An Experimental Investigation of Comparative Operating Efficiency of a 4GL (Focus) and COBOL in a Workstation Environment

Journal of Microcomputer Systems Management 26 Manuscript originally submitted July 18, 1989; Revised October 11, 1989; Accepted January 16, 1990 for publication. One can hardly browse through the major MIS/DP journals today without finding frequent reference to the term Fourth Generation Languages (4GLs). Commercial 4GLs have been one of the most significant software technological innovations in the 1980s for MIS organizations (Bernknoph, 1985; Martin, 1985). In contrast to the conventional procedural languages such as COBOL, FORTRAN, PL/I, etc. commonly referred to as the Third Generation Languages (3GLs) 4GLs are generally more “English-like” in syntax, and are essentially nonprocedural in nature. With these languages programmers typically need merely to specify what the result is to be, not how to obtain it. The programmer is not concerned with the low-level procedural details of how the system is to accomplish the computing task (as is the case with Current MIS practitioner literature commonly asserts that commercial 4GLs should completely replace 3GLs in the area of business information processing. One important question in this regard is whether the processing speed of applications written in 4GLs is comparable to that of their 3GL counterparts. This paper examines this central question, presenting the results of a controlled experiment that compared the processing efficiency of FOCUS and COBOL in an IBM PC/AT environment. The experimental results indicated that FOCUS programs, on the average, execute slower compared to their COBOL counterparts. Operating efficiency of applications is considered to be an important contributory factor toward users’ perception of system quality. Thus, given the current state of 4GL technology, the assertion that the commercial 4GLs should completely replace COBOL may be premature. For those applications where efficiency is an important issue, the traditional 3GLs such as COBOL may still be preferred to commercial 4GLs such as FOCUS. AN EXPERIMENTAL INVESTIGATION OF COMPARATIVE OPERATING EFFICIENCY OF A 4GL (FOCUS) AND COBOL IN A WORKSTATION ENVIRONMENT