About This Issue

Computer networks and computer communications have become an integral part of many working environments. The use of dial-up telephone lines, couplers, and modems for such communication is routine in North America. Particularly within the research community, networks such as the ARPANET, CSNET, BITNET, and the network UNIX 1 also provide valuable (and frequently rapid) means of communication. It is often easier to use a network to send a message to someone than to reach the same person by telephone, and faster to communicate via network than to send a message by mail. Much of my correspondence as Editor-in-Chief of COMPUTING SURVEYS is electronically transmitted. Networks also exist within an organization and serve to promote internal communication and resource sharing. At the University of California, Berkeley, for ex* ample, a collection of time-shared systems and high-performance workstations are connected via a local area network. Services within the network include remote login, copying of files across the network, and remote execution of commands.One machine is used to drive a laser printer, another to drive a printer-plotter, and others to provide a variety of specialized functions. To the user of these services, though, much of the distribution of services is invisible, with file transfers occurring below the user command level. Also, nodes in the network serve as gateways to external networks--the UNIX network and the ARPANET, for example, at Berkeley. Messages may be transmitted across either of these networks from any of the machines in the local network. Furthermore, users at remote sites can (and do) send messages through the network at Berkeley, using it to connect a site on one network, for example, the UNIX network, to a site on another, for example, the ARPANET. This connection is made