Tubes: A Journey to the Center of the Internet

by Andrew Blum

  At some point soon after I arrived home from Oregon, I don’t recall exactly when, I fished my laptop from my bag and opened it up. Then, silently, effortlessly, its hidden antenna latched on to the white wireless hub behind the couch, the one with the single green eye. That meant very little in logical terms, but it meant so much to me: I was home, back at my place on the network. When the squirrel nibbled on the cable a couple years earlier, I could watch him (or was it her?) from my desk in the small room I then used as an office. In the interim, the space had been given over to my daughter, her crib now occupying the same spot. The squirrel was still there. My daughter was big enough to stand up and look out the window and wave at him. Her corner of the world was a magical place, where animals told stories, baked cookies, and said good morning and good night. And it was a small place, a constrained geography; its specificity mattered. It mattered to me too.

  It reminded me that while I’d seen many of the biggest monuments of the Internet, I hadn’t answered one of the questions I had started out with: Where did the cable go from here? How did my piece of the Internet connect to the rest? On the sidewalk around the corner was a metal enclosure the size of a steamer trunk, which I suspected held the answer. A sign on it read CABLEVISION, my Internet provider. It was decorated with stickers advertising bands, and when I walked by late in the evenings—inevitably preoccupied with these words—I could hear it buzzing quietly.

  But in a cruel irony, after so many of the Internet’s doors had opened, this one stayed mostly shut. Cablevision is a notoriously tight-lipped company, and only after months of phone calls did I finally get a friendly engineer on the phone who sketched the outlines of my cable’s path. From the living room it passed through a hole in the wall, down into my hundred-year-old apartment building’s basement, out into the backyard, past the squirrel, across two neighbors’ yards, and landed beneath the steamer trunk in a thick bundle of cable—thicker, by far, than the cables stretched across the ocean. Beside the steel trunk was a manhole marked CATV. Inside of it was a fiber junction box, a cylinder that looked a lot like a muffler, where all the cables from the immediate neighborhood were aggregated into a few strands of glass. On a street map this spot would be labeled CARLTON AVE.; on Cablevision’s network map, it was node 8M48, the M denoting this area of north Brooklyn. The cable TV network was first laid in the 1980s; since then, it had been steadily upgraded, which in physical terms meant that fiber-optic cables reached out closer and closer to customers’ homes, expanding like the roots of a tree, and with the capacity increasing each time. For now, the fiber stopped at the curb; it would soon inevitably come all the way to the door.

  In the other direction, it went to the “head-end,” a small, industrial-looking building nearby surrounded by a fence, containing a piece of equipment known as a cable modem termination system, or CMTS. This was a special kind of router and looked the part: a steel machine the size of a washer, sprouting yellow wires, humming in a lonely room. All of Cablevision’s head-ends then plugged into just a few “master head-ends.” The one in Hicksville, Long Island—where Cablevision had its corporate headquarters—was also the broadband Internet services center, or BISC, always pronounced like the soup. The large routers there were the same kind I’d seen at the PAIX in Palo Alto. They aggregated all the signals coming and going between Cablevision’s customers and the rest of the Internet. And there, the trail got interesting.

  Cablevision may not like to say much, but the company’s network engineers can’t help but be chattier. At a web page on the otherwise unused domain of cv.net, they maintained a list of the places where Cablevision connected to other networks. It looked familiar: 60 Hudson Street, 111 Eighth Avenue, Equinix Ashburn, Equinix Newark, Equinix Chicago, and Equinix Los Angeles. And since the logical routes were inherently visible, with a little extrapolation I could even get a sense for which networks Cablevision connected to: Level 3 (my favorite regen hut in Oregon), AT&T (I wonder if the company got a new sticker for its landing station mailbox), Hurricane Electric (with Martin Levy’s slideshow of routers), and KPN (next door to the AMS-IX core). I knew that by virtue of being in New York I wasn’t physically far from the center of the Internet, but it was striking to see how logically close I was.

