Bringing management skills to micros. William Godbout.

The use of microcomputers in daily business operations gives managers direct, local control over information resources that is impossible when dealing with a mainframe maintained by "high priests" in a data processing department.

The real value of microcomputers in the workplace is in the amplification of individual human effort and human skills, raising the individual's productivity by improving his creativity. Increased productivity is vital to bringing about the reindustrialization of the U.S. and its conversion from a "smokestack society" to an "information society."

One of the ways to speed this conversion to an information society is by developing a matrix of management skills within which to employ microcomputers, workstations, mainframes, communications, networks, and all the other rapidly developing electronic tools available.

The result may be the redefinition of information management. And the time is right for bringing this new management matrix to bear on microcomputers in the office. There now is a window that is no more than 24 months open which will allow for a great deal of creativity and innovation in management science and skills, especially as they apply to the use of new electronic tools. After that period, the manager who has not taken advantage of the options now available may find those options no longer exist. The "Single Guru" Trap

The first key point in applying management skills to microcomputers in the workplace is realizing that these capable machines are everyman's tool, not just the wards of the DP department. If a manager starts developing another cult of high priests around microcomputers from a user's standpoint--the true application standpoint--the effort will fial. There will be so much overhead that a real cost/benefit analysis will prove that the value of having local computing resources will be lost.

Those "everyman computers" now exist and they are getting better, faster, and more capable every day. The improvement of work output through such tools as scientist/engineer workstations, dedicated administrative workstations, computer-aided design, graphics and manufacturing computers, and many other "desktop" applications is evident, even at this beginning stage of microcomputing in the workplace.

More than one project involving the employment of microcomputers has failed because the management approach was to make it a single designer project. The single guru was put in charge of all aspects of what was essentially a multi-designer project. That stifled creativity at the start, prevented synergy and interaction along the way, and inevitably invited failure when the single guru reached his level of incompetence or even worse, somebody made him an offer he couldn't refuse and he left. This is something a manager cannot allow to happen. What Micros Cannot Do

The second key point for managers to realize is what the single microcomputer--or even groups of stand-alone or networked micros--cannot do. In management of any project, administration and reporting are key requirements. General purpose micros, running word processing software, meet those requirements very nicely.

Another general requirement is "modeling," whether it is accomplished through physical models or mathematical models and computer simulation. At certain level of modeling requirements, general purpose micros using spreadsheet kinds of software with the capability to make rapid "what-if" manipulations are very satisfactory.

The major problem in those three key areas--administrative reporting, modeling, and communications--is that no manager today can pull a plain vanilla general purpose microcomputer off the shelf, drop it on a desk, and say, "This will do your word processing, modeling, and number crunching and handle all your communications." That machine does not yet exist, and may never exist, though it is a machine that all of us manufacturers are trying to build.

What is more likely to happen is that a general case solution will involve microcomputers and workstations that will satisfy about 80% of all engineering, administrative, scientific, and business applications, at the individual's desk under his total and local control.

But the other 20% will be met by the capabilities of highly specialized machines with rigorously optimized hardware and software that makes them ideal for highly specialized control tasks, guidance tasks, database management, program development, and special types of administrative and management tasks. However, those machines are basically unsuitable for general computing.

When a task calls for dealing with a huge database, a medium-sized general purpose microcomputer will not meet the requirement. If large matrix inversions are necessary, a desktop business computer will not do the job.

The management skill enters in that 80th percentile. The best manager will know enough to ask the question, "Do we go with a piece of general purpose hardware, marry it to some software, bash it together to fit, then paint it to hide what we've done?" There are times when to answer, "yes" to that basic mangerial question is the right decision.

But there are probably many more times when bashing-to-fit and painting-to-hide are far more costly than installing special purpose hardware and software. A large econometric modeling program running on high speed, number crunching hardware is an example that comes to mind.

Summing up those two key points in the managerial question, the person charged with deciding to install microcomputers must be able to answer, first, the question, "Can we use electronic equipment as an amplifier of individual human effort in this application?" And if the answer is yes, he then must be able to decide, "Can it be done with a single microcomputer, or will we need additional special and dedicated hardware and software?" Thankfully, managers are not faced with those decisions daily.

Managers charged with bringing micros into the workplace have a great opportunity in the next two years to do it right. There is enough experience with micros to avoid the time-consuming and extremely expensive mistakes of the early '80s. There is enough solid direction from industry leaders to make very good estimates of where the future lies (16-bit machines, CP/M, and CP/M look alike operating systems, 327x communications protocols), and there is enough actual experience already in the workplace to provide a history of what works and what doesn't.

Now it is up to the managers to superimpose on that rapidly expanding body of knowledge and hardware their own styles of management expertise, which must include the avoidance of the single guru trap and educating individuals in the best ways as to the use of microcomputers.

The result? Dramatic increases in worker productivity through extension of each one's unique abilities.