In this paper, we describe the how active learning techniques have been successfully applied in a CS 2 course.

Multiparadigm programming languages have been envisioned as a vehicle for constructing large and complex heterogeneous systems, such as a stock market exchange or a telecommunications network. General-purpose multiparadigm languages, as opposed to hybrid multiparadigm languages, embody several prevalent programming paradigms without being motivated by a single problem. One such language is Leda, which embodies the foundational paradigms of imperative, functional, logic, and object-oriented programming. We explore aspects of solving complex problems using Leda, in order to illustrate the benefits of using a multiparadigm language in expressing solutions to complex systems. We claim that general-purpose multiparadigm programming languages like Leda greatly expedite solutions to a variety of complex problems.

Abstract Object‐oriented programming languages are noted for their ability to allow users quickly to construct large software systems. They achieve this ability by allowing the programmer to concentrate on what it is they want to do, ignoring details of how that functionality is achieved. Such characteristics should make the object‐oriented style of programming attractive to casual or novice computer users, indeed one of the best known object‐oriented programming languages, Smalltalk, was initially designed with just such novice users in mind. The Unix operating system is widely regarded as a good environment for doing software development. Nevertheless, its large number of often terse and cryptic commands make it difficult for novice or casual users to use. In this paper we argue that one solution is to create an object‐oriented command interpreter, or shell, between the user and the underlying system. This paper reports on the construction of one such prototype shell, and the lessons we have learned from the use of this system.

Although vector processors have been available for over a decade, the software necessary to make effective use of these facilities has generally not been available.In this paper the design of a compiler for the programming language APL, which produces code that allows most operations to be performed in parallel, is described.This design is based on a similar sequential compiler currently in use.Both compilers use a technique called dragthrough to reduce the size of intermediate values generated during the course of execution.

In testing for program correctness, the standard approaches [11,13,21,22,23,24,34] have centered on finding data D, a finite subset of all possible inputs to program P, such that