  I no longer saw the network as an amorphous blob, but as specific paths overlaid on the more familiar geography of the earth. The images in my head were precise: a short and familiar list of specific places. Admittedly, some of them were banal; I’d seen a lot of plain concrete buildings and linoleum-tiled, fluorescent-lit corridors. But just as many were beautiful—their beauty rooted in knowing the network’s truths and the simple act of paying close attention to the world. To look for the Internet, I had gotten off the Internet. I had stepped away from my keyboard to look around and talk. No wonder then that some of the most vivid moments—the ones in which I felt most connected to these places—came outside the electronically locked doors. I remember in particular the evenings and—as any traveler would—the meals: the boardwalk fish shack in Costa da Caparica, Portugal, with the sun setting over the Atlantic (and the cable in its depths); the four-hundred-year-old country inn in Cornwall, where farmers in high rubber boots leaned on a stone fireplace; the brew pub in Oregon filled with skiers flipping through Facebook’s blue-bordered screen on their phones (the data stored nearby).

  But when I think of those moments I also think of being homesick—especially surrounded by people who were themselves home. I remember watching as Rui Carrilho, the Tata station manager in Portugal, greeted his wife and in-laws, who’d come to see the cable land on the beach (and the reason he’d missed dinner so many nights in a row). And I think of Jol Paling, who gave me a tour of Penzance’s fishing docks after we dropped his son off at soccer practice—and who took a call on the way there from a colleague half a world away (and only halfway through his workday). Or Eddie Diaz, who, after spending all night underneath the Manhattan streets, headed home for a quick shower before going back out again for his wife’s birthday. Or Ken Patchett setting down his giant mug of coffee to read the text message that arrived from his son—a sniper in the air force—who at that moment was sitting in a transport plane on the tarmac in Qatar. These guys aren’t Steve Jobs or Mark Zuckerberg. They didn’t invent anything, reshape any industries, or make a whole lot of money. They worked inside the global network and made it work. But they lived locally, as most of us do.

  What I understood when I arrived home was that the Internet wasn’t a physical world or a virtual world, but a human world. The Internet’s physical infrastructure has many centers, but from a certain vantage point there is really only one: You. Me. The lowercase i. Wherever I am, and wherever you are.

  Acknowledgments

  When I began looking for the physical infrastructure of the Internet, I knew only in the vaguest terms how it might all fit together. From the very first moments, and throughout the entire process of researching and writing this book, I benefited from the extreme generosity of time and spirit on the part of many of the people who built and operate the networks that comprise the Internet. I’ve listed all of them in the chapter notes that follow. There are a few supreme experts whose contributions pervade the entire book, and to whom I’m especially grateful: Rob Seastrom, Eric Troyer, Anton Kapela, Martin Levy, Joe Provo, and Ilissa Miller. Internetworking is a complicated business that I’ve tried to make accessible and correct, but all errors, inaccuracies, or misunderstandings are entirely my own. Expert fact-checker Erik Malikowski helped avoid many.

  My editor at Wired (now at Gizmodo), Joe Brown, saw the possibilities in this topic immediately and supported the early reporting that opened many crucial doors. My editor at Metropolis, Martin Pedersen, encouraged me for years to get to work on a book already and gave me space and moral support when I finally did.

  This book, and I, benefited immeasurably from a community of intellectual makers—writers, journalists, editors, teachers, filmmakers, curators—who are also all friends.
I’m grateful to Tom Vanderbilt, Anthony Townsend, Ethan Youngerman, Kenny Salim, Beth Schwartzapfel, Stu Schwartzapfel, Mark Lamster, Astra Taylor, Kazys Varnelis, Tony Dokoupil, Alexis Madrigal, David Moldawer, Greg Lindsay, Sarah Fan, James Sanders, Jason Hutt, Paul Goldberger, David Schwartz, James Biber, Rupal Sanghvi, John Cary, Kenny Caldwell, Rosalie Genevro, Anne Rieselbach, Cassim Shepherd, Varick Shute, Greg Wessner, Nick Anderson, Seth Fletcher, Geoff Manaugh, Nicola Twilley, Ted Relph, Kanishka Goonewardena, Kirsten Valentine Cadieux, Nik Luka, Zack Taylor, and Laura Taylor.

