Static analysis tools report software defects that may or may not be detected by other verification methods. Two challenges complicating the adoption of these tools are spurious false positive warnings and legitimate warnings that are not acted on. This paper reports automated support to help address these challenges using logistic regression models that predict the foregoing types of warnings from signals in the warnings and implicated code. Because examining many potential signaling factors in large software development settings can be expensive, we use a screening methodology to quickly discard factors with low predictive power and cost-effectively build predictive models. Our empirical evaluation indicates that these models can achieve high accuracy in predicting accurate and actionable static analysis warnings, and suggests that the models are competitive with alternative models built without screening.

Program analysis techniques are used by software engineers to deduce and infer targeted characteristics of software systems for tasks such as testing, debugging, maintenance, and program comprehension. Recently, some program analysis techniques have been designed to leverage characteristics of traditional experimentation in order to analyze software systems. We believe that the use of experimentation for program analysis constitutes a new program analysis paradigm: experimental program analysis. This research seeks to accomplish four goals: to precisely define experimental program analysis, to provide a means for classifying experimental program analysis techniques, to identify existing experimental program analysis techniques in the research literature, and to enhance the use of experimental program analysis by improving existing, and by creating new, experimental program analysis techniques.

Program analysis techniques are used by software engineers to deduce and infer characteristics of software systems. Recent research has suggested that a new form of program analysis technique can be created by incorporating characteristics of experimentation into analyses. This paper reports the results of research exploring this suggestion. Building on principles and methodologies underlying the use of experimentation in other fields, we provide descriptive and operational definitions of experimental program analysis, illustrate them by example, and describe several differences between experimental program analysis and experimentation in other fields. We show how the use of an experimental program analysis paradigm can help researchers identify limitations of analysis techniques, improve existing experimental program analysis techniques, and create new experimental program analysis techniques.

End-user programmers develop more software than any other group of programmers, using software authoring devices such as multimedia simulation builders, e-mail filtering editors, by-demonstration macro builders, and spreadsheet environments. Despite this, there has been only a little research on finding ways to help these programmers with the dependability of the software they create. We have been working to address this problem in several ways, one of which includes supporting end-user debugging activities through interactive fault localization techniques. This paper investigates fault localization techniques in the spreadsheet domain, the most common type of end-user programming environment. We investigate a technique previously described in the research literature and two new techniques. We present the results of an empirical study to examine the impact of two individual factors on the effectiveness of fault localization techniques. Our results reveal several insights into the contributions such techniques can make to the end-user debugging process and highlight key issues of interest to researchers and practitioners who may design and evaluate future fault localization techniques.

End users develop more software than any other group of programmers, using software authoring devices such as e-mail filtering editors, by-demonstration macro builders, and spreadsheet environments. Despite this, there has been little research on finding ways to help these programmers with the dependability of their software. We have been addressing this problem in several ways, one of which includes supporting end-user debugging activities through fault localization techniques. This paper presents the results of an empirical study conducted in an end-user programming environment to examine the impact of two separate factors in fault localization techniques that affect technique effectiveness. Our results shed new insights into fault localization techniques for end-user programmers and the factors that affect them, with significant implications for the evaluation of those techniques.

End-user programmers, because they are human, make mistakes. However, past research has not considered how visual end-user debugging devices could be designed to ameliorate the effects of mistakes. This paper empirically examines oracle mistakes - mistakes users make about which values are right and which are wrong - to reveal differences in how different types of oracle mistakes impact the quality of visual feedback about bugs. We then consider the implications of these empirical results for designers of end-user software engineering environments.

Although researchers have begun to explicitly support end-user programmers' debugging by providing information to help them find bugs, there is little research addressing the proper mechanism to alert the user to this information. The choice of alerting mechanism can be important, because as previous research has shown, different interruption styles have different potential advantages and disadvantages. To explore impacts of interruptions in the end-user debugging domain, this paper describes an empirical comparison of two interruption styles that have been used to alert end-user programmers to debugging information. Our results show that negotiated-style interruptions were superior to immediate-style interruptions in several issues of importance to end-user debugging, and further suggest that a reason for this superiority may be that immediate-style interruptions encourage different debugging strategies.

Emerging research has sought to bring effective debugging devices to end-user programmers. This research has largely focused on how well such devices bring genuine "functional" rewards to end users. However, emerging models of programming behavior indicate that another, often ignored, type of reward-perceivable rewards-can play an equally vital role in how well debugging devices serve end users. Using an empirically evaluated fault localization device, this paper investigates the impact such perceivable rewards can have on end-user debugging. Our results indicate that perceivable rewards alone can significantly improve the effectiveness and understanding of end users performing debugging tasks.