Jorge Aranda, from the University of Toronto, recently told us that he has presented a paper that he co-authored at the Requirements Engineering 07 conference. In a study of seven small companies it was found, among other things, that their practices seemed to depend considerably on the context in which they operated; and we observed that practices should not be prescribed generally, but with context in mind.
Jorge tells us more: Besides continuing our work with small companies, we are studying software development in two other contexts. The first one is scientific research groups. Most of today's relevant scientific research requires the development of some software component - in fact in many cases it depends entirely on such software. However, it is usually developed by small teams of people that, though highly knowledgeable in their areas of expertise, do not have proper software development skills.
We're studying several scientific research groups to find how they work around software development and project management challenges, and what problems do they consider particularly important in their programming experience.
The second domain that we are studying is much larger software companies, and in particular, which techniques and organizational structures do they use to address the problem of excessive information overload, so that their teams reach a shared understanding of their project's requirements and status.
I presented the paper at the ITiCSE conference the week before the PPIG 07 workshop. So the details may read better as follows:
Sue is currently a Medici Fellow at the University of Nottingham, UK. This is a one year fellowship concerned with technology transfer, focusing on the commercialisation of research outputs from CS, learning sciences and the business school.... while still trying to complete her PhD thesis - so keeping her somewhat occupied!
Sue presented her paper entitled "Spatial skills and navigation of source code" at the ITiCSE conference in Dundee at the end of June 2007. The results of her study showed relations between spatial ability and patterns of navigation of source code when carrying out a code comprehension exercise. A week later, Sue presented a paper at the last PPIG workshop, discussing the interaction between various individual differences, including spatial ability, on programming performance:
Jones, S., and Burnett, G.E. (2007). Spatial skills and learning to program.
Micheal, who works in the University of Limerick has successfully defended his PhD thesis. The abstract is presented below. Congratulations!
Evolving a Model of the Information-Seeking Behaviour of Industrial Programmers
Several authors have proposed information-seeking as an appropriate perspective for studying software maintenance activities. However, there is little research in the literature describing holistic information-seeking models in this context. Instead, researchers have concentrated on the related fields of software comprehension and software tool development. The work in software comprehension, while providing a cognitive basis for describing software maintenance activities, is abstract in nature and cannot provide strong guidance to those that aim to support software maintenance engineers. In addition, work in this area has been marked by a distinct lack of empirical studies of programmers' actual information needs, during their real-world maintenance tasks. Correspondingly, the work on software tool development, which largely depends on this work, also suffers.
This thesis focuses on maintenance programmers' information-seeking behaviour to address this gap and makes three core contributions to the field. Firstly, it proposes a holistic model of programmers' information-seeking behaviour, derived from related information-seeking research from other domains. Secondly, it derives and presents an analysis schema (Coding Manual) that allows programmers' talk-aloud to be characterised and subsequently analysed in the context of this model. Thirdly, it presents eight empirical studies that serve to evaluate and refine the proposed preliminary information-seeking model for programmers involved in software maintenance activities (using this schema). This evaluation largely validated the model but also suggested several important refinements. Indeed, the results are highly consistent with the 'Concept Location' research of Rajlich et al. and with Marchioninni's information-seeking work.
The case studies, their results, and their impact on the proposed information-seeking model are discussed in this thesis along with recommendations for further research based on its findings.
John Rooksby, one of the organisers of the forthcoming PPIG 2008 workshop, also organised a small workshop at Lancaster university entitled Testing Socio-Technical Systems between 20th and 21st September 2007, following on from the earlier successful 'ethnographies of code'.
This workshop focused upon socio-technical issues in systems testing. Despite a growing interest in human and organizational issues in the development and use of technology, research on testing has remained stubbornly technical. Whilst this technical work has made great advances and has debunked the idea that testing has to be a person spotting errors, it has not and cannot remove the fact that testing takes place within organizational constraints and must meet organizational demands.