  I cannot imagine that there exists a more professional and talented team than the one that transformed an idea, and then a manuscript, into this book. My agent, Zoë Pagnamenta, was always two steps ahead: she made this happen. The manuscript benefited from more than its fair share of expert editing, with people looking out for it on both sides of the Atlantic. Jim Gifford at HarperCollins Canada enthusiastically gave it a home in my second home. Will Hammond at Viking in London offered subtle and profound suggestions the whole way through. At Ecco, Dan Halpern, Shanna Milkey, Rachel Bressler, Allison Saltzman, and Michael McKenzie together put out amazing and beautiful books; I’m grateful for the support and attention they’ve given this one. Hilary Redmon generously and enthusiastically adopted the project. My editor, Matt Weiland, pointed out the most scenic paths, unfailingly improved each draft, and brought an intensity to the production process that every book should be lucky enough to have.

  And every writer should be lucky enough to receive the kind of support from the Blum and Pardo families that I did; not only were they always there cheering but—to a woman—they also offered many astute suggestions. I’m especially grateful to my parents for not once in twenty years suggesting that being a writer wasn’t a noble, practical, and worthwhile profession. Phoebe was born with this book; so much of the thrill of writing it was the thought of her someday reading it. Above all, I’m grateful to Davina for her patience, love, and insight: my center.

  Notes

  Prologue

  In the F. Scott Fitzgerald story: F. Scott Fitzgerald, My Lost City: Personal Essays 1920–1940, ed. James L. W. West III (Cambridge: Cambridge University Press, 2005), p. 115.

  Senator Ted Stevens of Alaska described the Internet: The comments were made in the Senate Commerce, Science, and Transportation Committee’s hearings for the “Communications, Consumers’ Choice, and Broadband Deployment Act of 2006” on June 28, 2006. The full audio can be downloaded at http://www.publicknowledge.org/node/497.

  The New York Times fretted: Ken Belson, “Senator’s Slip of the Tongue Keeps on Truckin’ Over the Web,” New York Times, July 17, 2006 (http://www.nytimes.com/2006/07/17/business/media/17stevens.html).

  “The cyborg future is here”: Clive Thompson, “Your Outboard Brain Knows All,” Wired, October 2007 (http://www.wired.com/techbiz/people/magazine/15-10/st_thompson).

  The Silicon Valley philosopher Kevin Kelly: Kevin Kelly, “The Internet Mapping Project,” June 1, 2009 (http://www.kk.org/ct2/2009/06/the-internet-mapping-project.php).

  Sure enough, one stepped forward: Lic. Mara Vanina Oses “The Internet Mapping Project,” June 3, 2009 (http://psiytecnologia.wordpress.com/2009/06/03/the-internet-mapping-project/).

  a “hard bottom,” as Henry David Thoreau said of Walden Pond: Henry David Thoreau, Walden and Other Writings, ed. Brooks Atkinson (New York: Modern Library, 1992).

  1: The Map

  My education in mapping the Internet, as well as many basics of the Internet’s geography, came thanks to the patience of the excellent people at Tele-Geography, including Markus Krisetya, Alan Maudlin, Stephan Beckert, Bonnie Crouch, Roxanna Tran, Nicholas Browning, and former employee Bram Abramson—who also fished TeleGeography’s first Internet report from the depths of his files and mailed it to me. In Milwaukee, Dave Janczak gave a great tour of the Kubin-Nicholson printing floor and filled in the company’s history; Dr. Steven Reyer of the Milwaukee School of Engineering—and keeper of the “Milwaukee Architecture” website—rounded out some historical details about the building. I’m especially grateful to Jon Auer for opening up his particularly vivid piece of the Net. At the Oxford Internet Institute, Mark Graham helped my understanding of the challenges of mapping cyberspace.

  “Industrially, Milwaukee is known…”: The WPA Guide to Wisconsin (New York: Duell, Sloan and Pearce, 1941), pp. 247–48.

  In one well-known incident in February 2008: For an excellent analysis of the outage, see the Renesys Corporation’s report The Day the YouTube Died, June 2008 (http://www.renesys.com/tech/presentations/pdf/nanog43-hijack.pdf). For a song about the outage, see http://www.renesys.com/blog/2008/04/the-day-the-youtube-died-1.shtml.

  2: A Network of Networks

  A shelfful of books helped me to understand the Internet’s history; I’ve listed them below. At UCLA, I’m grateful to Leonard Kleinrock who gave the better part of an afternoon to sharing stories. My understanding of the murky history of MAE-East was thanks to Steve Feldman, Bob Collet, and Rob Seastrom. On Tysons Corner, Paul Ceruzzi’s book Internet Alley was indispensable. And out of nowhere, Matt Darling sent me the 1980 ARPANET directory, which he’d pulled out of the trash twenty years before.

