Structure and Interpretation of Computer Programs - 2nd Edition (MIT Electrical Engineering and Computer Science) (Hardcover)

To use an analogy, if SICP were about automobiles, it would be for the person who wants to know how cars work, how they are built, and how one might design fuel-efficient, safe, reliable vehicles for the 21st century. The people who hate SICP are the ones who just want to know how to drive their car on the highway, just like everyone else.

Those who hate SICP think it doesn't deliver enough tips and tricks for the amount of time it takes to read. But if you're like me, you're not looking for one more trick, rather you're looking for a way of synthesizing what you already know, and building a rich framework onto which you can add new learning over a career. That's what SICP has done for me. I read a draft version of the book around 1982 and it changed the way I think about my profession. If you're a thoughtful computer scientist (or want to be one), it will change your life too.

Some of the reviewers complain that SICP doesn't teach the basics of OO design, and so on. In a sense they are right. The book doesn't directly tell you how to design and write an object-oriented program using the subset of object-oriented principles that show up in the syntax of Java or C++. Rather, the book tells you what those principles are, how they came to be selected as worthwhile, how they can be implemented from the ground up, and how a different combination of principles might be more appropriate for a particular problem. This approach requires you to understand the range of possibilities, and to think about trade-offs as you go through the design process. Programming is a craft that is subject to frequent failure: many projects are started and abandoned because the designers do not have the flexibility, experience and understanding to come up with a suitable design and implementation. SICP gives you an approach that will succeed, but it is an approach based on principles and wisdom, not on a checklist. If you don't understand the principles, or if you are the kind of person who wants to be given a cookbook of what to do rather than to think creatively, or if you only want to work on problems that are pretty much like the problem you worked on last time, then this approach will not work for you. There are other approaches that will be more reproducible for a limited range of simple problems, but there is no better way than SICP to learn how to address the truly hard problems.

Donald Knuth says he wrote his books for "the one person in 50 who has this strange way of thinking that makes a programmer". I think the most amazing thing about SICP is that there are so FEW people who hate it: if Knuth were right, then only 1 out of 50 people would be giving this 5 stars, instead of about 25 out of 50. Now, a big part of the explanation is that the audience is self-selected, and is not a representative sample. But I think part of it is because Sussman and Abelson have succeeded grandly in communicating "this strange way of thinking" to (some but not all) people who otherwise would never get there.

I have learned enough to write a couple books on Lisp that (currently) have four to five stars. Yet SICP, which is pretty much the bible of our world, has only three? How can this be?

Reading the reviews made it clear what happened. An optimistic professor somewhere has been feeding SICP to undergrads who are not ready for it. But it is encouraging to see how many thoughtful people have come forward to defend the book.

Let's see if we can put this in terms that the undergrads will understand -- a problem set:

1. Kenneth Clark said that if a lot of smart people have liked something that you don't, you should try and figure out what they saw in it. List 10 qualities that SICP's defenders have claimed for it.

2. How is the intention of SICP different from that of Knuth? Kernighan & Ritchie? An algorithms textbook?

3. Does any other book fulfill this purpose better?

4. What other programming books first published in the mid 1980s are still relevant today?

5. Could the concepts in this book have been presented any better in a language other than Scheme?

6. Who is al? Why is his name in lowercase?

The real point is to teach some very core foundations of computer science that show up everywhere. For example, supposedly revolutionary XML looks a heck of a lot like a nested scheme list, first described in 1960. And processing an active server page (or Java server page) is very much like the textbook's specialized language evaluator. Finally, c++ polymorphism through vtables and part of Microsoft's COM mechanics are the exact same thing as the book's data-directed programming section.

This is very deep material for a programming newbie to learn outside a course, but for an experienced nerd who's looking for a systematic framework, it's absolutely terrific. I had done lots of lisp and compiler work before reading the book, so many of the concepts were not new. But it's with this framework in mind that I learn new technologies, and this approach greatly speeds up how long it takes to understand each week's "new" hot product/language/tool/mindset. Put another way: why do so many popular computer books take 1000 pages to describe a few trivial concepts?