There are many theories as to what a book review is supposed to do and it is in the application of one or the other of those theories that we can determine how successful a review is.
Following on the article re: NYT book review going into chador, I found one condemning statement made therein provocative. The article condemned the NYT reviews for not intriguing a reader into reading the book but for providing a kind of cliff's notes to it so that in a passing conversation the NYT reader could avoid reading the book but also converse intelligently about it. And I would say that the statement identified what I think central about a book review--intriguing the reader, coaxing the reader into deciding that the book is worth the time.
On this basis, I have noted that two reviews from Bookmunch in a row--one yesterday and the one posted just below this post--have successfully performed the functions of a good book review. The two books they have reviewed in the past two days are now added to my TBR list.
My view is that the purpose of a book review, as opposed to a critique or piece of criticism on the work, is to inform a reader in such a way as to encourage the reader to take up the book (or to leave it lay). I tend to be on the side of encouragement, largely because I know my own dislikes with respect to literature are highly idiosyncratic and often driven by mood, day of the week, and other extraneous factors. No writer should be punished for my inconstancy.