Sigh! Unlike the reviewer pond-jumper, I have actually read "QED", in the proper sense. In fact, I have proofread it. Being a colleague of the author, I am hardly an unbiased reviewer, and though I regard "QED" as a gem of a book, the potential buyer needn't accept my opinion. Please peek inside "QED" to judge for yourselves.
What I would like to do is to address pond-jumper's criticisms of "QED". To begin, he (it's gotta be a he) objects to the author's characterisation of a mathematical proof. pond-jumper doesn't specify his worry here, but it is fair enough to be concerned or confused over exactly what constitutes a mathematical proof: mathematicians and philosophers have been debating this for thousands of years, and there is plenty of room for disagreement. The approach in "QED" is to avoid the pedantry, to emphasise the clear, intuitive ideas at the heart of some mathematical theorems. As such, the book does not contain completely rigorous proofs, with the last I's and T's dotted and crossed. But the arguments are clear and convincing (and beautiful), AND the arguments are correct: the mathematicians/pedants CAN easily fill in the details if they so wish. There is no sleight of hand in "QED", no "professor's trick".
pond-jumper's substantive complaint is that the author's proof that .9999999... = 1 is incorrect. In fact, the author is absolutely correct; I will briefly explain how pond-jumper has led himself astray.
Any use of infinity is problematic, prone to confusion, and infinite decimals are no exception: in high school (and, sadly, often at university), the difficulties are simply ignored. Here, the question is, what happens to .999...999... when it is multiplied by 10? The author (correctly) claims the result is 9.999...999..., each 9 moving one place to the left. pond-jumper claims that "a 0 (ZERO, not nine) fills in at the end", giving the result 9.999...9990. This is his mistake: neither a 0 nor a 9 is placed at "the end", because there is no end! That is what the dots after the last 9 indicate, that the pattern goes on forever, without end. It's pond-jumper's leaving off those dots (writing .999...999 instead of .999...999...) which has permitted his error.
It surprises many people that .999...999... could equal 1; in fact pond-jumper claims that it is impossible, that "a fundamental premise of mathematics is that no number is equal to any other number". Here, pond-jumper confuses the number 1 with the possible REPRESENTATIONS of that number. For example, the "fraction" 5/5 is the same number 1, even though it looks very different on the page. The same is at least possible for .999...999..., and it is in fact true.
Of course, none of what I have written here PROVES that .999...999... equals 1: for that, I urge the potential buyer to peek inside "QED", to see for themselves this (six pages in), and many other, beautiful proofs.