Lately I've been thinking (again) about books and book reviews, mostly because I'm looking to see if it makes sense for PASS to continue doing them. I hope you'll drop me a comment or two at the end.

I think Amazon has a pretty good system of reviews in that they get enough of them per book that you can get past the one guy who loves/hates the book for whatever reason. I typically read a few reviews before buying unless I know Im going to buy the book anyway. Mostly the reviews are right, but there are still times when I liked a book more or less than I thought I would based on the review. Not perfect, but useful.

For the most part those views only help me once I'm looking for a book based on a few keywords - do I buy this one or that one? But what about the book I didn't know about? Big Brown for example. The harder part is discovery, and that's where I think reviews can highlight something I wouldn't know about otherwise.

I used Big Brown as a recent example, and I also recently reviewed a book from Michael Coles on Full Text Search. Maybe a few of you discovered those books here? Even if you didn't decide to buy, maybe those are topics you hadn't thought about yet? I bet even if you read my notes you were at least tempted to read the Amazon reviews? Nothing wrong with that.

Would it be just as interesting to just feature a new SQL book each week with the standard ad copy to with it? Kind of like the word of the day? Or how about a running list of books I own? Or no longer own!

Why do I care? No, I'm not writing a book! I think that Books Online is a useful resource, as are sites like SSC and MSSQLTips.com, but sometimes you need the deeper/longer learning that comes from a book (or a class, but often that's a different focus). I want book authors to keep writing (not a huge profit margin for most) and I want to spot books that I might not yet know I want to read.

Do you like book reviews? Do you often buy based on them? What makes a good review? Are there other approaches that make more sense?