Zwiebach has written a book on string theory specifically for advanced undergraduates, and on this merit alone, there is a temptation to give many stars to this text, and this I believe is reflected in the existing reviews. Well deserved praise to Zwiebach for performing this valuable service for the physics community. It will be especially useful to serious undergraduates on helping them decide on whether or not to embark on string theory as a field of research. It provides a faster than normal overview into the subject. Instead of having to invest years in learning the subject, and then (maybe) decide if you do or don't believe this is the correct approach to unification of the fundamental laws, you can decide (maybe) sooner than later if your research efforts could have been directed to greener pastures, or if you are indeed safely on the path to the `holy grail' of physics.
Traditionally, advanced topics in theoretical physics require an undergraduate to first prepare himself with a firm grounding in classical physics (mechanics, EM, thermal, relativity) and quantum mechanics, and a solid grounding in mathematics. Then, after a certain maturity is achieved, the student can study Quantum Field Theory and general relativity and then, finally, string theory.
Zwiebach is attempting to shorten and the even circumvent the traditional learning curve. One might ask, is this possible and if so how? The method Zwiebach uses is to start with introducing 4-vector notation, and explain how to calculate in local coordinates. This is similar to the approach in many field or gauge theory texts, which is not a surprise since they also rely on 4-vector notation. No tensor analysis or differential geometry is provided, but this is fine for this level of a text. After the 4-vectors are introduced, the standard advanced topics are developed as needed. (Lagrangians, Hamiltonians, Maxwell, etc) Finally, the advanced concept is generalized to higher dimensions and the string theory analog is studied.
This approach works to explain the theory on a very elementary level, which was the intent, and the student is able to naively calculate in local coordinates.
I found it slightly annoying that Zwiebach seems to constantly overstate the case for string theory, or else he gives that impression because does not bother to address concerns sufficiently that any bright undergraduate would naturally have, and it is a tone that is present throughout the text. For example (there are more than a few, but for brevity, I list only one example): He says "Are we sure that string theory is a good quantum theory of gravity? There is no complete certainty yet, but the evidence is very good". (pg. 7) A scientist must be objective and explain the good and bad aspects of the theory with a dispassionate objectivity, and doubly so when there are no experiments to moderate one's theoretical speculations. The experimental fact is we don't observe 10 dimensions in the lab (i.e., we only still see 3 space and one time coordinate). The experimental fact is we don't observe compactification of dimensions as physical phenomena. There is zero evidence for this compactification, and this compactification explanation is almost epicycle in nature, as a way to explain why we don't observe those 10 dimensions to begin with. By contrast, Polchinski in his "String Theory", vol. 1, explains how the curling up, or compactification, is consistent (i.e., it is not forbidden) with the geometry of general relativity, since in GR, space-time is dynamic. Also by comparison, Kaku in his "Introduction to Superstrings and M-Theory" seems to take the objections to string theory more seriously, and presents a nice list of the more important objections to the theory. Kaku's book, incidentally, would be a rival text for Zwiebach at the advanced undergraduate level, except it does not have exercises at the end of the chapters, and so is more useful as a reference.
At times, it seems Zwiebach demonstrates occasional lapses in physics erudition...We are informed on page 32 that Planck's constant first appeared in the famous E= (h-bar) w equation, where w is the angular frequency of the photon. Even Freshman physics courses teach Planck was quantizing oscillators in 1900, and Einstein's theory of the photoelectric effect in 1905 (for which he later received the Nobel Prize) where the photon was introduced, was more than a few years away. We are told that the Born-Infeld and related nonlinear theories are as fundamental as the Maxwell equations, if not more so. This is a more advanced error, but amazing nonetheless. Maxwell is a classical, non-quantum theory only. The Born-Infeld equation attempts to explain nonlinearities that are quantum mechanical in nature, where Maxwell does not apply. This is explained even in the introduction to Jackson. Born-Infeld and the related non-linear theories also have an upper bound on the field strength, which Maxwell does not. Coincidentally, the electric fields on D-branes also have an upper bound, so now you can guess as to why Born-Infeld has been elevated to the same status as Maxwell -- because Born-Infeld agrees with string theory, of course. I expect gaps in the rather varied and advanced mathematics one must know, but not in such basic physics --how does this happen at MIT, and in a Cambridge University text? My guess is that string theory is a very demanding mistress, to the point that only strings and mathematics can be concentrated upon, sometimes unfortunately, to the detriment of equally important areas of physics. Perhaps this should be a consideration at least, for the budding undergraduate string theorist.
Despite the bias and the occasional lapses, a good text. Recommended.