Rating: 
Summary: A book about computer science, not programming
Review: It is clear that most of the negative reviewers of this book simply do not understand what it is about. Many have confused it with being about learning programming or an argument that Scheme is the best language. This is simply wrong. The book is an entire CS curriculum compressed into a 1 semester course. This is an intellectually demanding approach that evades most readers on the first pass. I read this book as a freshman at Northwestern (a Chicago-area school that requires it for the 1st and 2nd quarter intro CS sequence) and failed to understand it. I knew that I had read about some tremendously powerful ideas and techniques, but many of the key ideas lay just outside my grasp. Nevertheless, the book so impressed me that I declared and completed a CS degree. In the years following, I took several courses: OS, Algorithms, Compilers, AI, Databases. I then went back and re-read SICP. I discovered that the book had prepared me for the material in all of the upper-division courses by surveying, breadth-first, all of the major ideas of computer science. The upper-division courses provided the depth and comprehensive coverage that SICP did not provide. On the second reading, I discovered the beautiful synthesis of ideas that SICP, and only SICP provides. I now believe that the particular strength of using SICP as an introductory text is as by providing a foundation for understanding all of computer science. It provides methods for understanding and evaluating programming paradigms and languages. It teaches you the fundamental tradeoffs and constraints involved in system implementation. It is a tour de force of the CS world. If you're interested only in programming, don't read SICP. There are many other texts more oriented towards the beginning programmer. However, if you seek to understand computer science, do not hesitate to read this book.
Rating:
Summary: Most difficult programming book I've ever read
Review: This book is, in a word, challenging. I'm a Computer Science student and this is the most difficult book I've ever read. For example, there's no FOR or WHILE loops used in this book. All loops use recursion. As you go on with this book, the programs tend to get larger and larger, because it builds on programs introduced previously, and at some point it's hard to keep track of them. This book uses the language Scheme, which I believe has no practical use, yet it is a good language for your programming skils (such as recursion), and it's an interactive language, too, which means I don't have to compile every time after I modify my codes. Let me warn you again, this book is very challenging. If you're not ready for it or have bought it and found it difficult, may I suggest Simply Scheme by Brian Harvey. My CS class uses SICP as the textbook, and many of my fellow classmates also bought Simply Scheme as an aid.
Rating: 
Summary: Worth the trouble.
Review: Working through this book reminded me of a couple of lines from "Zen and the Art of Motorcycle Maintenance": "How do you paint the perfect picture? Make yourself perfect, then paint naturally." How do you write perfect code? Make yourself perfect, then write code naturally. This book is a great way to perfect yourself for coding. If you find the approach a little too dry and academic, then try "The Little Lisper" or "A Little Java, A Few Patterns." These don't cover the same range of subjects, but they get you thinking in a good way about how to develop solutions.
Rating: 
Summary: junk
Review: There's no useful material here, whatever your level of sophistication. Beginners will find that Scheme, and the book's approach in general are confusing. Those who are advanced will find the text is full of bull, and a book like "Introduction to Algorithms" by Cormen et.al. will provide you with many more theoretical and practical topics alike, but it's advanced (and hence more interesting than SICP drivel). If you are a beginner, "C++ how to program" by Deitel will teach you all about object oriented and top down programming, plus you'll learn the C++, the most common language out there. Whatever is presented, is of such poor quality, and so dry, I couldn't stay awake long enough to finish my assignments. You can learn how to be a good programmer from any computer science course or text, and there are only several sections about that in this book. If you are looking for interesting theoretical ideas, look for other texts, such as the ones mentioned above. From what I've read, I get the impression this book was written in a rush, and with the purpose of cramming tons of nonsense and filler in just to have an excuse to present Scheme (a pathetic language). The authors claim to present a view of programming that's widely applicable, and state that top-down (i.e. procedural) programming is by far not the best way to think about programs abstractly. Well, they don't stick to their promise. They present a biased, one sided promotion of scheme as the best language out there. No attempt is made to make comparisons between functional languages like scheme and normal languages like C. The end result, you come out knowing scheme and a few boring applications of it, but can't apply any of it to real programming tasks. This proves this book is just a fancy concotion of some egghead professor who has no idea what happens in the real world. I found many explanations to be incoherent and contradictory. The whole approach is divorced from the computing practice. This is no 21 day book, in fact, you can spend 21 months on it and still not learn a thing, because there's nothing but worthless garbage here.
