The effect of the graphical programming language FPL (First Programming Language) on the occurrence of conceptual bugs in programs written by novices was studied. The type and location for each bug, and the frequency for each type were all recorded following procedures developed in an earlier Yale University study of novice Pascal programming. The findings were compared with those of the earlier study, and suggest that FPL may help beginning programmers avoid some common conceptual errors in their programming.

No abstract available.

Instructional use of computers is now common at many secondary schools and in elementary schools. The computer is recognized as a new and important educational medium, as a source of change in existing curriculum, as a new curriculum area. Thus there is growing awareness that pre-service elementary and secondary school teachers need some computer education. What are colleges of education doing to meet this challenge? A panel of faculty from several colleges of education will present their views and insights. They will point out successes, both at their own schools and on a national scale. They will also point out what still remains to be done. There will be time for input from the audience.

For both practical and transcendent reasons, large numbers of school personnel should be trained in computing as soon as possible. Given the educational backgrounds of most such personnel, that training will have to be largely remedial from the computer scientist's point of view. At the same time, to be attractive to such personnel, the training has to carry a graduate label. This paper suggests what these practical and transcendent reasons are and briefly describes an example of the sort of training program needed.

As the field of data processing grows, it becomes increasingly important for the computer professional to keep abreast of new developments. Because one of the most significant of these new developments is structured programming, we decided to use knowledge of structured programming concepts as a measure of how up-to-date a programmer is. Thus, we undertook to answer the following questions:

Because of the rapid growth of computing and its influence in schools and community colleges, we find we must teach a growing number of graduate teachers and administrators how to program. Because we firmly believe that the only programming worth doing is well-structured programming, this is the sort of programming we want to teach these graduates to do. Because BASIC, FORTRAN, and COBOL are so widely implemented around the world, we know that one or more of these languages will be the ones most of our students will end up programming in, after they leave Columbia. Because such languages are not particularly well-suited for teaching structured programming to beginners, we decided to develop an indirect approach to this goal, one which would make the learning of these languages the learning of second languages in a second phase of learning to program. The approach which we have been developing, including the major components of a first programming language - FPL, is suggested in the remainder of this paper. For us, it has proven to be an indirect but effective way to teach raw beginners how to write well-structured programs in archaic, ill-designed, high-level real languages.