Review: Tussle In Cyberspace

        The Internet has evolved considerably over the past few decades.
Initially, it was the DARPANet, and its stakeholders had a common set
of goals. This made designing and implementing it in some sense
simpler: though a certain amount of fighting about how something should
be architected or designed may have existed, the end result converged
to a solution that everyone could be happy with. However, the situation
is much different today, with a myriad of different stakeholders
participating in the Internet, all with different goals. And there
appears to be no relief of this problem in the near-future, as more
people and organizations become involved in using and wanting to define
the evolution of the Internet. This situation is what is examined by
the paper, "Tussle in Cyberspace: Defining Tomorrow's Internet".

        The paper begins by examining the different tussles and fights that
are pulling the Internet in different directions. Examples of
stakeholders that have diametrically opposed goals include that of
music downloaders and the rights holders of that music; people who want
to communicate privately and governments who want to be able to snoop
people's communications; and ISPs in competition with each other.

        Not only do the authors examine what tussles currently exist, they go
on to suggest design principles that can be applied in practice so that
the tussle can be resolved in the fairest possible way. These design
principles are as follows: design for variation in outcome; modularize
the design along tussle boundaries, so that tussles are isolated from
each other; and design for choice.

        Overall, I thought the contribution of this paper was limited. Some of
the suggestions that the paper makes seems to be driven by ideology
more than sound engineering principles. For example, the insistence on
providing choice to consumers in order to stimulate competition seems
to be based on the ideological principle of capitalism that competition
is somehow able to effect the best solution to a given problem.
Consider the following problem discussed in the paper: Consumers
currently may not be able to obtain the best quality of service
possible in wide area networks. A reason for this, the authors suggest,
is that consumers lack the ability to choose a wide area provider. The
authors go on to say, "The Internet should support a mechanism for
choice of source routing that would permit a customer to control the
path of his packets at the level of providers." The authors present an
idealogical solution, but fail to realize that it would probably never
work in practice because most users would not want this level of
choice.
Received on Sat Nov 25 2006 - 12:37:47 EST