Click for next page ( 125


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 124
The User Era Martin B. Zimmerman * In the late 1970s Richard Nolan described a six-stage model for the evolution of computers and information systems {`Harvard Business Review, March-Apri} 19783. This mode} provides a framework for the U.S. Army's present initiatives in automation. A brief description of the model will help clarify the motivation behind these initiatives. Stage 1: Initiation. The first use of computers in most organi- zations was to solve well-defined problems and to unburden those involved in repetitive functions such as payroll ant! accounting. For the U.S. Army this occurred during the period 1956-1962. Stage 2: Expansion. This stage was characterized by an explo- sion in the use of computer technology. Hardware was king. Little was known of the software problem. Computers were placed be- hinc! glass walls and the uninitiated were paraded past the myste- rious devices in semireligious ceremonies. Companies and organi- zations depended on the computer industry for total system solutions in what were described as "turnkey" systems con- tracts. This seconc! stage can best be described by its laissez-faire *Martin B. Zimmerman is deputy assistant, deputy chief of staff for operations and plans, U.S. Army, where he serves as technical advisor on automation. 124

OCR for page 124
TlIE USER ERA 125 management and decentralized decision making. The result was "good news and bac! news." The bac! news was that clecentralized management led to the clevelopment of duplicate functional sys- tems. The Army, for example, hac3 49 automated payroll systems in 1968. The good news was that without such an expansion stage the industry would not have been able to develop the corporate professionals required for subsequent phases of growth. The un- controlled nature of the expansion naturally led to the next stage. Stage 3: Control. In this stage steering committees were es- tablished, budgetary control procedures developed, and central design agencies organized. Technological talent was concen- trated within the organization, and companies believed they could solve all their own problems. Development using in-house assets was in; contractual support was out. This stage resulted in users' frustration causer! by the inability of central development departments to solve all user problems. Stage 4: Integration. This stage is characterized by increased interactivity, data management, and management initiatives to integrate information- horizontally. The concept of developing a corporate database and sharing information among corporate staff is inherent in stage 4. A return to decentralization also marks this stage, since more systems are user-developed. Nolan, in fact, believes that stages 1-3 can be categorized as the "era of the computer," while stages 4-6 can be defined as the "era of the user." Stage 5: Data Ownership. This stage is characterized by the need to assign data ownership and a focus on solving the natural friction between the data processing professional and the increas- ingly literate ant! active user. Stage 6: Maturity. This final stage occurs when organizations are experienced enough to design a corporate planning moclel de- rivec3 from corporate databases. Today, most organizations are moving out of stage 3, central control, and into stage 4, integration from the end of the com- puter era to the beginning of the user era. Three phenomena are stimulating this move.

OCR for page 124
126 MANAGING MICROCOMPUTERS The first phenomenon can be caller! limits to the design activity. The software factories of the late 1960s and early 1970s were es- tablished to solve most if not all software clevelopment problems. They have not, however, succeecled as envisioned. As systems de- veloped and proliferated more users were trained. Over time these users requested more and more changes to the basic systems. Such changes, part of the software "maintenance" process, have used up 70 percent of the in-house programing resources of most organizations. Unfortunately, this floes not mean that 30 percent of the corporate programing staff will always be available for new functions. A hypothetical example shows why: If an organization began development of a system with 1,000 programmer/analysts, at the end of the effort 700 would remain for maintenance and 300 would be available to develop a second system. This second de- sign activity, when finished would require 210 personnel (70 per- centJ for maintenance, leaving only 90 programmer/analysts for the third system. It is obvious from this example that the number of systems any single organization can both develop ant! maintain is finite. And user frustration is the ultimate result. The second phenomenon is increased user literacy. For more than 20 years computers represented a mysterious technology, understood only by data processing professionals. Today, how- ever, computers are everywhere, and computer literacy, aicled by software products that permit unique system development with- out the neec! to understand COBOL, FORTRAN, or any other high order language, has greatly increased. The third phenomenon that has spurred movement out of the control stage is the technology explosion. The cost performance curve in logical devices has led to a sizable expansion in their use. If we continue to acquire computers in the future at the same rate as we have in the past, we will compound deliveries at 25 percent annually. Theoretically, computer growth requires an equivalent expansion in trained programmers. Tociay there are 300,000 pro- grammers. By one estimate, approximately 3 million would be needed by 1994 to maintain the current computer-to-programmer ratio. Since such expansion is unlikely, greater dependence must by placed on the user community. Fortunately, the software in- dustry has the capability to help solve the problem. The dynamics of change in the information industry and soci- ety's continuing thirst for technology-aided solutions demand re- sponses from organizational management. One user of computer

