Some notes on handling errors in APL

This paper explains the original method used at the University of Massachusetts for allowing APL programs to handle their own errors (Section 1). It then discusses the changes that were made to improve the method (Section 2). Remaining problems are discussed in Section 3.It is not the purpose of this paper to convince the reader that the facilities discussed here are needed. This has been addressed elsewhere [1, 2, 3].This paper was first presented informally at the Queen's APL Workshop, Queen's University, Kingston, Ontario, Canada (May 1976). It was revised in October 1977.