Rating:
Summary: An amazing piece of work
Review: While there certainly are better books for every topic covered in SICP, no other book does such a good job covering such a wide range of CS topics. It's not for the faint of heart and initial chapters will probably scare the mathematically inclined. I find it interesting that many of the poor reviews are from the Chicago area. I can only presume one of the local universities included this book in their 2nd semester weedout course. In any case, I have a request for the future negative reviewer. I was hoping you would suggest a better choice for prospective purchasers looking for a book surveying important concepts in CS.
Rating:
Summary: An Exceptional Book
Review: SICP is a great book. If you are a patient reader and willing to work the exercises, then this is the single best book you could read about programming in general and about the Scheme dialect of Lisp in particular. I found it well-written, thorough and clear. Read this book with an open mind. It is not "Teach Yourself Programming in 21 Days". Many people don't appreciate programming theory; they would rather get on with it and start programming. If you are one of these, then you may find yourself irritated by the theory SICP includes. However, if you want to excel as a programmer it is worth taking the time to learn the theory. Some ideas may penetrate slowly but when they do, the rewards will make up for the effort. This book can be used as an introductory computer programming text, but I suspect that students won't get full benefit until they have a couple of years of programming experience under their belt. This book will make you a MUCH better programmer--something that the many books that teach a programming language but ignore computing principles can not do.
Rating: 
Summary: What is your purpose for reading this book?
Review: If your school uses this book as a text of a certain introductory cs course that you have to get a "good" grade for, you definitely want to have some help to understand it in such a short time. That is, at most, a 16 week semester. On the other hand, if you really want to explore the world of programming on your time and on your own pace, it is the ultimate source of programming magic. Some people from MIT and UC Berkeley say these wonderful things about this book, yet they are only true when you are really after the knowledge not the grade.
Rating:
Summary: shoddy text
Review: The book is just that, a big bore. It's too confusing for beginners, and it'll put to sleep anyone who knows a little somthing about computer science. I would agree that some of the general topics presented are important, but how they are presented is a disgrace to computer science. Instead of talking about relevant ideas in compiler and language design, authors try to implement a scheme interpreter in scheme. The material was completely divorced from the real world. I showed it to my buddy, doing a Ph.D. in compiler design, and he laughed long and hard at the whole chapter 4, which also manages to completely mispresent databases and query languages. Being a grad student in computer engineering, I found something to scoff at in chapter 5 & 3, where they give a backwards presentation of assemblers, digital circuits, and circuit simulators. Come to think of it, the only worthwhile material was in chapter 2 & some of chapter 3, but that was on data structures, and Cormen's "Algorithms" is infinitely better. My impression is, and others have suggested, that the goal of this book was self gratification for the authors. They wanted to have a book with all kinds of stuff in there, that would be used by copycat instructors all over the world to confuse beginners and bore knowledgeable students. This book would get the prize for the most obfuscated computer science book.
Rating:
Summary: intro text to Computer Science at both MIT and Berkeley
Review: This text was written by two professors at MIT for MIT's introductory computer science course. The fact that it is still used today as the intro text at both MIT and Berkeley speaks for itself.
Rating:
Summary: you'll find this a good book if you can understand it.
Review: I had to read this book for my intro. to computer science as a freshmen in UC Berkeley. At first, this book appears quite dry and I don't understand most of the concepts. So I read each chapter at least 2 times try to figure out what Abelson is trying to explain. When I finally understand the concepts, I find it a great book for college students who what to have a soild start as a computer science major. Good luck on reading.
| 
