Michael Connolly

CPS 182s Midterm Exam--Question #1.

Throughout society, there exists a common notion that cyberspace is like some vast frontier where the constraints of the "real world," both legal and natural, do not apply. One tiny example of this general view can be found in a warning that I've heard stated at least a few times: "Don't write in an email what you don't want the entire world to see." This statement derives from an understanding (in some cases entirely inferred) of the architecture of today's Internet, and given that structure, the statement itself is somewhat useful.

In his book Code, Lawrence Lessig puts forth a detailed argument that our views and understandings of the Internet are only valid relative to the architectures involved. Today, for the most part, the internet works upon a framework that validates both the warning about emails, and the more general statement that code on the Internet can't be regulated. However, Lessig suggests that the current architecture is not necessarily absolute, and that code on the Internet can and will change and be regulated. Furthermore, he suggests that open code promises to make the Internet less regulable. While I agree with these propositions, I do not believe that an Internet comprised solely of open code would be entirely unregulable since the immense desire for regulablity will inevitably cause businesses and governments to focus their sanctioning efforts on hardware. A powerfully encrypted Palladium-type system, for example, will still provide at the very least a modest means of regulation.

Businesses and governments share a tremendous desire for a regulable internet. The examples and reasons why are numerous and easy to conjure up. An unregulable Internet can often render government ordinances impotent and can even be used to circumvent sales tax. Furthermore, an Internet free of regulation can also decimate traditional business models, particularly in the entertainment industry where perfect reproductions of original works are easily and freely distributed. These facts (and there are many more) and their implications alone are sufficient enough to show why governments and businesses must look at the regulation of the Internet as imperative.

Given that the world's most powerful forces have a critical interest in the ability to regulate what things happen in cyberspace and how they happen, it is then useful to examine how this regulation will be imposed. Lessig repeatedly points out that regulation will come from code. He maintains that the imperatives of e-commerce will lead to strong regulablity, and he suggests that government will be more than willing to help out. Then in chapter eight he sheds new light on the situation by offering the hope of the open source software movement and "open code." He shows how open code could serve as a check to balance the prospect of increased regulablity. I find his suggestions to be believable, and I think that nature of open code in general is in fact less conducive to regulability. Therefore, I suspect that businesses and governments will decide that they can achieve the ends they desire more easily and more efficiently by focusing on the regulation of hardware instead of software. Therefore, for the sake of argument, and since software code itself is not the most promising avenue for regulation, it is possible to assume that the Internet will comprised solely of open code.

It is clear that there exists an overwhelming need for regulation, and it is also fairly reasonable to assume that the internet may someday be running entirely on open code, but given all this, the internet will still not be entirely devoid of regulation. The key to regulability is hardware. The barriers to composing software are small. And once composed, the barriers to distributing software are trivial. On the other hand, the barriers to building hardware are formidable, and the barriers to mass-producing a hardware product are nearly insurmountable. Regulability desires trusted methods of identification, authentication, certification, and non-repudiation, and these methods will be more solidly implemented in hardware specifically because hardware is far more fixed than software code, and, and unlike software hacks, those who create hardware are huge targets. In the words of Lessig: "Even with open code, the techniques of identity, tied to code that has been certified as compliant, will still give government plenty of power. Thus, much of the argument from part 1 survives this point about open code--if the world become certificate-rich, regulability still increases." (Lessig; Code, p. 108)

At the bottom of that same page, Lessig adds the following footnote: "Another constraint would arise if more code were burned into hardware rather than existing as software. Then, even if the code were open, it would not be modifiable." This is the basis for my position that despite the potential for an architecture of open software code, the internet will not become entirely unregulable. Hardware devices will be supplemented with components such as the "Fritz" chip, which will work to provide an extensive set of controls over individual use and also provide the means for powerful certification. Therefore it will bring at least some regulability to the internet. Software could be entirely open, but if that software does not conform to the standards demanded by the government via hardware implementations such as the Fritz chip, then it won't run. Large businesses will welcome concepts such as the Fritz chip, and they will work to ensure its proliferation and eventual universal acceptance first creating and offering products that are subservient to Fritz, and then by using pricing schemes and feature options to leverage the public into acquiescence. In sum, while it is true that open code does provide a check against the seemingly inevitable rise in regulability, open code will not make the internet entirely unregulable since both government and business will seek to reach the goal of increased control through the means of seemingly unavoidable hardware implementations.