Semiotic engineering principles for evaluating end-user programming environments

Abstract

End user programming (EUP) environments are difficult to evaluate empirically. Most users do not engage in programming, and those who do are often discouraged by the complexity of programming tasks. Often the difficulties arise from the programming languages in which users are expected to express themselves. But there are other difficulties associated with designing extensions and adjustments to artifacts that have been originally designed by others. This paper characterizes EUP as a semiotic design process, and presents two principles that can be used to illustrate the distinctions between the various kinds of techniques and approaches proposed in this field. The principles support a preliminary theoretical model of EUP and should thus facilitate the definition and interpretation of empirical evaluation studies. They also define some specific semiotic qualifications that more usable and applicable EUP languages could be expected to have.