There is an urgent need for new, automated techniques to analyze software and to guarantee its quality, argues part-time professor of Automated Software Testing and CWI group leader Jurgen Vinju in his inaugural speech at Eindhoven University of Technology on Friday 12 February. The quality of software is an insidious problem. When mistakes come to light or software is completely incomprehensible, it is often too late and the damage already suffered.
"Does software always implement the policy or not? Could this software leak personal information? And how much will it cost to adapt this software? These are questions that still remain often unanswered for large software systems. The enormous complexity of software makes that errors lack visibility and adjustments are made without being clear what consequences this has for the quality," Vinju says.
"That software systems are often so large and complicated is an almost inevitable effect of creating software," he adds. According to Vinju the quality of software can simply not be guaranteed nowadays. Moreover, maintenance costs are rising high because of this high complexity: more than half of the software costs goes to maintenance.
Vinju advises the government and other parties to be aware of this reality and let others evaluate the quality of the software instead of the supplier. "And just do not do some things," says Vinju. "No vulnerable smart energy grid, for example, no voting machines and don’t store personal CITO test results on a web server."
Harvesting and sowing
According Vinju the solution must be sought in better automated software analysis. In his inaugural speech on 12 February he elaborates on new technologies and tools to automate the analysis of software and on the challenges involved. Vinju advocates investing heavily in source code and system analysis, in addition to relatively easy existing software-based innovations. Vinju: “Now we could perhaps harvest what we already have, but we also have to sow now for a better future.”
Jurgen Vinju (1977) received his PhD degree in 2005 at the University of Amsterdam. He gained experience as a guest researcher at Alcatel-Lucent Bell Labs and the IBM T.J. Watson Research Center. Since 2012 he has been the leader of the research group Software Analysis and Transformation (SWAT) at Centrum Wiskunde & Informatica (CWI), the national research centre for mathematics and computer science in the Netherlands. As of September 1, 2014 he has been appointed as a parttime professor of Automated Software Analysis at the Faculty of Mathematics and Computer Science of the TU/e.
Picture and text source: TU/e