• Politics
  • The physicality of the internet: Andrew Blum


The physicality of the internet: Andrew Blum

INTERVIEW! Disruption Network Lab is back with "Deep Cables" (Jun 17-18 at Kunstquartier Bethanien), an exploration of the internet's workings. On Sat, Jun 18, author Andrew Blum takes us through the tangibility of what's behind the world wide web.

Image for The physicality of the internet: Andrew Blum
Photo by Davina Pardom

Part of the next series of Disruption Network Lab, “Deep Cables”, author Andrew Blum will speak on Saturday, June 18 at 4:30pm.

The Disruption Network Lab is back! The new series opens with “Deep Cables: Uncovering the Wiring of the World,” a two-day conference gathering writers, activists, engineers, and security experts to discuss how the rise of the internet has altered the politics of surveillance and privacy. In typical Disruption Network Lab fashion, the event as a whole will attempt to deconstruct and educate, all the while inspiring people to find new ways to take political and social action.

We sat down with the author of Tubes: A Journey to the Center of the Internet, Andrew Blum, for a chat about the physicality of the Internet (it’s not all in the air!), his experience at the Google Data Center and how the power struggle over internet ownership has changed in the wake of his book’s publication, a year before the Snowden revelations shook the internet world. His book could not have been published at a better time, and it clearly made quite a splash: it has been translated into nine languages and even won Design Observer‘s Book of the Year Award in 2012. Mr. Blum is a contributing editor for Metropolis.

So what is it you’re going to talk about on the 18th at the Disruption Network Lab.

I am going to talk about what the internet is, and how it fits together. I’ll describe how the undersea cables fit into the physical structure of the internet; to discuss the logic of that infrastructure…it’s most notable for being really intimate; there’s just not that much of it. I point out how there are about 12 buildings in the world that are by far the most important places where the networks of the internet connect to each other. (Such as the AT&T Building in NYC, which houses miles and miles of telecommunication equipment.) When I started there were no places on the Internet that said how the Internet worked and there was no place that said where the most important places were, a sort of storytelling gap. But when I started asking network engineers about what the most important places were, the answers were really consistent, so that’s what I relied on. And the control of those buildings and the ownership of those buildings has major ramifications for surveillance, for privacy, for freedom of speech, and for this broader question of who owns and who controls the Internet – which is not a simple question to answer, and it has a lot of pieces to it.

You say in your book Tubes: A Journey to the Center of the Internet that you’ve always been attracted to architecture, that you used to write about buildings, and that the places that any building creates fascinates you.  What kind of place do you think wires and ports in the internet create?

A male place, that’s one answer. This is a place that is dominated by male technicians and male network engineers. It’s a highly secure place, and that security is often mistaken for secrecy. But often the security is based on the need to protect really expensive equipment.  It’s often a pretty ugly place, because these are not only quasi-industrial spaces, but also there’s this notion that they have to be anti-monumental – they have to be anonymous. It became a bit of a running joke for me that it was just another unmarked building in a suburban business park. But the flip side to that is that there has been a movement to celebrate these buildings for what they are. And there are some older buildings that are kind of amazing monuments, like the lower Manhattan buildings, the 1930s telephone buildings, that do celebrate their purpose, even if that purpose was originally for the telegraph, and not the internet. And then some crazy developments have been that – and I’m beside myself – that Google has evolved from not having a sign on the side of their data centres and being completely secret to painting murals there. I’ve been joking for years that eventually there’s going to be a brewery tour and a gift shop. I think we’re actually close to that.

One incredible part of your book is your experience with the Google data building and how unwilling they were to educate you about the internet compared to other tech companies you visited.  Was it really as simple as you going through a gate, looking at a few things, asking “Can you tell me how it all works?” and getting non-answer answers, and then leaving?

My experience reporting and visiting that Google data centre was – to be honest – an incredible gift from them. I thought one of two things were going to happen. Either they were going to say “no” and that’d be the end of that. Or they were going to say, “Yes, we’ve been wanting to tell this story. Come on in! We’ll tell you everything!” and then the nature of a book – that very likely in the six-12 months between when that happened and my book being published – this story would go up somewhere else and I would be left with a generic story. And instead they went this third way, which was to pretend they were giving me something substantial, but in fact gave me corporate emotional nonsense about what we as Google users expect and need from them. And obviously everything in the book is true and actually happened, but it was just remarkable to me that they could be so misunderstanding what my project was.

