My rating system

I expect books to be good. The scoring system I use is meant to hold books to a high standard. Any book that I give two or more stars in my reviews is a book that I would give four or five stars on Amazon (unless I think it is already severely-overrated by other people). For non-fiction, my ratings are as follows:

** (two stars) I would recommend this book in the sense that I don't think that reading it would be wasting your time at all, but I wouldn't recommend moving it to the top of your reading list. I think that the average reader would benefit more from reading this book than from reading most works of non-fiction. However, I would not be at all surprised if the content of this book received a much better treatment somewhere else.

**** (four stars) This book has fundamentally altered the way I think about its primary subject matter. Practically every thing it says about the subject with which the author claims expertise seems fundamentally sound to me, and I never expect to go back to thinking about this books central subject the way I did before I read this book.

**+ (two and one to three plus signs) The standards I set for four stars are too constraining to capture capture all forms of excellence. Nothing about this book truly revolutionized the way I think about the world so I can't give it four stars, but everything about this book is excellent.

*** (three stars) This book has content that deserves four stars, but it also has significant flaws that prevent me from considering it worthy of four stars. It is either poorly written, or the author demonstrates excessive ignorance of a topic that is highly relevant to the book, though it isn't necessarily particularly relevant to the author's field of expertise.

***** (five stars) The book exceeds the standards for requiring for stars. It is also extremely well written. Additionally, the author is presenting original research that he or she is uniquely well-qualified to comment upon. Moreover, when the author did that original research, he or she was producing ideas or results that contradicted the established views of his or her field.

****** (six stars) Five stars isn't enough. This book is one that I would nominate to be among the ten most important books ever written. (So far, the only two books I've read that deserve six stars are Darwin's Origin of Species and Pearl's Probabilistic Reasoning in Intelligent Systems.)

* (one star) I would not recommend this book, but it has some material and or ideas that I hope somebody else picks up and gives a better treatment in another work.

(0 stars) I would not recommend this book, and based on my evaluation, this book does not contain any material that would contribute to a book worth recommending if it were placed in a better context or given a better treatment.

+++ (three or four plus signs) A valiant effort. Something about the author's premise makes the book of limited use as a source of information and/or ideas, but the book itself is very well executed. The book itself gets enough right that I want to praise it as an example of how people should write non-fiction, but I cannot recommend its informational content.

Zero stars does not mean substantially worse than average. Most books deserve zero stars. For the most part, I try to avoid reading those sorts of books, and will seldom review them.

I have not yet figured out what I'm going to do with my rating system for fiction. In general, I think that devising an adequate rating system for fiction is much harder than devising an adequate rating system for non-fiction. For now, the basic rule is more stars is better than fewer stars.