Gürcan Gülesir

Experimental evaluation of a tool for the verification and transformation of source code in event-driven systems

In event-driven systems, separating the reactive part of software (i.e., event-driven control) from the non-reactive part is a common design practice. The reactive part is typically structured according to the states and transitions of a system, whereas the non-reactive part is typically structured according to the concepts of the application domain (e.g., the services provided by the system). In such systems, the non-reactive part of software stimulates the reactive part with event calls. Whenever the non-reactive part is modified (e.g., during evolution), the existing event calls may become invalid, new event calls may become necessary, and the two parts of software may become incompatible. Manually finding and repairing these types of defects is a time-consuming and error-prone maintenance task. In this article, we present a solution that combines source code model checking and aspect-oriented programming techniques, to reduce the time spent by developers and to automatically find defects, while performing the maintenance task mentioned above. In addition, we present controlled experiments showing that the solution can reduce the time by 75%, and enable the prevention of one defect per 140 lines of source code.

Vibes: A visual language for specifying behavioral requirements of algorithms

Manually verifying the behavior of software systems with respect to a set of requirements is a time-consuming and error-prone task. If the verification is automatically performed by a model checker however, time can be saved, and errors can be prevented. To be able to use a model checker, requirements need to be specified using a formal language. Although temporal logic languages are frequently used for this purpose, they are neither commonly considered to have sufficient usability, nor always naturally suited for specifying behavioral requirements of algorithms. Such requirements can be naturally specified as regular language recognizers such as deterministic finite accepters, which however suffer from poor evolvability: the necessity to re-compute the recognizer whenever the alphabet of the underlying model changes. In this paper, we present the visual language Vibes that both is naturally suited for specifying behavioral requirements of algorithms, and enables the creation of highly evolvable specifications. Based on our observations from controlled experiments with 23 professional software engineers and 21 M.Sc. computer science students, we evaluate the usability of Vibes in terms of its understandability, learnability, and operability. This evaluation suggests that Vibes is an easy-to-use language.

Creating high-quality behavioural designs for software-intensive systems

In todays industrial practice, behavioral designs of software-intensive systems such as embedded systems are often imprecisely documented as plain text in a natural language such as English, supplemented with ad-hoc diagrams. Lack of quality in behavioral design documents causes poor communication between stake holders, up to 100 times more costly rework during testing and integration, and hard-to-maintain documents of behavioral designs. To address these problems, we present a solution that involves the usage of (a) data-flow diagrams to document the input-output relation between the actions performed by a software-intensive system, (b) control-flow diagrams to document the possible sequences of actions performed by the system, and (c) Vibes diagrams to document temporal or logical constraints on the possible sequences of actions performed by the system. The key benefit of this solution is to improve the separation of concerns within behavioral design documents; hence to improve their understandability, maintainability, and evolvability.