Books, Books, and more Books!
If you’re a fan of books on tech and tech policy this is a particularly good time for you because so much new stuff is hot off the presses. Here’s a short list of the books in my reading queue at the moment, along with a couple of longish journal articles.
Obama FCC Chairman Tom Wheeler’s From Gutenberg to Google: The History of Our Future won’t be available until Feb. 26th, but it should be an interesting read. Wheeler is an amateur historian who published an e-book, Mr. Lincoln’s T-Mails: How Abraham Lincoln Used the Telegraph to Win the Civil War in 2009. This one looks like it will be a fleshed-out version of the e-book. Wheeler’s speeches while he was at the FCC tended to contain more history than current events, but he often got his facts mixed up. It will be interesting to see whether his first hard copy publisher provided him with a fact-checker.
Wheeler has broken from convention by acting as highly vocal critic of Chairman Pai despite investing time in the book. Whether it’s insightful or not, the views of influential public figures, fundraisers, and lobbyists are interesting.
Former Internet Activities Board Chairman David Clark’s book, Designing an Internet, summarizes a decade of research on 27 different visions of the Internet as it should be. Before he gets to that, he recounts the Internet’s history as it unfolded from his point of view. Clark has a better appreciation of the tradeoffs made by Internet designers because he was there at the time. His perspective is more pragmatic than that of the cheerleaders, which could be either a cause or an effect of his preference for hard research problems such a Quality of Service.
Given Clark’s grounding in history and his job as a reviewer of National Science Foundation grants for future Internet architectures, his predictions about future Internets are worth reading. Here’s the main one:
We are at a transition at which the future advances in network capability will be not higher access speeds but instead more diverse delivery services. The next measures of utility will be resilience, security, availability, and consistency of service delivery. (Kindle edition location 6381)
His summaries of the research proposals are thinner than I’d like, and I don’t think he does justice to John Day’s RINA by lumping it in with Joe Touch’s RNA. But to his credit he doesn’t waste much ink on RNA. Here’s his summary of the difference between RINA and TCP/IP on the critical subject of addressing:
RINA stresses that functions at different scopes should be independent, not entangled. The current Internet, because of its history, entangles its layers in ways that are not optimal. The addresses in the Internet do not include explicit identifiers for ASs. Initially, the 32 bits in the Internet address were divided into two parts: the first 8 bits identified the network and the last 24 bits the end point inside that network. But we quickly realized that 8 bits were not enough for the number of networks that would be built, and a much more complex mechanism was invented (called CIDR, for classless interdomain routing), in which different parts of the Internet use a different number of bits at the beginning of the address to identify the network. Today, the interior routing protocols and the inter-AS routing protocols have to work with the same 32 bit field, dealing with all the complexity created by CIDR. The RINA alternative is that each scope should have its own addresses, which are suited to the needs of that scope. (Kindle location 2781)
Clark points out the fact that RINA supports a uniform cross-layer interface, another feature the Internet lacks. He suggests that RNA designer Touch is working on a blend of RNA and RINA called DRUID, but DRUID is several years old.
Harvard Law School Clinical Professor Susan Crawford’s new book, Fiber: The Coming Tech Revolution—and Why America Might Miss It, wants to keep you up all night worrying about Comcast coming to take your Internet away. While she argues that only one network technology can deliver the future – direct fiber optic connections – her most exciting examples rely on 5G wireless:
Welcome to the future, the first hints of which were visible at the Winter Olympics of February 2018 in Pyeongchang. Korea Telecom (KT) unveiled its very-high-capacity but very-short-range 5G wireless communications network (at least ten times faster than 4G) to the world during those games, so South Koreans really could be part of speed skating or ski jumping events from the perspectives of the athletes involved, and ride autonomous buses on an eighty-eight-acre test site—the world’s biggest. It was a major collaboration: the enormous South Korean companies Samsung and Hyundai, together with Intel and Ericsson, committed to the enterprise and built hardware to handle the flood of data moving around the Olympic venues before a global standard for 5G communications—not expected until 2020—even existed. South Korea simply wanted to be first. And South Koreans with 5G-enabled devices were the most engaged sports fans on the planet during those Olympics. They could see and hear anything they chose, in real time, from any angle, including the athlete’s view, while surrounded by their friends and other South Koreans as fellow commentators. Interactive holographic projections, along with augmented reality (the digital overlay made possible with glasses or smartphones or interactive bus windows) allowed for new forms of engagement. KT’s “360 Virtual Reality” technology made viewers feel as if they were right there in the stadium, no matter where they actually were, and its “time slice” technology allowed watchers to roll these visuals back and forth at will. (Kindle location 119)
Crawford understands that 5G depends on fiber backhaul, but she misses the fact that the common broadband technologies that wire the world today – 4G, cable modem, and VDSL – are also hybrid networks that drive fiber deployments deep inside neighborhoods.
What Korea, Japan, Hong Kong, China, Singapore, and the Nordic countries have that the rest of the world does not are fiber optic cables running physically, directly, into neighborhoods, homes, and businesses in both rural and urban areas—the so-called “last-mile” network. Inventors have found ways to encode stunning amounts of information on pulses of light vibrating billions of times a second, and then send that light on its journey through a channel made of the purest glass on earth.
We do have fiber optic cables running deep inside neighborhoods even where they don’t directly touch houses. The reasons we don’t directly connect to houses have to do with our installed base of high-quality coaxial copper cable – capable of 10 Gbps – and the difficulty of splicing and troubleshooting hair-thin strands of fiber. It’s necessary to have fiber to each cell tower, cable head end, and VDSL switch; it’s not yet necessary to run it inside homes, and it probably never will be.
