Sites Inria

Version française

In focus

Article awarded prize thirteen years after it was first published

Nicolas Anquetil, a member of RMOD, the joint research team with Lille 1 University and the French National Centre for Scientific Research, has, along with Janice Singer (NSERC), Norman Vinson (NSERC), and Timothy Lethbridge (University of Ottawa), just been awarded the CASCON prize for First Decade High Impact Papers for an article they wrote in 1997.

This year saw the 20th edition of the IBM Center for Advanced Studies'  international conference on software engineering, CASCON. To mark the occasion, the CASCON judging panel selected fourteen of the 425 articles published between 1991 and 2000, based on their direct impact on industry (accessibility and applicability of results) and the quality of their approach (creativity, originality, exemplary methodology and interpretation).

The article "An examination of software engineering work practices", published in 1997 by Nicolas and his partners, proposed that new requirements should be set in software design in order to help engineers optimise the performance of certain tasks. The tools in existence at the time did not meet these conditions.

To arrive at this conclusion, the four authors followed several groups of engineers for six months in order to study their practices. The document presents the results of the study in terms of the activities performed (e.g. research, publishing, compilation, etc.), the time devoted to each activity, and the types of tools used. The authors suggested that the working practices of engineers upstream of software design should be considered. At the end of the study, they produced a piece of software that met user requirements.

Nicolas Anquetil looks back at the context of the article and how things have changed in the years since:

"Although it is thirteen years old, this article remains relevant today in that it looked at the non-technical aspect of software engineering.  Software Engineering is just about the worst name you could use to describe all the activities that go into designing, developing, installing, operating and maintaining a computer system. First of all, in essence, it is not about software, but rather about properly organising the work of a group of professionals to build a system (which happens to be a piece of software). And secondly, it's not an engineering discipline like electronics, physics or chemistry, because computer software is not based on any natural laws. Software is invisible, intangible and limited only by the constraints of our imagination and our capacity for abstract thought.

Researchers in this field, myself included, are typically technicians with a strong interest in computers. So we are the ones who tend to put these solutions forward. But two of the co-authors of the article are psychologists and so naturally opted to invert the research process, putting human beings at the centre of things and identifying their needs so as to be able to develop suitable solutions. These days, few people would deny that computer science needs the help of social sciences to find solutions suited to its users, but that idea was not so widely accepted at the time. "

Keywords: Equipe RMOD Génie Logiciel Maintenance logicielle