  Janet Abbate, Exploring the Internet (Cambridge: MIT Press, 1999).

  Paul E. Ceruzzi, Internet Alley (Cambridge: MIT Press, 2008).

  C. David Chaffee, Building the Global Fiber Optics Superhighway (New York: Kluwer Academic, 2001).

  Katie Hafner and Michael Lyon, Where Wizards Stay Up Late (New York: Simon & Schuster, 1996).

  Carl Malamud, Exploring the Internet (Englewood Cliffs, NJ: Prentice-Hall, 1993).

  Stephan Segaller, Nerds 2.0.1 (New York: T.V. Books, 1998).

  Kazys Varnelis, The Infrastructural City (Barcelona, Spain: Actar, 2008).

  “the Internet lacks a central founding figure…”: Roy Rosenzweig, “Wizards, Bureaucrats, Warriors, and Hackers: Writing the History of the Internet,” American Historical Review 103, no. 5 (December 1998): 1534.

  I should have known things wouldn’t be so clear-cut: Janet Abbate, Exploring the Internet (Cambridge: MIT Press, 1999), p. 2.

  “Not ideas about the thing but the thing itself”: Wallace Stevens, The Collected Poems of Wallace Stevens (New York: Alfred A. Knopf, 1954).

  As the philosopher Edward Casey writes: Edward S. Casey, “How to Get from Space to Place in a Fairly Short Stretch of Time: Phenomenological Prolegomena,” in Senses of Place, eds. Steven Feld and Keith H. Basso (Santa Fe: School of American Research, 1996).

  I tried to get a little postmodern kick: Walter Kirn, Up in the Air (New York: Doubleday, 2001).

  “The Work of Art in the Age of Mechanical Reproduction”: As published in Visual Culture: Critical Concepts in Media and Cultural Studies, eds. Joanna Morra and Marquard Smith (New York: Routledge, 2006).

  Columbia law school professor Tim Wu points out: Tim Wu, The Master Switch (New York: Alfred A. Knopf, 2010), p. 198.

  Winston Churchill said about architecture: In a speech to the House of Commons, October 28, 1943, as quoted by the Churchill Center and Museum (http://www.winstonchurchill.org/learn/speeches/quotations).

  a claim repeated in James Bamford’s bestselling 2008 book: James Bamford, The Shadow Factory (New York: Anchor, 2008), p. 187.

  It’s to this that Gore owed his purported claim: For a good overview of the government’s role in the development of the Internet, see NSFNET: A Partnership for High-Speed Networking, Final Report 1987–1995, available at http://www.nsfnet-legacy.org/about.php.

  Anthony Townsend has pointed out: Anthony M. Townsend, “Network Cities and the Global Structure of the Internet,” American Behavioral Scientist 44, no. 10 (June 2001): 1697–1716.

  3: Only Connect

  From the earliest stages of this project, Eric Troyer at Equinix was a constant source of information and guidance; I couldn’t have understood the Internet without his expertise. Also at Equinix, Aaron Klink, Dave Morgan, and Felix Reyes were genero
us with their time on both coasts; David Fonkalsrud at K/F Communications opened the door. And I’m grateful to Jay Adelson for the day he spent going down memory lane—and the faster ride, to the future, in his electric roadster.

  As E. B. White said of New York: E. B. White, Here Is New York (New York: Little Bookroom, 2000).

  the local venture capitalist John Doerr once described: Andy Serwer, “It Was My Party—and I Can Cry If I Want To,” Business 2.0, March 2001.

  As the MIT sociologist Sherry Turkle describes: Sherry Turkle, Alone Together (New York: Basic Books, 2011), pp. 155–56.

  a “major global connectivity hub”: Rich Miller, “Palo Alto Landlord Sues Equinix,” Data Center Knowledge, September 20, 2010 (http://www.datacenterknowledge.com/archives/2010/09/20/palo-alto-landlord-sues-equinix/).

  John Pedro would earn US patent 6,515,224 for his technique: Accessible at the United States Patent and Trademark Office’s online database, http://patft.uspto.gov/.