My initial response to this book was awe and pleasure. It's the kind of book I might put on my coffee table (if I had a coffee table). The use of photos of people and places from around the world helps intensify the feeling of diversity that is the hallmark of the world's 6,000+ languages. I found the charts, drawings, and maps very helpful, though I do share with an earlier reviewer a little dismay about the accuracy of the maps. As that reviewer wrote, this is, after all, an atlas, which is a book of maps; if the maps are askew, then the value of the book is seriously weakened. I seriously hope the authors and the publisher will take this criticism to heart.
In addition to the errors pointed out by the other reviewer (regarding Gascon/Basque on page 213; the languages of Belgium on page 41--French & Flemish, not just Flemish) there is no line indicating where Azeri is spoken (p 50); furthermore, I believe that the major language of Iran is Farsi, an Indo-European language, not Azeri, and not any Altaic language. Finally, the authors do need to include a chart of symbols used in the text. For example, one reviewer complained of the use of something that looks like a question mark "?" next to words such as "mi?" on page 28. Actually, it's not a question mark; it's a symbol that indicates a "glottal stop". (For a discussion, readers may consult a linguistics textbook or use a search engine like Google to look up the term.) In any case, if the authors use it, they should explain it.
All of this said, I still feel pleased about the overall quality of the book. The discussion of the relatedness of languages is very important because this is a hotly contested issue in the study of languages. There are many people who are opposed to their language (or language family) being related to any other. Some of their reasons are because of the difficulty of providing proof of such things as Proto-World or Nostratic. Some of their reasons are because they've invested lots of time in their own specialty and really cannot accept alternative explanations. Some are because of professional jockeying for position and jealousy and all the other "sins" of scholarly omission and comission which academics are liable to.
For example, Prof. Comrie was quoted on one website as refuting the notion of the Altaic language family, but that was in an article written 20 years ago. In this book, written and compiled under his direction, there is a sizeable discussion of this (putative) language family (e.g. pages 46-47). The most controversial part of the discussion is the relationship of languages like Korean and Japanese to the rest of the world. Are they related to each other, are they the offspring of a long distant Proto-Altaic, are they sui generis, perhaps like Basque? As a student of these two languages, I see the relatedness, and Prof. Roy Andrew Miller has made what to me stands as a significant contribution to the idea that they are part of this larger Altaic group. However, his is apparently a minority opinion. Seeing Prof. Comrie and his team include the possibility of the relationship tips my opinion in favor of this book as an important contribution to the field.
In short, it's a beautiful book which I am glad I bought and which makes an important contribution to our understanding of the great and fascinating diversity of languages around the world. However, authors and publisher take note: fix those maps!