Paper Review: A Protocol for Packet Network Intercommunication

Reviewer: Jie Zhou


1.      This paper describes a protocol that supports the sharing of resources that exist in different packet switching networks.


2.      The paper has solved a new problem. At the time when the paper was published, there existed many kinds of networks, which implemented packet switching in different ways.  It required a common protocol to share resources among these networks. Even now, the situation of networks is not reversed. Actually, more kinds of networks, such as wireless, have been developed in recent years. In future, it is very likely that networks will be more diversified. So, the problem will still be very important.


3.      Main idea: the protocol provides for variation in individual network packet sizes, transmission failures, sequencing, flow control, and the creation and destruction of process-to-process associations.


4.      I rate the significance of this paper on 5. The proposal of TCP laid the base for the development of Internet.


The paper is quite convincing. It describes the protocol by arguments and analysis. For many issues, it gives detailed explanation of the process and uses many examples to illustrate it.


The most important limitation is there is no prototype of TCP provided. As a consequence, no experiments have been done so that we cannot evaluate its performance in reality.


5.      One important lesson from the paper is: more work should be done at the interfaces, such as the GATEWAY, so as to hide the complication inside the individual implementation.


The question that the paper leaves open is how to produce a detailed specification of the protocol.