Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Managed Innovation Controlling End-User Computing in the Federal Government Ray Kline* In 1982 the General Services Administration (GSA), respond- ing to the long-awaitect microcomputer revolution, convened a small group of knowle(lgeable people in government to help cleter- mine what the federal government's policies should be in this emerging area. The group met every two or three weeks to Took at various aspects of the problem and to see what might be done. The quandary it faced was change versus control the government's need to be responsive to technological change and the need to maintain management control. In the area of computer technology, of course, change is essen- tial and inevitable. Decades ago the federal government was in the vanguard of automated data processing and related technolo- gies. Today it lags behind. Studies during the Ford administra- tion (1974-1976) placed government about 10 years behind the private sector. By 1983 the Grace Commission reported that the gap was only 6.7 years. It is possible that by 1990 we will have caught up with or even surpassed the private sector. However, the driving force to initiate change is not merely to keep up with the private sector but to improve the efficiency of government and to increase the productivity of the federal worker. President Reagan has impressed upon his entire cabinet the need to cut gov *Ray Kline is acting administrator of the U.S. General Services Administration. 52
MANAGED INTO VA TION 53 ernment spending, to increase productivity dramatically, and to perform the functions of government at a lower cost with fewer people. At the same time there is the neec! to maintain control. And in the federal bureaucracy controls are imposed somewhat differ- ently than in the private sector. The government is not, as one often reads, just another large corporation. There is a General Accounting Office (GAOl, congressional legislation, a board of di- rectors (Congress) not always in agreement with the chief execu- tive officer (the President. All of these players must be consic3- erecI when developing effective controls for microcomputers in the federal government. Compounding the problem, the GAO has issuer! an audit report stating that a runaway condition exists in the use of personal computers in the federal bureaucracy. This situation has arisen even sooner than originally preclicted. Federal employees aren't waiting for guidance. They are bringing in their own equipment and putting it to work. In my trips around! the country I have seen the growing number of computers, and although the numbers vary in different departments no one can deny that there is a lot of activity. In the budget for fiscal year 1985, for example, new auto- mation requirements are in the hundreds of millions of dollars. These requirements serve to indicate what employees want to clo with personal computers as well as with some of the larger sys- tems. Some degree of control in this process is imperative. With these two considerations in mind the need to make change and the need to control it GSA's microcomputer assess- ment group performed its work and in June 1983 released a re- port, Managing End-User Computing in the Federal Govern- ment. The assessment group identified three components of end-user computing: mode, users, and technologies (Table 1~. From a management perspective the report concludes that the integrated mode requires the most attention. In reviewing users and~technologies the assessment group observed that in the past heavy emphasis was placer! on clerical applications. Now the tech- nology has useful applications throughout the work force. There- fore, although the group focused on personal computers, other technologies were considered, along with the broad range of po- tential users. The group considered two possible courses of action to deal with the growth of personal computers in the fecleral government.
