"Most reviews give a bell-shaped curve of star ratings;"

Not true now [1] and not true back then [2].

[1] http://portal.acm.org/citation.cfm?id=1134743 [2] http://econpapers.repec.org/paper/eclstabus/1670r.htm (Fig. 1)

The same was true for YouTube comments, and that's why they switched to Like/Dislike buttons. When it comes to judging the value of things, people tend to be all or nothing.

That's more the case if you show some sort of aggregate 5 star rating on the front page, I suspect. If you look at peoples iTunes libraries they probably use 2-4 stars a lot and 1 or 5 stars more rarely. But on Amazon, people who think a book is 4 stars but see it rated as 3 stars are going to vote for 5 stars to push it up more.

They should aggregate the reviews by taking a median. That way there's no incentive to game the system.

Only works if people understand there's no incentive...

Which they won't, cause math is so hard...

Only the median voter(s) would need to understand that there's no incentive.

> When it comes to judging the value of things, people tend to be all or nothing.

Not really. The sample is self-selecting. People who neither love nor hate don't care enough to vote.

Also see pg's review, referenced but not linked from Norvig's: http://www.amazon.com/review/R3G05B1TQ5XGZP/ref=cm_cr_pr_per...

Lecture videos: http://groups.csail.mit.edu/mac/classes/6.001/abelson-sussma...

Although note they are for the earlier edition of the book.

I don't understand how someone could give it anything but 5 stars.

Imagine if all you need to do everything is turn PSDs into HTML and your job title is programmer.

And then someone shows you this book and tells you that every good programmer loves this book and it is confusing. The reaction is always negative.

That is not the intended audience, it's a book about programming, not about some specific technology.

Well, yes, but the people giving it one star appear to not understand that they weren't the target audience. That's the point.

The book is not really about programming. It is more about thinking.

You can find two well-reasoned critiques at http://what-the-rain-knows.com/tag/krc/

The first one is by Philip Wadler, who's quite famous in Haskell circles. The second one is by Matthias Felleisen, is quite active in using Scheme to teach computer science.

Nobody of those says that SICP isn't a really great book.

Wadler claims that the book and a 'better' language would be an improvement.

Felleisen et.al. claim that the book does not prepare the students for software engineers in industry, is very difficult and thus is not the best book for a typical introductory course.

Even though I understand both critiques, I see their point of view and find both views valid, SICP is still * out of five stars.

It's sitting open on the desk next to me. I have loved seeing SICP-related links show up on HN over the past 6 months that I've been reading it.

I loved SICP. That review definitely made me want to read TAoCP.

Even for those who love SICP (and I count myself among them), reading TAoCP cover to cover would not necessarily be enjoyable or fruitful. It's useful as a reference, but not as a teaching text. (IMO.)

Sounds like a good description of "sensing" vs "intuition" in the Myers-Briggs personality type indicator.

http://www.myersbriggs.org/my-mbti-personality-type/mbti-bas...

"Sensing" types will definitely hate this book, as they hate all high-level conceptual talks and writings.