OCR for page 124
THE USER ERA 127 technology, the U.S. Department of the Army, has iclentifiec3 seven initiatives to help deal with information systems: 1. Establish a single management source. The Army is in the process of creating a new corporate "officer" who will provide central control for planning, programing, policy setting, and cle- veloping resources for all facets of information systems (the infor- mation itself, automation, and communication facilities). 2. Establish a single developer for critical nonbattlefield infor- mation systems. This concept does not conflict with the plan to involve the user community in a major portion of future systems clevelopments. It does recognize, however, that some systems are so complex that they can only be cleveloped by professional pro- grammers. Further, systems developed by the individual user be- come candidates for organizational standardization. The central developer ultimately becomes responsible for the "maintenance" of systems selected for such stanciardization. 3. Emphasize technology. In an attempt to seek homogeneous technological solutions, thus minimizing software development costs, the Army has acquired or is in the process of acquiring a variety of hardware and software that will support this effort. It includes standardized minicomputers and a family of microcom- puters; a set of hardware and supporting software that satisfies the department's worIc~wide administrative and mobilization in- formation needs; a broadband local area network within the Pen- tagon to integrate multifunctional information needs; video tele- conferencing technology to be installed initially at 18 locations, with options for additional installations; and a Data Base Man- agement System (DBMS) for use on one vendor's stanclarcI main- frames. The Army has also begun efforts to develop a software package that will combine the best features of relational, net- work, and hierarchical DBMS techniques. 4. Improve planning. Nolan articulated the need to develop horizontal corporate database systems in lieu of the present verti- cal, single-function systems. To do this requires a unique analyti- cal method. IBM, among a limited number of companies, has cle- velopecl a procedure called Information Systems Planning (ISP) that details the steps required in such an analysis. The Army has adopted the method ant! is in the process of performing ISP stud- ies worIdwicle. 5. Develop a new technologist. The melding of communication

OCR for page 124
128 MANAGING MICROCOMPUTERS and automation technologies demands inclivicluals skilled in both fields. The Army is engaged in developing programs that will train such persons. 6. Define the relative roles of users and data processing profes- sionals. The increase in the ability of users to perform their own data processing functions has led to friction between these two groups. Yet both nee(1 to play key roles. Specifically, the profes- sional shoul(l: clesign, develop, and manage the common user net- work; develop software that satisfies the needs of more than one agency or activity, such as payroll and budget; develop and en- force standards throughout the organization; establish and staff the organization's information center, which instructs users on new tools; provide fourth-generation software tools; and act as the database administrator, ensuring that every data element has a single owner who has identified procedures for its use. Finally, this professional should clevelop a procedure to provide visibility for user-developed software. The Army has developed a software clearinghouse that will function in two ways. Users can find out if software aireacly exists that meets their requirements before initiating new development. Professionals can use the clearing- house to identify software products that are canclidates for stan- ciarclization. 7. Establish standarcis. In the public sector establishing stan- darcts is the only real way to ensure both economic and effective systems development. Although the computer industry has de- fined standards in certain areas, it has not formulated standards in other areas for purely business reasons. The Army has set its own standards in five broac! areas: Languages. COBOL, FORTRAN, BASIC and Acia (for battle- field systems) are establisher! standards. SIMSCRIPT, LISP, C, ant! PASSAIC are other languages that have been used for specific and obvious functions. Data Base Management Systems (DBMSJ. The Army has a variety of DBMSs. On most of its large IBM mainframes, DATA- COM D/B is used. For the future the Army believes that the solu- tion lies in the creation of a simple but stanciard data manipula- tion language. The software behind the language could be developed to manage databases using hierarchical, network, or relational procedures. With such a solution present databases could be preserved.

OCR for page 124
THE USER ERA 129 Communication protocols. The Army is committed to the De- fense Data Network (DDN3 protocols (based on the seven-layer reference stanciard of the International Organization for Stan- ciardization). Systems that do not satisfy DDN will not be purchased. Operating systems. The Army will establish as preferred stanciarcis CP/M for 8-bit micros and both UNIX (version V) and MS DOS for 16/32 bit-micros, with UNIX preferred. Local area networks. The Army will initially use ETHERNET protocols (CSMA/CD), but recognizes that it is probably too early in the development of this technology to commit to a single stan- darcI. Without doubt, society is in the midst of major change in its use of computer technology. Clearly, the seminal force is the availabil- ity of ever more powerful technology at relatively Tow cost. Com- puter literacy has increased at an exponential rate. The pressure on the software industry for more software products for the pro- fessional has exceedec! its capability to respond. However, the industry has made specific user-oriented software products available. To shape a program that benefits from these varied forces the Army has chosen to provide top-clown planning and broac! archi- tecture; bottom-up, user-initiated development; a procedure for giving visibility to user-cleveloped software; a definition of the relative roles of the professional and the user; and a set of pre- ferred standards to increase compatibility and minimize the cost of software. The specific course of action the Army has chosen will clearly be iterative. In nineteenth-century England a band of workmen known as Luddites tried to prevent the use of labor-saving machinery by destroying it. Obviously, this was not a very successful response to technological change. Managing today's computer technology also requires the ability to implement change appropriately. Those organizations that remain static will be the Luddites of tomorrow.