54 TABLE ~ Components of End-User Computing MANAGING MICROCOMPUTERS Mode Technologies Users Stand-alone Word processing Clericals Integrated Optical character recognition Office professionals Electronic mail Scientists Micrographics Facsimile Personal computers Dictation systems Terminals and networks Document storage/retrieval Graphics Distribution data processing One was to institute a moratorium on the purchase of computers until things could be sorter! out. The other was to allow activity to continue and try to provide an interim environment for learning. Needless to say, judgment ran in favor of the latter. The group then began to look at what kinds and levels of con- trols and supports wouIci be helpful in creating this learning envi- ronment. In the process members of GSA's Advisory Board, com- posed mainly of vice-presidents of private sector companies, were asked about the controls they imposed on their own end-user com- puting. Approximately half said they hac! tight controls. The other half said they had rather loose controls, and some wished they conic begin again and include more control features. The assessment group began to shape a new microcomputer management environment around the idea of sticks and carrots. Control was the stick; support or encouragement was the carrot. Sticks include the kinds of standards that shouIc! be imposed on people and organizations and the levels at which such restrictions apply. The group identifier! three main types of carrots procure- ment vehicles, support structures, and education tools. In each area the group concluded that more ant! better carrots were needed. For example, there was room for much improvement and streamlining of the federal procurement system. Support struc- tures in departments and agencies were not at the levels they should be. Improvement was needler! in both managerial support
MANAGED INNOVATION 55 of changes and technical support to new users. More and better education tools were required at all levels. The most obvious needs in the education area were for better training plans, refer- ence tools such as buyer's guides, and a clearinghouse for infor- mation on what other people were doing. From a broac! perspective, what was needled cluring this transi- tional learning period! was an environment for growth at a pace in keeping with each organization's ability to effectively apply a new technology. The assessment group evolved several broact goals to underlie its management recommendations for this tran- sition period: · Create a friendly but well-managed environment. · Don't scare people with too many control systems. · Encourage the use of technology to improve productivity. Improved productivity must be the end result because, without the private sector's bottom line, productivity is the only way to jucige whether efforts are truly cost-effective. Beware of the run- away train syndrome, which could result one day in a pile of un- productive equipment and a situation that is out of control. The group called its approach the "managed innovation pro- gram." Its objective was to meet the needs of a one- or two-year transition period. During this time microcomputer users couIct cle- velop an understanding of what is going on in the marketplace, and the market itself might stabilize somewhat. There are two parts to the managed innovation program: a set of 13 government- wide initiatives for which GSA is responsible and 12 individual agency initiatives. These initiatives are summarized in Table 2. Governmentwicle initiatives fall into four categories of purpose: policy, agency assistance, education, and in-house learning. These served as common denominators for the recommendations and suggestions macle by the assessment group. The group hac! to evolve policy not for the Tong term, but for the transition period. Thus, it recognized the need for a fact-finding facility to stay current with the activities of different depart- ments and agencies, many of which were quite advancecI. Local clata-network policy was another area that needed consideration. An interagency group is now working to develop a deeper under- standing in this area that may evolve into policy guiclelines. A final policy area involved revision of the regulations ant! the
56 ·_' C) be ¢ ·_1 ·¢ ,~ o 1 U) · - Ct ._ ·= be o o . - o - ¢ Cal ¢ Ed .~> . - I_ o C 3 ~3 ~ ~> ~ U) o ~ ~ ~ =' ~ . . ~ ~ a ~ ~ ~'5 ~ ~ ~ ~ I, ~ ~ be,=~ D O O ~ I,, ~ _~ ~ ~C v ~ ~ ~ ~ .. . . . . . . . . . . Cal CO ~ ~ IS: ~ 00 ~ O ~Cal AD .= C) .~m .0 o ~O PA G ~.a _ ~ 0 ~ ~ ~ ~ 3 O ~ ~ 3 C~ ~ ~ O ~ ~ ~0 3 ~ ~0 A,, ~bD b.O.~, C ~ GAPE Cam ~0 ~ ~ ~ v ~ .. . . . ........ Cal CO ~ ~ CO ~00 ~O ~CalCO ~AD ~· - 4 O · - 0 ~=0
MANAGED INNOVATION 57 guidelines themselves to make them more responsive to actual needs. Governmentwide initiatives shouIcT help people get started. The managed innovation program calls for agency assistance in two main areas. One involves obtaining and maintaining equip- ment, providing procurement vehicles for equipment and soft- ware and cleveloping a nationwide cluster maintenance contract to streamline procurement modes. The other involves helping the user by providing buying and proceclural guides. Governmentwide initiatives have already begun in the area of education. In 1983 senior executives from federal agencies met to discuss problems and make suggestions about how GSA could improve its central management performance and about what in- dividual agencies could do. In the area of in-house learning there have been some notable achievements within individual agencies, such as the Department of Agriculture's graduate school program, the GSA Interagency Training Center, and other departmental efforts. New initiatives are needed. GSA conducted its own encI-user pilot project, one of the recommenclec3 initiatives in the in-house