PML 2007

The main goal in a software projects is to deliver the desired functionality with desired quality and within budget and schedule. To achieve this goal, engineers apply existing knowledge and skills so the problem can be solved with least effort and in shortest time. Project Managers want to achieve the goal with minimum risk.
As any experimentation can increase risk as well as effort, both project managers and engineers are averse to experimentation during a project.

However, innovation is the buzzword now - organizations need to continuously innovate.
An innovation inevitably has foundations in some new knowledge or insight. Hence, to facilitate innovation, one needs to facilitate creation of new knowledge and development of subtler and better understanding of various factors at play during a project. Generally, knowledge creation is left to researchers, who use many different paradigms to develop new knowledge. Experimentation is one of the paradigms of knowledge creation that is used in many disciplines, including software engineering. While researchers often conduct controlled experiments which can give precise knowledge, project people can also experiment and create knowledge by treating their projects as potential experiments and innovating suitably to execute the project in a manner that experimental knowledge can also be obtained. As software engineering knowledge is often based on experience, such an approach can move some part of experimentation from labs to projects, which can widen their scope. It will also have a strong benefit of adding research orientation on project people, which will help facilitate innovation in the long run. The talk will also discuss some experiments that can be conducted in projects for this purpose.