The Design Philosophy of the DARPA Internet Protocols
David D. Clark
M.I.T. Laboratory for Computer Science

This paper introduces the goals of the Internet protocols¡¯ design and explains how the protocols achieve these goals. The author focuses on three points: survivability, types of service, and varieties of networks. According to the author¡¯s design philosophy, these issues can be well addressed. However, it is not enough. Lots of problems become or would become apparent due to the incomprehensive thoughts. So it is not recommended for publishing.

The origin of the Internet design is for military, but not for commercial. From this point, it is obvious that the public security cannot be considered when it was designed. The networks for military are limited in a comparably small scope. The communication channels are protected. It is unlikely for commercial use. A variety of commercial companies or individual persons need a safe environment to communicate with each other. Unfortunately, the Internet protocols don¡¯t provide us with this critical feature. Emails, telnet session, file transfer, and so on, are all subject to eavesdropping or interception. No one expect his credit card number is known by a malicious hacker. Since the Internet protocols don¡¯t intuitively protect users¡¯ data, people have to introduce a lot of encryption schema to protect their information. The result leads that a variety of security clients are invented and normally they are not compatible to each other. Today there¡¯s no a common email client with security feature for public use. For instance, PGP email clients are not compatible with SMIME email clients. People are still unable to send an encrypted email to anyone who can use another kind of email client to read it. If the issue is invoked at the beginning, the current situation will be changed a lot. Fortunately, network researchers have noticed it. The next generation Internet protocols will not miss the opportunity to eliminate this trouble.

Another key point not mentioned in this paper is robustness. The current Internet protocols are not strong so that it is easy to be attacked. A very recent example occurred two or three days ago. A worm variant attacked nearly the entire Internet. In a quite short time, millions of, billions of dollars were lost. Can you imagine that a single person is able to corrupt parts of the Internet, even the entire one, with several kilo-byte codes? That¡¯s true. The Internet is so vulnerable to attack. With the popularity of the Internet and people¡¯s dependence on the network, this kind of loss might grow faster in future attacks. If the author realized the importance of robustness in the design philosophy, the loss would be definitely reduced. Therefore, how to design a robust Internet is another critical issue that researchers are facing.

Only from these two facts, it is shown that the author¡¯s design philosophy is not deliberate. The protocols due to this philosophy might be dangerous. So this paper is strongly not recommended.