“Secret Servers”, an article by James Bridle originally published in issue 099 of Icon magazine, looks at the relationship between architecture and the physical infrastructure of the internet. I found Bridle’s last few paragraphs particularly provocative:

“What is at stake is the way in which architects help to define and shape the image of the network to the general public. Datacenters are the outward embodiment of a huge range of public and private services, from banking to electronic voting, government bureaucracy to social networks. As such, they stand as a new form of civic architecture, at odds with their historical desire for anonymity.

Facebook’s largest facility is its new datacenter in Prineville, Oregon, tapping into the same cheap electricity which powers Google’s project in The Dalles. The social network of more than 600 million users is instantiated as a 307,000 square foot site currently employing over 1,000 construction workers—which will dwindle to just 35 jobs when operational. But in addition to the $110,000 a year Facebook has promised to local civic funds, and a franchise fee for power sold by the city, comes a new definition for datacenters and their workers, articulated by site manager Ken Patchett: “We’re the blue collar guys of the tech industry, and we’re really proud of that. This is a factory. It’s just a different kind of factory then you might be used to. It’s not a sawmill or a plywood mill, but it’s a factory nonetheless.”

This sentiment is echoed in McDonald’s description of “a new age industrial architecture”, of cities re-industrialised rather than trying to become “cultural cities”, a modern Milan emphasising the value of engineering and the craft and “making” inherent in information technology and digital real estate.

The role of the architect in the new digital real estate is to work at different levels, in Macdonald’s words “from planning and building design right down to cultural integration with other activities.” The cloud, the network, the “new heavy industry”, is reshaping the physical landscape, from the reconfiguration of Lower Manhattan to provide low-latency access to the New York Stock Exchange, to the tangles of transatlantic fiber cables coming ashore at Widemouth Bay, an old smuggler’s haunt on the Cornish coast. A formerly stealth sector is coming out into the open, revealing a tension between historical discretion and corporate projection, and bringing with it the opportunity to define a new architectural vocabulary for the digitised world.”

Though Bridle does not make this link explicit in the article, the idea of a potential “new architectural vocabulary” is clearly related to the “New Aesthetic” that Bridle began talking about this past May.  (I’ve always liked Matt Berg’s description of it as a “sensor vernacular”, and Robin Sloan’s “digital backwash aesthetic”.  I’m not sure either of those capture exactly what Bridle’s been talking about — more like pieces of it — but they all dance around the same set of things, or at least similar sets.)  Here’s Bridle’s original description, pinched together:

For so long we’ve stared up at space in wonder, but with cheap satellite imagery and cameras on kites and RC helicopters, we’re looking at the ground with new eyes, to see structures and infrastructures.

The map fragments, visible at different resolutions, accepting of differing hierarchies of objects.

Views of the landscape are superimposed on one another. Time itself dilates.

Representations of people and of technology begin to break down, to come apart not at the seams, but at the pixels.

The rough, pixelated, low-resolution edges of the screen are becoming in the world.

And when that — a new aesthetic vocabulary — gets linked to a “re-industrialization”, pulling together aesthetics, culture, economics, and politics, you’ve got a pretty significant project.  I’d like to talk about this at more length later, but for now I will just quote from Dan Hill’s fantastic 14 Cities project.  (Independent of the concerns in this post, the whole project is worth a read.)  This is the fourth of the fourteen fictional future cities Hill describes, “Re-industrial City”:

“The advances in various light manufacturing technologies throughout the early part of the 21st century — rapid prototyping, 3D printing and various local clean energy sources — enabled a return of industry to the city. Noise, pollution and other externalities were so low as to be insignificant, and allied to the nascent interest in digitally-enabled craft at the turn of the century, by the early 2020s suburbs had become light industrial zones once again.

Waterloo, Alexandria and the Inner West of Sydney through to Pyrmont once again became a thriving manufacturing centre, albeit on a domestic scale, as people were able to ‘micro-manufacture’ products from their backyard, or send designs to mass-manufacture hubs supported by logistics networks of electric delivery vans and trains. Melbourne had led the way through its nurturing of production in the creative industries and its existing built fabric.

In an ironic twist, former warehouses and factories are being partially converted from apartments back into warehouses and factories. Yet the domestic scale of the technologies means they can coexist with living spaces, actually suggesting a return to the craftsman’s studio model of the Middle Ages. The ‘faber’ movement — faber, to make — spread through most Australian cities, with the ‘re-industrial city’ as the result, a genuinely mixed-use productive place — with an identity.”

