Break Up the Programming Groups?

As one who has general direction over an “in-house programming group” as well a Management Science department in a large corporation, I have come to the conclusion that there is no one best organization of programming activity—even for Engineering, Research or Management Science.

Let us restrict our consideration to programming support for Management Science. There are two types of extreme environments, among countless others, that one can visualize: (1) The Management Scientists use no real data, request programming assistance for “one-time programs” and solve particular problems with no attempt to build problem-solving systems, (2) The Management Scientists make extensive use of real data and attempt to develop, perhaps in a modular fashion, problem-solving systems.

I doubt that there is any Management Science department that corresponds to either of the extremes listed above. My experience tells me, however, that there are tendencies on the part of Management Science departments to move in one or the other of the above directions.