How to evaluate quality of a book

Question

I'm reading quite a lot professional literature, it's often about information technologies, programming books, computer science etc.

The problem is that there are many books that are poorly written. On the other sides there are books that are very enriching, their authors know how to explain things, how to go to the core ideas and create good background for advanced topics. Some authors are really masters. Some authors give you also guidelines and references for further readings.

Could you describe me your guidelines or algorithm, how to find out that the book is high-quality before you buy it and start to read? How you will narrow the choice of 100 books about technologies (programming language, computer science, electrical engineering) and choose a few that are pearls?

Answer 1

I often use the costumer reviews from amazon. Not only for books. I read the reviews with one or two stars and the comments of these reviews. If these reviews contain good points which are backed up by arguments I can follow, these reviews give me more hints then the five or four star reviews.

Also the comments on these review reveal more pros and cons.

For example (not a book) I searched for a laser printer. I found one that could be the right one for me, but the one star reviews revealed that there was a piece of plastic that often broke and made the use impossible without long lasting repair. So I dropped the printer from my list. The same it is with books. If there are professional reviews, that claim, that the book is of little use, maybe you shouldn't take it or search for more professional reviews revealing the same issues.

Answer 2

My approach is somewhat like this:

First, I like to reduce the number of books to look at. If I am at a library (which for me as a student happens a lot) I already have a small selection, so this number is already limited. But suppose you want to buy a book: Then you can try to get recommendations from colleagues or online reviews. Another approach is to look for books from a publisher you trust: I am very confident that books from O'Reilly or Galileo won't be bad (in the worst case have a different focus or target audience).

Now we have a small number of books which we will have a closer look. I now try to get physical access to the book (library, brick and mortar store, lending from other people) or use amazons "look inside" feature. Then my approach usually is to read the introduction (where the authors will explain who the target audience is). The next stop for me is the table of contents to get an idea of how the structure of the book is, or if it covers the topics I care about. Then I just choose a topic which I'm interested in and look for the chapter where this is covered. This way I have a look at the index of the book, and then can read some pages to get a feeling for the writing style.

Answer 3

Visiting blogs that has book reviews can be pretty useful. The problem is that not all the books have proper reviews, and it's sometimes hard to find one if any. However, when you can find the blog, sometimes it's really useful.

For example, I use Eli Bendersky's blog to check some of my programming books before buying them.

Jon Skeet's review on Effective C# in his blog was very useful to me.

Answer 4

How To Read a Book has an excellent chapter written on just this. You should be able to skim the quality of a book well within an hour, perhaps down to 5-10 minutes with practice.

Start with reading the title and summary at the back of the book. It seems obvious, but a lot of people actually don't do this and even fail to read the title properly, misjudging the angle which the book takes. This should give you a hint of who it is written for, what it's about. A technical book written for novices and one for experts are very different things. You'll always want one above your level, not too high, but never want to read a book that's below your level of skill.

Read the table of contents. If the topic covers things that you already know or are being vague, chances are that it's not a good book. But in today's books, the table of contents takes a lot lower priority, so some books may be great even without a good table of contents. Subtitles can be especially useful here.

Read the last chapter, if applicable. The summary and not the epilogue. Most narrative books will summarize everything in the last chapter.

Read the introductory chapter and the publisher's blurb. The author will usually write his/her motivations for the book, as well as what kind of research and experience went into the book. This can help you decide if the author is a reliable source of information.

By now, you should already have a good idea of the quality of the book. You should make a decision whether it's still worth considering and comparing to other books by now, or you should put it down.

Once you have a small stack of books you might be interested in, go through each chapter in the book. Focus mainly on the start and end of each chapter. The start will introduce what it's about, and the end often summarizes. Technical books do not often summarize chapters, but the good, more instructive ones will. It's a good way to choose the easily readable quality books from the hard to read, but quality books. Feel free to read parts of the chapters that catch your eye, but you should mostly want a superficial view through it at this point.

This method takes a lot of practice to get used to and will be tiring, even humiliating when you start. But it's a good way to differentiate between a lot of different books on the same topic in a short amount of time.

Answer 5

I read a lot of books (or at least, bits of books) and so I've learned a lot on how to choose good ones.

Recommendations

My main guiding principle is to read books that have been recommended to me: either personally or by a professional whose work and opinions I already know and respect. The more recommendations a book receives, the more likely I am to read it. For example, I read, "Code Complete" because it was recommended by so many other developers. I am glad that I did.

Names I Already Know

I also tend to choose books by authors whose work I have enjoyed in the past. Two non-programming examples are the works of Steven Covey and John C Maxwell.

Amazon Reviews

As a last-but-one resort, I look at Amazon reviews. Quite apart from their use in choosing books, they are contain very insightful comments on an author's work... as well as numerous - er - "less useful" opinions.

Read the Book

If all else fails, I may even try reading a book to see how good it is. This is easiest where chapters are available for free download, but libraries and even bookshops are OK for a quick flip through before I buy. For example, "Getting Real" by 37 Signals is available free from their website. I downloaded it on the off-chance that it was good. It turns out that it is.

---

I nearly forgot - there is one other heuristic I use (but not always deliberateley):

Judge by the Cover

Despite the old adage, this actually does work... to some extent.

Notice, for example, the publisher of the book: some publishers seem to be more choosy than others regarding the quality of book that they're willing to be associated with. In my experience the Pragmatic books are generally good, as are many of the O'Reilly ones.

It doesn't matter how good the content of a book if its appearance puts you off reading it. Back in the day I remember receiving "free" technical books with computer magazines. They were invariable printed on poor quality paper in a poor font and with poor binding, so I never could muster up the enthusiasm to read them.

Conversely, if the pages of a book are laid out nicely and if the illustrations are clear then there is hope that the book will be able to communicate its message effectively.

Answer 6

The Table of Contents is usually my indicator of the quality of the book. If it seems comprehensive and interesting , then I read a few pages at random and decide if I like it enough to want to read it.