I am sure that there are many among us who remember the early days of networking, when the general feeling was that anything which could be shipped over a piece of copper wire was fair game. I am equally sure that there are many among us who know of networks where such is still the case. Our own networking world, however, has reached a level of civilization sufficient to admit of a certain amount of politesse.
Correct social behavior at the network level is in many ways more important than in the so* called “real” world. In everyday life, a sufficiently gross social blunder will sometimes mean that people will not speak to you. In the networking world, a gross protocol blunder invariably means that people will not speak to you.
Mr. Protocol is an arbiter of excruciatingly correct network behavior. It is his belief that in a more perfect world, correct adherence to all the nuances of network protocol will smooth away the vicissitudes of network life. He invites your questions and comments and will attempt to deal fairly with all comers (as is his wont).
• • • •
Q: Who writes these protocols, anyway?
A: Certainly not Mr. Protocol. He has spent a lifetime assiduously absorbing all the details of the network protocols, and is far too busy to write any himself. Protocol* are written by true wizards who spend their lifetimes figuring out how these things actually work. RFC’s (Requests for Comments) are the typical means for promulgating protocol standards. Mr. Protocol thinks that issuing rules as “Requests for Comments” is the soul of gentility. Copies of any RFC which one might want to read, together with an index of all RFC’s, are available from the CIC (of course).
• • • •
Q: What’s the biggest thing I can mail?
A: How long can you talk? Mr. Protocol believes that PhoneNet is designed to allow people to exchange mail, and is distressed by the various attempts to use it to itemize the national debt. While he recognizes the need to send chip descriptions to giant fabrication systems, he regards the use of a 1200 baud connection to do so with some degree of distress. Sending a multi-megabyte chip description file through the Relay at 1200 baud is about as polite as using the public address system on a crowded Greyhound bus to discuss the internal politics of Acapulco’s city government for five solid hours • which is about how long it takes to transmit such a file. Mr. Protocol also recognizes with some dismay that the gentle folk at the CIC have not always made it perfectly clear that this is the case. He therefore wishes it to be made known that any file larger than about 100,000 bytes should be regarded as “large.” A one-megabyte data file is regarded in some circles as “moby,” but Mr. Protocol never uses such language.
It is distressing to note that there appear to be gremlins in the telephone system which search gleefully for such lengthy telephone exchanges and break them - but only after they have been underway for some hours. Mr. Protocol ventures to suggest that in a more perfect world, there must be a better way. There are a large number of unemployed Olympic torch runners at the moment. Hiring one of these and handing him or her a magnetic tape would be faster and more certain than the current scheme.
If for some reason this proves unsatisfactory, or if there is reason to believe that the transfer will actually work, or if there is some degree of urgency (!), Mr. Protocol suggests a conversation with the CIC to reach the best solution possible in the given situation. On the grounds that an invasion of thousands of mice is preferable to an invasion by an elephant, it might be suggested that a single very large message be broken down into several smaller messages and reassembled at the destination.