If I was to rate this on the quality of writing alone, Blum could win a high 4, maybe a 5, for the richness of his descriptive passages, particularly in the parts on the cable landing stations in Cornwall or the modernization of The Dalles in Oregon. Let's face it, Blum can write well and engagingly. Nevertheless, even in the writing style there are a few nagging problems. His tendency to use quotes from literary sources like Emerson or J.G. Ballard is OK when limited to once or twice in a single book, but after a while the use of such quotes sounds a little too grad-student for my tastes.
There were passages that were just trite or silly, as well - the squirrel chewing up cable, complete with exclamation points in the text; the description of Silicon Valley as startup mecca, when that description even seemed dated and pedestrian in the 1980s; and the reference to The Dalles as a digital Kathmandu. Don't get me wrong, I like the Zen aspects of Blum's search, it's just that one must be careful in using these analogies at getting too starry-eyed. Also, he gets a trifle over-dramatic in confronting the secrecy of Google and other companies in dealing with data center locations. If Blum was like James Bamford, chasing down the location of snooping centers of intelligence agencies, he'd have reason to feel paranoid. Here, his fears just seem silly.
But there is another aspect of Blum's work that makes me rank the book in the high-3's, albeit moving closer to 4. I disagree with the nature of his quest and the way he chooses to pursue it. I know, I know, that sounds like a reviewer for a travel book who says he wished the writer had gone to Spain instead of Kazakhstan. But bear with me.
Blum rightly sees a certain spiritual quest in examining the communication protocol layers of the Internet, and there's an argument to be made for treating the Open Systems Interconnect seven-layer stack as a mysterious bardo. But Blum sees the bottom two layers, physical and data-link, as representing physical macro-geography. And that's where network engineers raise their eyebrows at his quest. Does it matter whether the data center is in The Dalles or Prineville? Does it matter whether a Cisco or Brocade router sits at the center? Does it matter the locations on the planet where networks aggregate? Some might talk about planetary magnetic fields and ley lines and say, "Oh yes it does." Maybe so, but by spending too much time on large-scale geography, you miss the spiritual layers underneath.
To really make some good analogies of the type Blum strives for, you need to understand the underlying chip architectures and middle-ware software responsible for dissecting packets and putting them back together. You need to understand the Zen of Ethernet switching, multi-protocol label switching, and dense wave-division multiplexing. Then you need to be able to translate that in a way your grandmother can understand. Does that mean one needs a BSEE or geekdom certification? No, but it means one needs to go deeper into the technology than Blum did.
A similar problem exists when he equates the physical backbone of the Internet with fiber optics. This is true today, but the optics might some day be replaced by millimeter-wave radio or some sort of quantum-computing "weird action at a distance." The key to the Internet's center is bandwidth itself, and optical switching is merely the best current manifestation.
The reason this matters is that several books that made a technology deep-dive on the history and nature of the Internet were released 10 to 15 years ago, such as Katie Hafner and Matthew Lyon's "Where Wizards Stay Up Late" (1996). A lot has happened to the Internet since then, but Blum had to show he could tackle the more recent nuances and still come out with something that moved beyond the Hafner/Lyon book.
I still think this book is worth a read for learning some details of specific place - the paranoid secrecy of Google officials in discussing their data centers, for example, teaches us that Google is a lot creepier than Facebook in its own way. Blum's talents could be put to future use - he would be a great candidate to join with James Bamford in dissecting the new NSA data center in Bluffdale, Utah, for example. But I can't help but feel this book would have been a lot more interesting if Blum had used his Zen quest to dive deeper into the underlying chips and software that make the Internet hum.