Modeling languages, like programming languages, need to be designed if they are to be practical, usable, accepted, and of lasting value. We present principles for the design of modeling languages. To arrive at these principles, we consider the intended use of modeling languages. We conject that the principles are applicable to the development of new modeling languages, and for improving the design of existing modeling languages that have evolved, perhaps through a process of unification. The principles are illustrated and explained by several examples, drawing on object-oriented and mathematical modeling languages. 1 Introduction The key difficulty in producing quality software is specifying and designing the conceptual construct that underlies the software [2]. This conceptual construct is usually complex. Complexity is an essential difficulty that cannot be dealt with by using more powerful programming languages or tools, or by using modeling languages that abstract it away. Comple...

Finding a useful abstract representation is fundamental to solving many difficult problems in software engineering. In order to better understand how representations are actually used in key collaborative software engineering tasks, this empirical study examined all of the spoken representations of soflware behavior in 9 domain analysis sessions. It found that about 70 % of them were metaphorical, representing system behavior as physical movement of objects, as perceptual processes, or in anthropomorphic terms ascribing beliefs and desires to the system. The pattern of use of these representations indicates 1) that they were not merely temporary placeholders, but rather their use persisted even when a specialized and more formal vocabulary had been developed, and 2) the metaphoric descriptions appear to reflect actual use of metaphor, rather than just a choice of vocabulary. The use of metaphor is explained in terms d how well they serve human cognitive abilities and collaborative needs. The predominance of metaphorical representations in synchronous collaborative sessions raises important issues about the possible misleading effects of metaphorical thinking. It also raises questions about the compatibility of the spoken representations with other representations (e.g., diagrams, specification languages) that trigger, and capture the results of, the verbal collaborative work.

Abstract. As modeling becomes a crucial activity in software development the question may be asked whether currently used graphical representations are the best option to model systems efficiently. This position paper discusses the advantages of text-based modeling over commonly used graphical representations. It is inspired through the advent of new extensible development tools like Eclipse. The discussion is illustrated by showing a textual version of UML state machines as Eclipse plugins. 1 Text-based Modeling Modeling becomes an increasingly important technique for the development of complex software systems. Nowadays, three different approaches are used for modeling: textual languages, graphical languages and combinations of text and graphic. The latter usually being dominated by graphics, where text is just a supplement. Currently, the UML is the most common visual approach for modeling, as it is both widely known and assisted by many tools developed in the last decade.

End-users must write programs to control many different kinds of applications. Examples include multimedia authoring, controlling robots, defining manufacturing processes, setting up simulations, programming agents, scripting, etc. The languages used today for these tasks are usually difficult to learn and are based on professional programming languages. This is in spite of years of research highlighting the problems with these languages for novice programmers. The Natural Programming Project is developing general principles, methods, and programming language designs that will significantly reduce the amount of learning and effort needed to write programs for people who are not professional programmers. These principles are based on a thorough analysis of previous empirical studies of programmers, as well as new studies designed to discover the most natural programming paradigms. Our proposed research is to extend these results, and apply them to different domains. The result will be n...

Keywords: POP-I.B barriers to programming, POP-II.A. novice programmers, POP-III.C all cognitive dimensions, POP-III.B new language. Research in Psychology of Programming (PoP) and related fields over the past thirty years has identified many important usability issues for programming languages and tools. However, when new programming languages are designed these findings do not seem to have much impact, so popular modern languages continue to exhibit many of the same old problems. This paper reviews the progress of an ongoing project to elevate the influence of PoP on the design of a new programming language. In the context of designing a new programming language for children, we cataloged and interpreted the prior work, performed new studies where questions remained unanswered, and have focused on usability throughout the design. In addition to producing a system that is easier to learn and use than existing systems, we hope to exemplify a process that could be adopted by other language designers to improve the usability of their systems.

A graphical notation for representing Timed CSP (TCSP) specifications is presented. The notation, which integrates features from a number of existing specification languages, including Statecharts, is aimed at providing the means for more easily constructing and managing large TCSP specifications, with the intention of forming the basis for tools and a methodology for applying TCSP in the large. The graphical notation extends TCSP by allowing specifications to be both processes and arbitrary predicates, thus increasing the expressiveness and applicability of the notation. An extendible tool framework, designed for the graphical notation and to be integrated with other tools, is presented. We discuss the features of this framework, especially how it aims to support reasoning about TCSP specifications.

Abstract. Modeling systems based on semi-formal graphical formalisms, such as Statecharts, have become standard practice in the design of reactive embedded devices. Statecharts are often more intuitively understandable than equivalent textual descriptions, and their animated simulation can help to visualize complex behaviors. However, in terms of editing speed, project management, and meta-modeling, textual descriptions have advantages. As alternative to the standard WYSIWYG editing paradigm, we present an approach that is also graphical but oriented on the underlying structure of the system under development, and another approach based on a textual, dialect-independent Statechart description language. These approaches have been implemented in a prototypical modeling tool, which encompasses automatic Statechart layout. An empirical study on the usability and practicability of our Statechart editing techniques, including a Statechart layout comparison, indicates significant performance improvements in terms of editing speed and model comprehension compared to traditional modeling approaches. 1

Three experiments were conducted, exploring issues concerning visual representations of conditional structures. The first experiment originated in response to a paper by Green & Petre (1992), in which they compared subject performances on textual and visual representations of conditional structures, and which concluded that the visual programs were harder to comprehend than the textual programs. The results of Green & Petre (1992) also supported the `match-mismatch' hypothesis, which holds that `forwards' structures, ("condition implies action, with nesting", Green & Petre, 1992), facilitate access to `sequential' information (Green 1977), whilst `backwards' structures. ("action is governed by conditions, with boolean operators in place of nesting ", Green & Petre, 1992), facilitates access to `circumstantial' information, (Gilmore & Green, 1984). Experiment One in this report compared an alternative forwards visual representation to the forwards textual representation of Green & Petre...

Abstract not found

The understanding of polymorphic typechecking and type errors is poorly supported by contemporary functional language implementations. Here, a novel visualisation of functions and their types is presented based on the generation of function specific icons with graphical type representations which change dynamically as functions are applied. This visualisation has been implemented for a Standard ML subset within a visual programming environment in which function combinations are constrained by type matching. Keywords: functional programming; polymorphic typechecking; visualisation; Standard ML 1 Introduction The benefits of static parametric polymorphic typechecking based on the Hindley-Milner scheme, in minimising semantic errors while optimising component generality, are well known. Indeed, a central strength of modern functional languages lies in the ease with which they incorporate polymorphic typechecking. However, systems based on the W algorithm and its variants tend to offer po...