This paper [1] talks about the Internet¡¯s tomorrow. The authors try to define some design principles that can accommodate the tussles between stakeholders.

First, the authors give some observations that existing in today¡¯s Internet. The fast development of Internet has exceeded its original design goal. Hence the Internet involves more social issues rather than engineering ones. The parties of Internet are diverse, including: users, commercial ISP, private network provider, government and content provider etc.

Then the authors propose some principles that can guide designers and mechanisms to recognize the divergent interests among various parties. The highest-level principle is design for variation in outcome. The two specific ones are: modularize the design along tussles boundaries, so that one tussle does not spill over and distort unrelated issues; and design for choice, to permit the different players to express their preferences. The authors also identify some implications of these principles and the reality of tussles.

To illustrate how to apply their principles, the authors analyze some tussles¡¯ nature. There is economics tussle, including lock-in IP addressing, value pricing and Internet access competition etc. there are also trust tussle and tussles of openness. The authors suggest that these areas can benefit from the application of their principles.

The authors also discuss how to revisit old principles. They describe the future of the end to end arguments as well as the separation of policy and mechanism.

Finally, the authors summarize the lessons they learn from the failure of Qos and multicast. That is anyone who design a new enhancement for the Internet should analyze the tussles it will trigger, and the tussles in the surrounding context, and consider how they can be managed to ensure that the enhancement succeeds.

Generally, this paper gives us some insight thoughts about today¡¯s Internet from high level as well as some historic perspective. It also provides us several principles that could guide the evolution of Internet. These principles are also valuable when we meet issues involves both engineering domain and social domain. It could be better if the authors can give more specific design examples.

[1] David D. Clark, John Wroclawski, Karen R. Sollins, and Robert Braden. Tussle in Cyberspace: Defining Tomorrow's Internet. Proceedings of ACM SIGCOMM, August 2002.