Hacks and the Hardware ProblemPosted: September 17, 2014
Apple’s announcement of the iPhone 6 last week, like every other product they unveil, highlights an easily neglected truth: The algorithms, bits, clouds, streams, and light beams—the elements of today’s most inspiring technological wonders—all run on hardware, though the inscrutability of those elements, and the metaphors by which we seek to know them better, conceal the underlying materiality that makes them possible, the tangled mess of wires behind the (proverbial) television set. Apple’s devices disappear seamlessly behind their glowing rectangular screens when used, but are nonetheless the exceptions that prove a widely-accepted rule: Information should take up as little space as possible in the physical world. What used to fill shelves and file cabinets and clutter our houses and offices should part from its physical body and ascend into the Cloud. In this narrative, if not in reality, hardware is becoming less important all the time, and Apple, by providing the lightest and sleekest portals to that alternate universe of information, light, and sound, has managed to make the most socially important hardware of all.
Devices like the iPhone and iPad, in fact, form a graphical user interface for the physical system that delivers a tweet from one person’s fingertips to another’s eyes. Contrary to the popular narrative, the “cloud” that enables such light traveling at the personal level is anything but light in the aggregate. A mass of true hardware, from routers to data centers to fiberoptic cables to cell towers, is the behind-the-scenes machinery that makes the internet tick. Andrew Blum describes this condition vividly in Tubes, his exploration of the Information Age’s hidden infrastructure, and I’ve addressed it previously here, here and here.
Kazys Varnelis recently reflected upon the data center’s status as the architectural symbol of network culture. Comparing data centers to factories, the buildings that most closely embodied the Industrial Revolution, he writes that “factories served as conspicuous symbols of power and modernity” while “data centers strive for invisibility.” The traditional factory gave some indication of its function and its role in society through its size, its outward appearance, and its location (usually near or within the city). Its presence, like that of the bridge or skyscraper, was often striking and dramatic. Data centers, Varnelis writes, are the opposite, housed in the “familiar, anonymous architecture of the big shed,” situated outside of urban centers and rarely even seen, much less noticed. Not only does the reality of Information Age technology differ greatly from its user-facing mythology; the design of its various layers reinforces the myth at the expense of the reality.
Frommer’s quip captures the essence of the “hardware problem” that reaches far beyond the smartphone, although Apple’s flagship device is a good starting point in the effort to understand the broader problem. In short, the issue is this: Hardware improves at a much slower pace than the exponential improvement of the information flow it makes possible. As a result, hardware is the main bottleneck that limits what our technology can accomplish, even when it’s all on an upward trajectory. In the iPhone, Apple solved a thousand problems that we didn’t yet know we had, but the constraints on its battery life and network connectivity still limit our access to its power in a way that harshly contrasts with the enormous impact of its software. That impact, so total in one domain, does not necessarily extend to the heavier and less pliable layers of reality: Recall the boss of Lena Dunham’s character in Girls, who jokes that she won’t sue him because there’s no app for doing so.
We’re trained to underrate hardware’s importance by a constant cultural emphasis on problems with fast, scalable solutions, like social networking, search optimization, or music streaming. The thornier problems, like the lawsuit example in Girls, are either treated as constants or ignored. The iPhone’s short battery life is obvious to its myriad users, but more subtle hardware problems, invisible as the data center in the woods, are the “unknown unknowns” (to quote Donald Rumsfeld) that we don’t even know we haven’t solved because we’ve focused our attention everywhere else.
Extending the metaphor beyond consumer devices, this hardware problem pervades the present-day urban landscape. Countless apps, open data portals, and smart city agendas promise to revolutionize or save cities, while the costly infrastructure upon which those solutions depend—bridges, roads, and power grids—steadily decays or requires increasing maintenance just to keep working. “Hacking” the city, or hacking anything, is a form of arbitrage that yields something for nothing by exploiting an asymmetry in information, but a massive substrate lies below those hacks that requires true work, in the mechanical sense of the word, to improve. Hannah in Girls can’t sue her employer using an app, nor can the Port Authority rebuild the functionally obsolete Goethals Bridge with anything we could call a hack. All the brilliant attention focused on such hacks and shortcuts at the expense of the underlying work ensures that we’ll keep running increasingly sophisticated apps on perpetually dying phones.