learning category. Initiatives of inclividual agencies form the second part of the managed innovation program. Although specifics will differ ac- cording to the needs of each agency or department, initiatives can be grouped in four broad areas: policy and planning review, pro- cess, use assistance, and education. In the policy and planning area the top management of each agency must assume responsibility. Further, the organization needs to understand top management's position on microcompu- ters to move through the transition phase and into the future. Strategic planning for microcomputers should address both verti- cal (top-down) and horizontal perspectives on the organization. Planning must also take into account the need for agencywide data regulations and periodic reviews to determine that resources are being used properly. Management must make sure that the public's money is being well spent. Agencies must move away from the earlier management mentality, which was clominated by the large, mainframe, automatic data processing departments. They must avoid the trap of believing that the only way to control the end-user computing environment is to employ the kind of con- trols imposed on the large systems in the 1960s. In its report the assessment group recommends that computing be placed in the
58 MAA[AG1NG MICROCOMPUTERS hands of line managers, giving them the resources they need to do their particular jobs. In the areas of user assistance and education recommended ini- tiatives include encouraging people with common problems to get together, provi(ling ways to help them, and cleveloping computer- literacy plans that meet individual and agency needs. After the assessment group completed its report GSA began to apply its principles through an end-user pilot project. Top man- agement informed GSA's 30,000 employees around the country that it was interested in their ideas on end-user computing. To evaluate the end-user computing proposals that resulted, the agency formed a two-person review panel macle up of one member who understood the world of encI-user computing ant! the other who understood the breadth of GSA's activities. Proposals hac! to include justifications not exceeding one page, a brief description of the application, and an estimate of the gain in productivity. GSA received over 100 proposals, and about half were approved. The people involved in the proposals that have been implemented are convincer! of the value of these tools to the performance of their jobs. To help them use the approved applications, GSA set up a three-member technical support group. The agency also set up a steering committee and authorized some broader applica- tions throughout the organization, such as automating many of the General Counsel's activities and tying in with the White House's system of electronic mail. The end-user pilot project has resulted in some real improve- ments. There has been a definite gain in productivity. Reports are coming out in a fraction of the time that had previously been re- quired. Capabilities now exist for "what-if" analysis that were not previously available. Finally, the quality and accuracy of the work product have improved. Perhaps just as important, the project also identified areas that need more attention. Proposal ideas came, by and large, from peo- ple directly involved with encI-user applications, and reflected their considerable knowledge. But the proposals also reflected the fact that people had obviously been working and learning pretty much on their own. They pointed up the need for a strong, effec- tive training component, inclucling a survey course of what an encl-user computer can do and a hands-on training session. The agency concluded that, in terms of training time, there should be at least a two-to-one ratio of hands-on training versus lecture. To
MANAGED INNOVA TION 59 limit training to broad brush exposure is not enough. All 50 peo- ple involves! in the pilot project stressed the need for much more initial hancis-on training in order to achieve competency in a much shorter period of time. In the area of technical support GSA found it indispensable to have an in-house group of people who are not vendors to advise and help users by answering questions, many of them procedural, on how to get things going. Such in-house capability eliminates the problem of a vendor walking off after delivering equipment or software and leaving users to their own devices as problems arise. In the area of procurement GSA found some isolated cases in which computers were being chosen on the basis of price alone, with no consideration for training and support components. In certain cases the hardware came from vendors hundreds of miles away, making on-site support extremely difficult. Such experi- ences emphasized the importance of procurement procedures that focus not merely on the hardware price alone, but include the total package of hardware, software, training, and technical support. GSA is now applying the knowledge gained from the pilot proj- ect to its microcomputer management system. The agency has also moved into requirements contracting to cover its end-user computing needs. GSA learner! through this enc3-user computing experiment that there are four or five applications that cover 85 percent of the needs. The requirements package being provided for the acquisition of hardware will meet the needs of those pre- dominant applications. GSA has made considerable progress in dealing with the prob- lems of end-user computing in the federal government. However, there is still much to learn. Looking to the future, the agency ex- pects the current initiatives of the managed innovation program to evolve into a management pattern that will benefit the federal government and ultimately the taxpayer in a cost-effective way.
