Converging end-user and corporate computing

Summary and Conclusions It appears that the convention of defining EUC in terms of the nature of the appliLcations is quickly becom- ing obsolete. In the minds of the top ISEs who participated in this study, EUC consists of a broader set of ap- plications than previously thought. In spite of the fact that, as antici- pated, a majority of the surveyed firms did include within their defini- tion of EUC the communications- intensive and database-enabled ap- plications typically associated with desktop computing, office automa- tion, and personal and work-group support systems, the findings re- ported in Table 5 and substantiated in Tables il0 and 17, suggest that EUC has e'~panded into application areas formerly thought to be the ex- clusive domain of the corporate IS function. From the perspective of these ex- ecutives, the definition of EUC now appears to include transaction pro- cessing systems; manufacturing sys- tems; ES; and IS development for local as well as corporatewide and even interorganizational systems. It is as if EUC has become a direct ex- tension of corporate computing, undifferentiated by hardware or software considerations. Moreover, it appears it is expanding its computing activities in a mirror image of the pattern of expansion taken by the traditional, corporate IS function. As the nature of the computing activities within these two domains continues to expand from opposite directions along the continuum of problem and task structure, as depicted in Figure 3, these once separate, distinct, and mutually exclusive spheres of com- puting activities are beginning to converge. Now the only distinguishing char- acteristic of EUC appears to be its reporting relationship within the organization. In the eyes of these top ISEs, computing which is outside the direct control of the IS function is EUC. Yet, as those ultimately respon- sible for the effective management of the information resources within their organizations, these same exec- utives are very concerned with the management of these end-user- controlled computing resources. Driven by the desire of users to gain control of their information re- sources and facilitated by technologi- cal advances, the EUC phenomenon expanded at a remarkable pace in the 1980s. In spite of this, it is en- couraging to see the degree to which mechanisms appear to be developing to manage the ever-expanding do- main of EUC resources within busi- ness organizations. And yet, there are indications that there is room for improvement in some organizations regarding the accomplishment of such objectives as integration and distribution, connectivity and flexi- bility, and the productivity and qual- ity of IS investments. It appears there is also room to refine and im- prove the techniques used for evalu- ating how effectively such objectives are being met. Support is one of the primary mechanisms for managing EUC. The evidence in Table 12 suggests that extensive mechanisms are already in place to support EUC in organiza- tions. Moreover, it is clear that out- sourcing such support for end users is, at the present time, a viable option for only a small percentage of these organizations. Likewise, chargeback arrangements for such support ser- vices are clearly the exception as well. Further research is needed to exam- ine which of these support mecha- nisms, and what combinations thereof, are the most effective, and when; and what roles outsourcing and/or chargeback systems might play in a successful EUC support strategy. Control is another important di- mension in the management EUC. The findings reported in Tables 14 and 15 suggest that control mecha- nisms are widespread with regard to EUC hardware issues, but less com- mon with respect to the software, application development, and secu- rity issues. This suggests there are some areas of potential organiza- tional vulnerability that may require additional attention. Although this inconsistency in degree of control is in harmony with the organizational philosophy regarding EUC as re- ported in Table 6, it does suggest the possibility, as one might expect given the four decades of experience with corporate computing, that hardware is easier to manage than software and people. Nevertheless, the findings reported in Tables 7, 8, and 9 con- vincingly point to the conclusion that the norm is simple and uniform stan- dards for EUC hardware and soft- ware. Additionally, Tables 14 and 15 suggest that progress has been made in the 1980s regarding managerial control of EUC (see [24]). Evaluation is the third primary mechanism for managing EUC. Evaluation mechanisms have the po- tential to provide management, and end users as well with the informa- tion needed to resolve the paradox inherent in simultaneously support- ing and controlling, facilitating restricting, enabling and disciplining, nurturing and regulating, fos- tering and restraining EUC. More effective evaluation may be key to finding the temporally dynamic and contextually dependent balance that every manager must find, one that best suits their individual organiza- tion, its end users, and themselves. And yet, Table 16 makes clear that the use of quantitative, objective evaluation criteria is the exception and not the rule. This is in spite of the fact that, as reported in Table 5, the majority of these participating executives report to financial execu- tives within their organizations. To rely too heavily on only subjec- tive evaluation measures is potenti- ally dangerous to the health of an organization. Purely subjective, per- sonal evaluations of IS effectiveness SO