one or more bugs even after years of successful operation. The bug may be in a subroutine that is used only infrequently or it may be in a very complex set of calculations where it is difficult to recognize an error. (You may say, but isn't the program tested and debugged when it is written? ln theory, yes. But where there are millions of possible paths through a program it generally is checked by another computerprogram - and what if there is an error in that one?) What are the consequences of such errors? In a word, scary! They're unpredictable, potentially very damaging, and, even after they occur, may not be correctly identified or even recognized. Not only that, but they could well cause physical harm as well as the more common inconvenience and financial loss. For example, the computerized control system on BART failed to take into account the possibility of a certain type of malfunction in a sensor. Result - a very damaging crash. What about the computer diagnostic systems used in virtually every large pathology laboratory? Let's say you return from Zaire feeling poorly. You go into a large hospital and the doctor orders blood tests, etc. The computer then diagnoses your illness - incorrectly. You're then treated for the wrong disease and you die four days later. Not a very pleasant consequence of a computer bug. Not only that, who would ever suspect the computer program? So the error goes on for years unchecked. Other Problems The problems mentioned above are only a few of the many threats to society facilitated by the computer. We } also have possible threats to national security when compromised timesharing systems are used for advanced defense or weapons research. We have the invasion of privacy which goes hand in hand with larger and more comprehensive data banks (criminal, medical, credit, consumer opinion). We have pranks and practical jokes such as the McDonald's contest in Southern California and the student who insured the life of his goldfish. We also have the threat of becoming excessively dependent on the computer to do things that we used to do via experiment. Thus an engineer or scientist today may be deprived of the practical and valuable learning experience which results from an experiment which fails. Are these undesirable problems the fault of the computer? Emphatically, NO. They occur more easily because the computer exists. The impersonal, mechanical computer is a convenient scapegoat to blame when the real problem may lie with the input data, the system's design, or the execution of the program. We must recognize that as the computer extends our intellect it also extends our capability and speed of making errors, of committing crime, of forcing change, of invading privacy, and of causing inconvenience. We must also recognize that along with the computer's tremendous power to facilitate beneficial advances for society is an equal power to cause problems. David H. Ahl [image] IT IS WISE TO RECALL THAT AMIDST THE FLOWERS BE THORNS