Monday, April 6th, 2009 at 4:46 pm (last modified 4/10 at 7:47 pm)
[My apologies for the length, but there was a lot of background to cover. If you’re already familiar with “The Future of the Internet — And How to Stop It” and some of the responses to it, you can skip or skim the first 4-5 paragraphs.]
Last week, I was digging around the CITP’s YouTube channel, and was pleasantly surprised to see that Jonathan Zittrain gave a talk about his book, “The Future of The Internet — And How to Stop It” last year. His provocative title doesn’t disappoint - Zittrain presents us with a bleak yet frighteningly-plausible view of how the computing world could look in a few years if we don’t do something about it.
If you’re unfamiliar with Zittrain’s book, I strongly recommend that you check out his talk. It’s an hour long, but Zittrain is a tremendously entertaining lecturer, and is capable of developing his thesis much more fully than I can here. Nonetheless, I’ll attempt to summarize the crux of his argument.
Zittrain begins by classifying technologies into two categories - sterile and generative. A classic sterile technology is AOL or Compuserve, since their walled-gardens restricted users to only what was given to them - they could not create anything new. Even early computers like the rented IBM mainframes of the 60’s are sterile, despite technically being universal turing machines, since only a team of trained IBM technicians could program them. This meant that the average user is forced to use whatever IBM gives them. With the rise of the PC, and later the Internet, generative technologies took center stage. With a PC anyone with some coding skill could get their machine to do anything they wanted, even things the original manufacturer hadn’t dreamed of. Zittrain points to the case of the Apple II and VisCalc, the world’s first widely-used spreadsheet program. Soon after the launch of the company’s first personal computer, Apple executives noticed that they had an inexplicable spike in sales to businesses. The reason? Some third-party had released the world’s first “killer app” - VisCalc - without the knowledge or involvement of the original designers of the Apple II.
Zittrain then points out a troubling trend - sterile technologies are slowly beginning to creep back into the mainstream. He argues that a fear of the power of generativity will fuel a backlash against the openness of these technologies. Generativity opens up the possibility for some particularly well-written piece of malware to insidiously gain control of millions of computers for use in a coordinated attack. His primary example of this trend was the iPhone, which presents the user with a limited menu of the applications Apple deemed appropriate for you to run, eerily reminiscent of Compuserve’s welcome screen. This example has lost some of its power with the release of the SDK, but some of the sterility remains - you’re technically only allowed to install Apple-approved programs from the App Store.
However, not everyone agreed with Zittrain’s ominous prediction of an impending crackdown on generative technologies. Some, like the Technology Liberation Front’s Adam Thierer, argue that the trend is actually towards “hybrid” technologies that allow for generativity within certain confines (like the current state of the iPhone). He also asserts that this trend is a natural reaction of the market to what consumers want, and is therefore not a bad thing. Sure, tech geeks like us might want the ability to install arbitrary executables on our phones so that we can run an SSH client to remotely restart our webservers over cereal, but for the vast majority of consumers that kind of openness is overkill and potentially confusing. Worse yet, it exposes them to serious vulnerabilities, since not everyone can tell that a pop-up window that’s styled to look like a system message is most likely trying to infect you with a virus of some sort.
For the most part, I agree with Thierer’s line of reasoning. But Zittrain still has me worried, and here’s why. Even though users’ desire for simplicity and security is a big driving force for the trend back towards sterile technologies, in some cases failed policy efforts by our own government are to blame. And that is most certainly not good, since it stifles growth and competition in the marketplace, ultimately hurting consumers.
Fortunately, recent news has provided me with an abundance of examples to make this point. I’ll begin with the most obvious. A few days ago, the Cybersecurity Act of 2009 was introduced on the Senate floor. This bill is a product of exactly the kind of paranoia that Zittrain talks about. According to the Center for Democracy and Technology, the bill would give the federal government unprecedented authority “over private sector Internet services, applications and software,” such as the ability to shut down any Internet traffic during “emergencies.” It also gives the Commerce Department the power to “override all privacy laws” in order to investigate cybersecurity threats. The CDT dryly observes that this “threaten[s] innovation, freedom, and privacy.” Aside from it’s obvious fatal flaws with regard to civil liberties, this bill would have a massive sterilizing effect on the entire Internet. I’m not so sure that I’d want to let Google store my email if the Commerce department could read it “without regard to any provision of law, regulation, rule, or policy restricting such access”(yes, that really is a direct quote from the bill).
While the Cybersecurity Act would be an obvious misstep by the government, there are many others that have a more subtle, yet still suffocating effect on technical growth. Recently, the Second Circuit Court of Appeals ruled that Google could be sued under trademark law for selling ads on a keyworded on a company’s trademark as part of it’s AdWords product. Although this ruling still requires a company to establish that selling such ads contributes to consumer confusion, the EFF notes that the prohibitively expensive costs of defending against such litigation could make Google hesitant to sell any trademarked terms as ad keywords at all. Why should we care, you ask? Aside from restricting Google’s ability to harness the generative power of the Internet to do business, it would also limit the ability of well-intentioned advocacy groups to raise awareness about their cause, since they could no longer do things like buy ads on the keyword “Wal-Mart” in order to take issue with the company’s labor practices.
My last example is the technological equivalent of throwing someone into a thermonuclear reactor core. DRM has managed to stifle innovation like no other force in the past decade. It is the embodiment of music industry executives’ clinical phobia of generative technologies, and it’s enabled by a terrible piece of legislation, the Digital Millennium Copyright Act. To quote our own Tim Lee (who is, incidentally, cited in the Wikipedia article about the Act):
The DMCA is anti-competitive. It gives copyright holders — and the technology companies that distribute their content — the legal power to create closed technology platforms and exclude competitors from interoperating with them. Worst of all, DRM technologies are clumsy and ineffective; they inconvenience legitimate users but do little to stop pirates.
Industry executives are slowly beginning to realize that DRM is not the answer to their problems, but in many other arenas, fears of generative technologies still fuel bad policy. There are already signs of this as we begin to decide what the role of government regulation should be in preserving net neutrality, particularly for mobile wireless networks. As lawmakers debate policy that will define the future of the Internet, they should be sure to take Zittrain’s warning seriously so that they do not contribute to this sterilizing trend.