Turing's 1936 Paper "On Computational Numbers" is often cited as a landmark in the history of computing, but it's details are not widely considered or well known today. If you're curious to know more about the Paper, and why it's important, you can do no better than read this book. It contains a complete transcript of the original Paper, with extensive commentary and explanation from Petzold that make the Paper accessible and understandable to a wider audience (and even for specialists, this book is probably a better choice than just reading the original Paper!). Petzold's enthusiasm for the topic shines through in an excellent writing style, striking a good balance between detailed technicalities and simpler descriptions, in a friendly helpful way that will neither confuse the layman nor bore the expert.
Petzold supplies invaluable historical context: some of the developments in mathematics in the late nineteenth and early twentieth centuries that undoubtedly influenced Turing, and which appear explicitly in the paper. This is a very useful aid to understanding Turing's paper for readers not expert in those topics of mathematics (which is almost all of us who don't have post-graduate degrees in very specialised areas of pure maths!). But the book is definitely aimed at readers with some mathematical background and aptitude. If (UK) O-level / GCSE maths was a mystery to you, you may struggle; if you have A-level maths or computing you'll be fine. If you're somewhere in between, Petzold's explanations will happily guide you through the details.
Two things this book isn't: First, if you want a book that starts from Turing's paper then delves into even more advanced mathematical research and theories, then this isn't the one for you (although it does helpfully include a summary of more recent work that follows on from Turing's ideas). Second, at the other extreme: although this book includes some biographical information, if you want a detailed non-technical biography of Turing you should look elsewhere.
But for all the rest of us between these two extremes, who want to understand what Turing machines are from the original source, then I wholeheartedly recommend this book.
My only complaint, and a very minor one, is that Petzold's description of Bletchley Park's location would place it in Suffolk rather than Buckinghamshire! But given the complexity of the book's subject matter, it is a testament to the quality of Petzold's research that this is his only error.