[For more on the New Aesthetic, read Rob Walker's recent interview with James Bridle at Design Observer.  It's also well-worth checking out the essay in Domus by Alexis Madrigal that the image at top is taken from.]

["Bundled, Buried, and Behind Closed Doors", a documentary short by Ben Mendelsohn and Alex Chohlas-Wood, looks at one of our favorite things -- the physical infrastructure of the internet -- and, in particular, the telco hotel at 60 Hudson Street. It's particularly fascinating to see how 60 Hudson Street exhibits the "tendency of communications infrastructure to retrofit pre-existing networks to suit the needs of new technologies": the building became a modern internet hub primarily because it was already a hub in earlier communications networks, permeated by pneumatic tubes, telegraph cables, and telephone lines, and thus easily suited to the running of fiber-optic cables. (This is important because it demonstrates the relative fixity of infrastructural geographies -- like the pattern of the cities they are embedded in, the positions of infrastructures tend to endure even as the infrastructures themselves decay and are replaced.)]

Alexis Madrigal points out an article in the New York Times this morning which starts to uncover some of the specifics of how the Egyptian government unplugged the internet.  Quoting from that article:

Because the Internet’s legendary robustness and ability to route around blockages are part of its basic design, even the world’s most renowned network and telecommunications engineers have been perplexed that the Mubarak government succeeded in pulling the maneuver off.

But now, as Egyptian engineers begin to assess fragmentary evidence and their own knowledge of the Egyptian Internet’s construction, they are beginning to understand what, in effect, hit them. Interviews with many of those engineers, as well as an examination of data collected around the world during the blackout, indicate that the government exploited a devastating combination of vulnerabilities in the national infrastructure.

For all the Internet’s vaunted connectivity, the Egyptian government commanded powerful instruments of control: it owns the pipelines that carry information across the country and out into the world.

As we’ve been continually interested at mammoth in how the digital ephemera of the internet is materially instantiated — both in the infrastructures that it depends on and the global landscapes that it produces — this is a fascinating story, though it is obviously also a sinister story.

Relatedly, if we hadn’t been at the tail end of our winter hiatus when it first ran, I would have definitely linked to Andrew Blum’s brief article from the morning after Egypt turned off the internet, also at the Atlantic Technology channel, in which Blum rightly notes that the efficacy of the Egyptian government’s actions should make us more aware of the importance of who controls the physical infrastructure of the internet.  (Of course, Blum would say that! — he’s writing a book on that infrastructure.)

In A Preliminary Atlas of Gizmo Landscapes, mammoth briefly described the Google data center in The Dalles; in an excellent recent article, local The Dalles Chronicle reporter Theodoric Meyer investigates the relationship between Google and local public officials, the impact of the arrival of a second data center — for Facebook — in this rural Oregon town, and the surprising contrast between the willingness of the two internet giants to share information about these physical instantiations of the Invisible City.

[Via Alexis Madrigal at The Atlantic; via Andrew Blum, commentary on Meyer's article at Data Center Knowledge.]

If anything, Zuckerberg looks, in some distant but discernible way, like the Robert Moses of the Internet, bringing severe order to a chaotic milieu. While several efforts have been made to create more open versions of social networks, none has found much success. We are still waiting for the Jane Jacobs of online “urban planning” to appear…

Urban-analogy bonus-points aside, what I found interesting about this article was the comparison between the types of targeted advertising Google is capable of, and what Facebook is [expected to be] capable of.

Because of its unparalleled demographic information, Facebook can sell ads that will appeal only to carpenters in one small town in Vermont, or to graduates of the Harvard Business School, or to residents of Manhattan who list “opera” as an interest. The site could also provide the most highly targeted political ads in history. Google can sell ads that will appear in a particular locality, as Scott Brown showed by buying up much of the online ad space for Massachusetts during the final days of his successful bid for the Senate. With Facebook Connect, it may be possible to show ads specifically targeted to Massachusetts residents who use words such as “Irish,” “Italian,” or “black” in their profiles, or who list their religion as Catholic, Protestant, or Jewish. So far, however, advertising has only provided enough revenue for the site to barely break even, and many believe the site can only claim to be profitable because of creative accounting.

