch.

And some of my interviewees have ended up moving into slightly more supervisory roles but none of them have left the technical track entirely. So clearly it's not a requirement to become a Pointy Haired Boss.

I asked them about how they liked working with other people--in what kinds of teams and how they liked to split up the work. So that gets at some of that. And of course their stories of how they worked on actual projects are also quite interesting. For instance Zawinski described the early days at Netscape were every body was loud and abrasive but they got a lot done and then later he and a partner worked on the Netscape mail client in a much more quiet but equally productive way.

I think at one point I asked for comments about questions people thought I should ask but I don't recall that that got as many responses as the basic call for names. I mostly made up the questions myself. In the end it really came down more to a half dozen topics that then got expanded back out to questions in a way customized for each interviewee: how did you lean to program, how do you design software, how do you debug software, what team organizations do you like, some stuff about programming languages and choice thereof, what is programming.

Bill Johnston wrote:how do these programmers seems to feel (or better yet think) about the general stratagems to programming (OO, functional, procedural ... )? Are any general trends apparent to you that would indicate general movement in one direction or another, and that would be useful for all the rest of us to realize?

Well, one big lesson from the book is there are lots of ways to write great software. Or at least, if you're as good as these folks, you can do it with all kinds of different tools and techniques. I also came away with a sense that programming languages probably don't matter as much as we like to think. (And I'm a big languages weenie.) I did come away thinking that Simon Peyton Jones and the Haskell, pure-FP guys are onto something. But maybe that's just because that was the area I was least familiar with going in.

I talked a bit about the phenomenon of long hours and intense concentration that often seems necessary for writing software. For some of my interviwees that lead to discussion of finding balance between work and the rest of life.

I suspect you might come away from reading the book with the beginnings of an understanding that there are no rules, at least none that everyone agrees on. I suspect a brand new coder would benefit from reading these interviews but might also want to come back and read them again after getting some experience of their own.

Everyone got roughly the same number of pages. The book about 1/3 the length of the raw transcripts so some people got cut more than others. (The people I traveled to interview I spent more time with than the ones in the Bay Area so I naturally had more text from them.)

I think there were probably a lot more in commonalities--they are all, after all programmers--than differences. Thus the differences are more interesting. For instance Ken Thompson described his extremely bottom-up approach to design and claimed to never have known a good programmer who worked more top down. Yet Bernie Cosell described himself as working in a fairly top-down way.

pooja jain wrote:what is emacs?

In addition to the link above, you can read all about its invention in the Guy Steele chapter.

indra negi wrote:What made you write this book? Does this book reflects the challenges faced by you or your knowns? What was the inspiration behind it?

Apress, the publisher, had the original idea for the book as a follow on to their book Founders at Work. And they knew me from when I wrote Practical Common Lisp for them and asked me to do this one. I tried to ask my interviewees about topics that would be of interest to any programmer with a broad interest in their craft.

You'll probably enjoy Brad Fitzpatrick's description of late nights at the data center trying to keep Live Journal running and inventing the software he needed to scale it up as it was melting down around him. And Zawinski has some war stories from Netscape days about having other groups come in and screw up his software. ;-)

I talked about Design Patterns some with various people though not particularly tied to any language. Joshua Bloch mentioned the Gang of Four book as one every programmer should read but Jamie Zawinski brought it up as a an example of a terrible book that he hated.