Paper review: The Design Philosophy of the DARPA Internet Protocols

Reviewer: Kevin Hofstra

1. Is TCP/IP the most effective protocol for internetworking, or have the new expectations of the internet made the protocol “non-effective”?
2. An evaluation of the motivation behind supporting TCP/IP as the Internet standard and a look into possibilities of changing the standard.
3. It is important to realize the context in which the standards for the internet were chosen. The priorities have changed since the migration from a largely military to a mainly commercial setting. The factor of survivability, which was seen as most important in the military setting, would now be overshadowed by the importance of performance and accountability.
4. If designed today the internet protocol would assign much different weights to the set of goals when defining “effective”. Some of the goals that were lower in importance when the internet was created were less effectively met and have now become serious shortcomings. The internet protocol was very effective in meeting its original set of priorities, but was not designed to meet the priorities of today.
5. Critique the main contribution
• Significance- 3 The article is very effective in showing how the internet has met its original intention, and recognizes that it is faltering in the changing environment, but offers no real insight on what we should do about this besides the vague question of the importance of the datagram.
• Convincing- 2 The author makes a suggestion of flow monitoring or “soft state”, but says very little to back this up. He says it may well permit us to reach our goals, but offers no reassurance, and leaves it open as a type of exploration, not a solution.
• Any type of protocol solution or flow management would have to be backwards compatible, cost effective, yet more robust. A new implementation may necessitate the elimination of the old.
6. System researchers and builders should recognize that in ever changing environment of the internet priorities often change. It is not the ability to immediately satisfy all priorities, but rather the ability to adapt to meet the priorities of tomorrow. It was a large step to move from networks which had been traditionally circuit switched to the packet switched internet. The use of datagrams allows much more flexibility in dealing with unlike systems then continuous streams would allow. It is also important to realize that sessions can still be used in the datagram model by creating a virtual circuit. The issue of survivability which was once so important to the military is pretty much non-existent due to the extreme redundancy built into the topology of the internet today. Much more important today is the issue of performance. Performance was not a large issue when the internet was created, but could now be considered the most important issue. TCP/IP has adapted to fit the current day needs, but it is evident by the ordering of priorities when created that the protocol was not designed for the present day internet.