taylor@hplabsc.UUCP (Dave Taylor) (10/03/86)
This article is from <decvax!yale!LEICHTER-JERRY> and was received on Thu Oct 2 16:47:22 1986 It's been pointed out - I don't remember the name of the economist who spends a lot of time on this issue, but it's certainly not, in broad outline, an astonishing observation - that the one thing that most closely correlates with salary is the amount of time a person works unsupervised. If you think about it, this measure also correlates pretty well with intuitive ideas about "responsible" jobs. People subjected to this kind of computer monitoring are in the position of being "supervised" continuously. Outside of prisons, this kind of supervision was virtually unknown until these systems came into being. Think of what you'd consider the lowest-level, least responsible positions around - jobs like minimum-wage cash register operators. The supervisor is always around, but will normally only check on the operators every once in a while - though of course typical managers in these kinds of places cultivate the feeling among those under them that they ALWAYS know what's going on. There's a long, long history of management systems that allow unskilled, un- trusted people to do "important" kinds of things. Consider bank tellers: These are near-minimum-wage jobs. A typical bank teller can handle several times his annual salary, in cash, every day. Banks can survive this because they can draw on many years of experience in how to prevent tellers from abusing the system - partly by catching them - there are so many interlocking safeguards that it's just about impossible to walk off with money and not be detected very quickly; partly by controlling who can make decisions: A bank teller can make virtually no decisions without authorization. Despite all this, bank tellers go through the day without being watched every minute. One lesson to be learned from bank management is that the kind of micro-manage- ment that the/usr/mail/taylor.lock monitoring systems provide is not necessary. With some additional effort, it's possible to run a business efficiently without them. For the most part, they are technological "quick fixes" for problems managers are not willing to deal with effectively. A good manager should know which of his workers is productive and which is not without these kinds of intrusive techniques. A poor manager won't be helped by these techniques. There's a lot of discussion about how modern technologies are leading to decen- tralization and more democratic styles of management. But in fact, the same technologies - high-bandwidth communication, tons of computational power - that allow for decentralization also allow for just the opposite: Centralized con- trol. Both trends are visible. Consider the attempted rescue of the hostages in Iran a couple of years back. One aspect that's been discussed is the close control over the operation: President Carter was in direct contact with the men on the ground, making decisions and giving orders from Washington. Every analyst I've read on this says it was a bad idea. Micro-managing almost always is. Low-level, continuous monitoring is almost always a poor idea. I question the ethics of those involved. I also question whether they are going to get any significant payoff, except over the short run. -- Jerry -------