Title:

Tubes

Author:

Andrew Blum

Category:

Non-fiction

Rating:

Date Reviewed:

May 12, 2013

picked up the book Tubes because it got a sterling review from NPR. Unfortunately, I disagree with that reviewer. Tubes is not enthralling, it is not filled with neat tidbits about the formation and expansion of the Internet. Tubes is actually rather dull. Blum spends way too much time focusing on the physical components of the Internet, while basically ignoring the ingenius software that makes it all work.

One thing that I missed from Tubes is any "sense of wonder" of the Internet. Blum gets so bogged down in traveling to various buildings around the world, that the whole miracle of routing and connectivity fails to come through. He does talk at one point about the amount of traffic going through certain hubs, but the raw numbers that he throws out don't convey just how impressive this technology truly is. It's not just the bandwidth that is impressive, it also the packet switching and routing, and the redundancy of the network. But Blum focuses so much on the physical parts of the internet that I felt he missed the true magic. It might have been interesting to show how an actual message gets passed through the Internet, perhaps showing a wireless link and then down into the backbone and then routed through various hubs before reaching its destination. But as I have already said, Blum focuses on the physical part of the Internet - the glass fiber cables and their connections at the various network switches - he spends little time explaining about the traffic that is carried by those fibers.

Still, a book that focused only on physical part of the Internet might still have been an interesting read if Blum had some interesting stories to tell. But mostly he just travels to different data centers, where he visits nondescript buildings that house much the infrastructure. Sometimes he even gets a tour of the place, but how much can you learn just by looking at a bank of servers and switches? Its like looking at a supercomputer and trying to guess what it might be computing. Blum's stories are rather dull, the little ancedotes about the building of the Internet are not very interesting. This book is light on content and interest, it is too bad Blum seemingly spent so long traveling around and writing it.