Information Foraging Theory in Software Navigation

Margaret Burnett - Oregon State University

Oct. 30, 2015, 2:30 p.m. - Oct. 30, 2015, 4 p.m.

Trottier 0070


In this talk, we reconsider the notion of how people go about navigation-intensive tasks like debugging, by exploring a theory that is consistent today's large collections of source code and modern programming environments. The theory we are exploring is called information foraging theory. Information foraging theory has become very influential in the field of human-computer interaction, and we are working to bring its benefits to the software engineering community. For example, in recent years, the software engineering community has begun to study program navigation and tools to support it. Some of these navigation tools are very useful, but they lack a human-oriented theoretical basis that could reduce the need for ad hoc tool building approaches by shedding light on what is fundamentally necessary to the people using such tools.

Thus, in this talk, we present our work on the PFIS family of models (Programmer Flow by Information Scent), which are behavior models and algorithms, based by information foraging theory, of programmer navigation during software maintenance. We also describe strengths, weaknesses, and open questions we have found in our empirical studies applying these models to expert programmers debugging real bugs described in real bug reports for real Java applications. Our results have been very encouraging, and also point out interesting differences between how information foraging theory applies to the web browsing versus information foraging theory's application to software development activities like debugging.

Biography

Margaret Burnett's research is in human issues of software development, which lies in the intersection of HCI and software engineering. Her current research focuses on end-user programming, end-user software engineering, information foraging theory as applied to programming, and gender issues in those contexts.


End-user software engineering is the first research area to rigorously consider the problem of dependability in end-user programming, and Burnett led in founding this research area. Her team's "WYSIWYT" systematic testing approach for end-user programmers initiated the groundwork, and in 2003, she co-founded and became Project Director of the EUSES Consortium. Under her leadership, this collaboration grew to 10 institutions whose contributions have helped ordinary end users achieve up to 10 times greater effectiveness at guarding against software defects, receiving wide recognition for technical quality (11 Best Paper recognitions).

Burnett's research also considers supposedly "gender-neutral" software development platforms. Prior to this work, gender investigations into software had addressed only gender-targeted software, such as video games for girls. Burnett and her team systematically debunked misconceptions of neutrality in software platforms, and then presented software features that help avert the identified problems. The team has reported results in over 30 publications, has given numerous invited talks, has attracted NSF and Microsoft funding, and has been featured in such popular press outlets as the Boston Globe, Chicago Tribune, Washington Post, CNN, CBS, Time, and USA Today.

Burnett's awards for her work include several Best Paper recognitions, IBM's International Faculty Award, and the NSF Young Investigator Award. She was recently honored with her university's Excellence in Graduate Mentoring Award, College of Engineering's Research Award, College of Engineering's Research Collaboration Award, and the Elizabeth P. Ritchie Distinguished Professor Award.

She is currently Workshops Co-Chair for ACM/IEEE ICSE'13 (International Conference on Software Engineering), Papers Co-Chair for IEEE VLHCC'13 (IEEE Symposium on Visual Languages and Human-Centric Computing), was Papers Co-Chair for ACM CHI'08, and serves on a variety of HCI and Software Engineering conference program committees. She also co-led the NCWIT "REU in a box" project (2011), was a panelist at ACM CHI'11 and ACM/IEEE ICSE'11 events on U.S. National Science Foundation "Broader Impacts", and co-chairs the Executive Committee of the Academic Alliance of the National Center for Women In Technology (NCWIT).