But is Facebook’s demographic information ‘unparalleled’, as Petersen claims? I think this forgets the massive amounts of data on personal preference collected by Google each time we make a search (or use Google maps, or add a website to our Google reader account, or compose an essay on Google docs, or send an email, or…). And in a way, I would think this information is far more valuable to advertisers than the personal information on Facebook. The latter is data that 1) we choose to share and 2) isn’t necessarily about us, so much as it is about the personality we choose to craft online. The data Google has is much more personal – it concerns the actions we want to take, the places we want to go, the knowledge we want to have. The Google might know us better than we know ourselves.

[link via James Fallows]

[A manhole near Halifax marks the Canadian arrival point for one of the eleven major cable lines carrying the bulk of trans-Atlantic Internet traffic; photographed by Randall Mesdon; from this excellent Wired slideshow on the physical infrastructure of the internet; the text accompanying that show is by Andrew Blum, whose forthcoming book on said infrastructure promises to be one of the most interesting books of, um, whatever year it will be released in.]

Daniel Hillis has an insightful answer to the World Question Center’s question of the year, “How has the Internet changed the way you think?”:

It seems that most people, even intelligent and well-informed people, are confused about the difference between the Internet and the Web. . . . The Web is a wonderful resource for speeding up the retrieval and dissemination of information and that, despite Wolfe’s trivialization, is no small change. Yet, the Internet is much more than just the Web. . . . By the Internet, I mean the global network of interconnected computers that enables, among other things, the Web. I would like to focus on applications that go beyond human-to-human communication. In the long run, these are applications of the Internet that will have the greatest impact on who we are and how we think.

Today, most people only recognize that they are using the Internet when they are interacting with a computer screen. They are less likely to appreciate when they are using the Internet while talking on the telephone, watching television, or flying on an airplane. Some travelers may have recently gotten a glimpse of the truth, for example, upon learning that their flights were grounded due to an Internet router failure in Salt Lake City, but for most this was just another inscrutable annoyance. Most people have long ago given up on trying to understand how technical systems work. This is a part of how the Internet is changing the way we think.

I want to be clear that I am not complaining about technical ignorance. In an Internet-connected world, it is almost impossible to keep track of how systems actually function. Your telephone conversation may be delivered over analog lines one day and by the Internet the next. Your airplane route may be chosen by a computer or a human being, or (most likely) some combination of both. Don’t bother asking, because any answer you get is likely to be wrong.

Soon, no human will know the answer. More and more decisions are made by the emergent interaction of multiple communicating systems, and these component systems themselves are constantly adapting, changing the way they work. This is the real impact of the Internet: by allowing adaptive complex systems to interoperate, the Internet has changed the way we make decisions. More and more, it is not individual humans who decide, but an entangled, adaptive network of humans and machines.

Hillis pictures human society as a cybernetic organism, in which individuals are but interchangeable constituent parts. We’re watching, in the memorable phrase from Ken MacLeod’s awkward-but-fascinating Star Fraction, the emergence of the blind watchmaker. One of the most striking things about this is the way that this reality — so intangible, so hard to understand in concrete terms — interacts with physical and visible reality; the exchange of light within servers in London and New York may determine who has enough to eat and who does not in remote villages. Increasingly, the physical infrastructure of the internet is not limited to server farms, wi-fi routers and trans-oceanic cables, though those direct infrastructures remain critical and yet poorly understood. Everything is becoming the infrastructure of the internet, or the internet is becoming the infrastructure of everything, and at either point the distinction between and order of the two collapses and becomes irrelevant.

Barring widespread societal collapse, managing patterns of emergent decision-making, which may be shaped even if they cannot be controlled, will likely become an increasingly central task for society, and so is incredibly fascinating as an architectural problem, as architecture is fundamentally more interesting when understood in terms of decisions than in terms of forms. My suspicion — though I remain interested in experiments such as the EDAW/AECOM merger which attempt to compensate for increasingly complex conditions by building increasingly complex design processes — is that this trend is another nail in the coffin of totalized design, another reason that we’ll never see a successful attempt at the sort of fully rational planning processes that last century’s modernists sought to deploy. Tomorrow’s architects have, as Lebbeus Woods recently said, “no faith in grand architectural plans to make a better world and especially not [the] best of all possible worlds”. But, as Kazys Varnelis noted in a piece mammoth recently quoted, this is more an exciting opportunity for new roles, design processes, and practices than it is an object of worry¹.

[Hillis link via Alan Jacobs' Text Patterns]