Community-based question answering services accumulate large volumes of knowledge through the voluntary services of people across the globe. Stack Overflow is an example of such a service that targets developers and software engineers. In general, questions in Stack Overflow are answered in a very short time. However, we found that the number of unanswered questions has increased significantly in the past two years. Understanding why questions remain unanswered can help information seekers improve the quality of their questions, increase their chances of getting answers, and better decide when to use Stack Overflow services. In this paper, we mine data on unanswered questions from Stack Overflow. We then conduct a qualitative study to categorize unanswered questions, which reveals characteristics that would be difficult to find otherwise. Finally, we conduct an experiment to determine whether we can predict how long a question will remain unanswered in Stack Overflow.

The software engineering process in video game development is not clearly understood, hindering the development of reliable practices and processes for this field. An investigation of factors leading to success or failure in video game development suggests that many failures can be traced to problems with the transition from preproduction to production. Three examples, drawn from real video games, illustrate specific problems: 1) how to transform documentation from its preproduction form to a form that can be used as a basis for production;, 2) how to identify implied information in preproduction documents; and 3) how to apply domain knowledge without hindering the creative process. We identify 3 levels of implication and show that there is a strong correlation between experience and the ability to identify issues at each level. The accumulated evidence clearly identifies the need to extend traditional requirements engineering techniques to support the creative process in video game development.

Clone detection techniques essentially cluster textually, syntactically and/or semantically similar code fragments in or across software systems. For large datasets, similarity identification is costly both in terms of time and memory, and especially so when detecting near-miss clones where lines could be modified, added and/or deleted in the copied fragments. The capability and effectiveness of a clone detection tool mostly depends on the code similarity measurement technique it uses. A variety of similarity measurement approaches have been used for clone detection, including fingerprint based approaches, which have had varying degrees of success notwithstanding some limitations. In this paper, we investigate the effectiveness of simhash, a state of the art fingerprint based data similarity measurement technique for detecting both exact and near-miss clones in large scale software systems. Our experimental data show that simhash is indeed effective in identifying various types of clones in a software system despite wide variations in experimental circumstances. The approach is also suitable as a core capability for building other tools, such as tools for: incremental clone detection, code searching, and clone management.