The management of distributed processing

1The environment of the information-systems manager has evolved significantly in the past three decades. Ever cheaper hardware, vastly more powerful systems software, and an expanding appetite for computer support on the part of the functional managers has increased the information systems manager's realm of responsibility steadily. The data processing department has grown in numbers and in influence. The major operating environment has shifted from batch to on-line. From a start in simple accounting-type record-keeping applications, the I/S manager now oversees real-time operations whose interruption can cost the corporation dearly. Something has happened to this pleasant progression recently. The Vice President, Management Information Systems, of a major corporation puts it this way: "Yes, the data processing environment has changed steadily. But somehow the last three years have been different. Everything has snowballed. All the trends have accelerated. Most significantly, some new ones (minicomputers, end-user languages, et cetera) have been added and these are threatening to move us from a nice tight, controlled operating environment (at which we've just barely arrived!) into a decentralized, dispersed era in which the managerial approaches, techniques, and tools which have been so painfully developed and implemented in the past few years appear inadequate. More important, in some cases, is that our current managerial methods appear just plain wrong for the future. It seems to me that distributed hardware, increasing user-oriented programming, soaring software costs, and soaring demands by users require, if not a whole new managerial approach, at least a thorough review of the way in which we manage in 'the distributed era'." 2This article reports on the highlights of a two and a half day conference convened to examine the new information systems environment and its implications for information systems management. Held at MIT'