Since then, Google seems to have changed their attitude about transparency. They’ve become a lot more open and willing to show the public what goes on behind the scenes. Should you be credited for that?

Well, [laughs] Tubes was released about four years ago, and it was that fall that the first pictures from the inside of a Google data centre had been released, and in the years since there’s been more and more transparency about what goes on. Now there’s a main Google data centre mini-website now that has a timeline and 2012 is the year of transparency [laughs]. So I’m trying to own that a little bit more. No one [from Google] has ever told me that their embarrassment over the way that they handled Tubes was the cause for this newfound transparency, but I did spend a lot of time poking at their lack of understanding, and certainly the terms have changed in the last four years. Not just the cloud generally, but also Silicon Valley’s stance and the government with the Snowden revelation, or the Apple/FBI stuff.

Do you think it really is as simple as a widespread misunderstanding that people don’t realize that the internet truly is a series of tubes within buildings and across oceans, and rather think of internet as that inaccessible, undecipherable ‘magic’?

I think there are two pieces that are worth pointing out. Firstly, the people who make the pieces of the internet, it’s unavoidable that the nature of coding is often different from the nature of storytelling, so the stories we tell about the internet lag in their sophistication and in their elegance behind the networks that are being built. I’m a part of this sort of old-school internet history LISTSERV, and it sort of amazes me that whenever a basic question is raised about the history of the internet, these sort of giants of internet chime in, in a way that’s both defensive and incomprehensible at times. And I think that those important storytelling skills have a long-term detrimental effect in understanding the internet.

The second piece is the commercial part of it. The imperative in Silicon Valley and tech culture is to scale. Nobody builds anything unless it scales. Nobody does anything just to be small. The entire goal is to scale. Which means that nobody is building any neighbourhood-scale institutions, everything is done to be worldwide and impersonal.  And I think that adds to the impulse to make the internet magical and ethereal. If it’s transparent and engaging and more accessible, then it doesn’t have that same aura of total domination.  Which – with very rare exception – is what everyone who builds a piece of the network wants. It’s like somebody who opens a restaurant or cafe. They don’t necessarily want to expand around the world – some do – but anyone who starts with something technological, they absolutely do.

Following this idea of physicality – you show how susceptible to humans (and human error) the internet is. For instance, there’s that example of a Pakistani internet worker who was able to accidentally shut down YouTube in 2008.

The internet works on a system of trust. And this is a moment where that system of trust failed. In this instance, it happened because someone screwed up. [A Pakistani telecom employee accidentally misconfigured his router, effectively becoming YouTube instead of blocking its access to other routers and computers]. It wasn’t nefarious, but of course there are instances where it is nefarious. Which again comes back to this crazy fact of how small the internet is, and how few people there are that are responsible for the connection of global networks. I remember I was at Microsoft (I don’t know how many employees Microsoft has: 40,000? 50,000? 75,000? Somewhere in there.) and I asked the question of how many people are responsible for running Microsoft’s global data network. And the answer was 250. And I was like “Okay, how many people are responsible for connecting Microsoft’s network with other networks in the internet?” and the answer was like: two. So it’s not just the operation of a network, but also the connections between networks that make it work, which are also the smallest and most intimate communities. 

It’s sort of uncanny how just a year before this stuff with Snowden, your own research sort of led you to similar suspicions that proved valid – this concept of intimacy can easily be read as secretive anonymity.

Yeah. It’s worth pointing out that it is this very intimacy which had made me suspicious of that legend of NSA spying… until the revelations came out. I didn’t believe that – given how much I’d seen, and how little hesitation people had in showing it to me – I just couldn’t believe that there was this parallel internet that was tapping it. And there isn’t – it’s more specific than that, but it’s still big enough to be a good chunk of the internet. But I think the intimacy of it and the trust involved is part of the same context – as is the surveillance – because it makes it that much more of a violation of the community.

Any surprise while researching your book. Or reactions to it?

I was certainly concerned when the book was coming out that people would say that I was doing something dangerous. The chapter about Equinix was excerpted on GIZMODO and we made headline “The bull’s-eye of America’s Internet” and we commissioned a map that drew a bull’s-eye to Equinix, and then when the article came out, Equinix tweeted “the bull’s-eye of America’s Internet” because they would rather be known as the most important place on the internet than be worried about being known as the most important place on the Internet. The commercial imperative almost always trumps the security imperative.

Disruption Network Lab: Deep Cables, Jun 17-18 | Kunstquartier Bethanien, Mariannenplatz, U-Bhf Kottbusser Tor