Most Korean households are apartments in high-rise buildings with fiber connections to switches in the basement and copper wire internally. In the US, homes with fiber connections to Verizon or AT&T always rely on Ethernet and Wi-Fi inside the dwelling. 10 Gbps data center Ethernets use copper wire because it’s cheaper and stronger than fiber. It’s not the medium that matters, it’s what you do with it. For limited runs of hundreds of feet, fiber is more trouble than it’s worth.
Harvard Berkman Center faculty associate Shoshana Zuboff‘s tome The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power is completely over-the-top account of Internet data collection and use. Zuboff’s last major book was on the role of automation in defining the nature of work in the information age. It was reasonably insightful and generally well received, but it was published before the fall of the Berlin Wall. London Times columnist Hugo Rifkin calls Surveillance “Das Kapital for the digital generation.” He’s about right.
In Zoboff’s estimation, those ads I saw for Poulan chainsaws after searching the Home Depot web site enslaved me. They took away my “right to a future tense” by predicting what I might buy and therefore subjugated me. Here’s how she explains it:
I suggest that we now face the moment in history when the elemental right to the future tense is endangered by a panvasive digital architecture of behavior modification owned and operated by surveillance capital, necessitated by its economic imperatives, and driven by its laws of motion, all for the sake of its guaranteed outcomes.
But I showed them: I buy half of my chainsaws from Amazon just to keep the monsters guessing. BTW, the term “panvasive” was coined by another great privacy hawk, Vanderbilt Law Professor Christopher Slobogin.
There are some facts and insights in Zoboff’s book, but they tend to be hidden or embellished. Consider her account of iTunes:
Apple thundered onto the music scene in the midst of a pitched battle between demand and supply. On one side were young people whose enthusiasm for Napster and other forms of music file sharing expressed a new quality of demand: consumption my way, what I want, when I want it, where I want it. On the other side were music-industry executives who chose to instill fear and to crush that demand by hunting down and prosecuting some of Napster’s most-ardent users. Apple bridged the divide with a commercially and legally viable solution that aligned the company with the changing needs of individuals while working with industry incumbents. Napster hacked the music industry, but Apple appeared to have hacked capitalism.(Kindle location 482)
Huh? Napster was cool because it made it radically easy to acquire digital music. Everyone knew it was illegal and somewhat dangerous to use. But it was right there all the time. Apple simply solved the legality problem by negotiating contracts with the record labels, something it was able to do because Steve Jobs was a Hollywood insider thanks to his acquisition of Pixar Studios. A guy making money by leveraging his connections seems pretty common in capitalism; I’m sure it was also quite common in the USSR.
Zoboff also tends to drift off into pseudo-poetic reveries that distract from whatever point she wants to make. Take a look at how Chapter 11, “The Right to a Future Tense” begins:
I wake early. The day begins before I open my eyes. My mind is in motion. Words and sentences have streamed through my dreams, solving problems on yesterday’s pages. The first work of the day is to retrieve those words that lay open a puzzle. Only then am I ready to awaken my senses. I try to discern each birdcall in the symphony outside our windows: the phoebe, redwing, blue jay, mockingbird, woodpecker, finch, starling, and chickadee. Soaring above all their songs are the cries of geese over the lake. I splash warm water on my face, drink cool water to coax my body into alertness, and commune with our dog in the still-silent house. I make coffee and bring it into my study, where I settle into my desk chair, call up my screen, and begin. I think. I write these words, and I imagine you reading them. I do this every day of every week—as I have for several years—and it is likely that I will continue to do so for one or two years to come.
I watch the seasons from the windows above my desk: first green, then red and gold, then white, and then back to green again. When friends come to visit, they peek into my study. There are books and papers stacked on every surface and most of the floor. I know they feel overwhelmed at this sight, and sometimes I sense that they silently pity me for my obligation to this work and how it circumscribes my days. I do not think that they realize how free I am. In fact, I have never felt more free. How is this possible?
I have to wonder how it was possible to get this book published. But the answer to her question is that she’s free because she contracted to write the book and her will allows her to fulfill the promise: “This act of will is my claim to the future tense.” OK. But I can agree to let Google and Facebook collect Internet data from me by using the Internet without any loss of freedom. The mere act of observation of the web sites I visit simply enables advertising networks to show me ads that they believe to a appropriate to my interests.
These ads aren’t forcing me to buy anything, and I certainly know why advertisers show me these things They’re observing me but I’m also observing them. And much of what I see is quite amusing. It might get scary if the ads were freakishly relevant to my interests, but the experience of Internet ads today is like kids running a lemonade stand. You know what they’re trying to do, and it’s charming that they have so little clue about how to do it.
You can see Zuboff tell her story at Kevin Werbach’s After the Digital Tornado conference here. While she’s an untidy thinker, Zuboff has her fans.
Honorable mention to UC Irvine Professor Scott Jordan, formerly Chief Technologist at the FCC during the Obama era. Jordan has written two long law review articles on net neutrality law from the engineering perspective. The first is on mobile broadband, “Mobile Broadband Internet Access Service Is a Commercial Mobile Service, and Hence Must Be Regulated as a Common Carrier Service.” The second is “Broadband Internet Access Service is a Telecommunications Service“. The second is free to download, but the first is paywalled.
Jordan mischaracterizes DNS and commits a number of more interesting errors. I’ll review the papers shortly.