Review: Tussle in Cyberspace: Defining Tomorrow's Internet

Review: Tussle in Cyberspace: Defining Tomorrow's Internet

Reviewer: Di Niu

This paper is quite interesting in that the author of the famous end-
to-end argument revisited the design of Internet with regard to
network economics and attempted to offer some new design guidelines.
Although the Internet was originally engineered to yield reliable and
predictable performances, yet one important observation in today's
Internet is that there are powerful players that make up the Internet
milieu with interests directly at odds with each other. Such
conflicts among different users of the Internet are called Tussles in
Cyberspace in this paper.

To motivate designs that accommodate the characteristic of
conflicting interests among different parties of users, the paper
proposed several design principles. First, it was proposed to
modularize along tussle boundaries; functions that are within a
tussle space should be logically separated from functions outside of
that space. DNS was offered as an example to illustrate this idea.
Second, it is suggested that protocols be devised to permit all the
parties to express choice, and in particular to express preference
about which other parties they interact with. However, in my view,
this could be merely a naive idea, since this will add much burden to
users, who actually want to enjoy the service without efforts.

Several facets of tussle spaces were discussed in the paper,
including issues concerning economics and trust. With regard to the
issue of economics, it is advocated that the consumer should have the
choice to move from ISP to ISP. GIven that, the Internet should
incorporate mechanisms that make it easy for a host to change
addresses and to have and use multiple addresses. However, it is much
easier to be said than to be done. It is also proposed that the
Internet should support a mechanism for choice of source routing that
would permit a customer to control the path of his pakects at the
level of providers. However, such an idea is too complex to be
implemented. With regard to the trust problem, for example, it is
suggested that firewalls of the future must be designed so that they
apply constraints based on who is communicating, as well as what
protocols are being run and where in the network the parties are.

In all, the paper provided a whole bunch of new design ideas based on
the conflicting interests among users observed in today's Internet.
While considering conflicting parties and tussles is beneficial to
the design of the Internet architecture, it is not clear at all
whether the ideas proposed here are viable in real world. In fact, it
will be very likely that they will add more to the hairiness of
engineering than the performance enhancement of the Internet.
Received on Tue Nov 28 2006 - 00:27:18 EST