Software engineering research results in industrial practice: a tale of two projects (invited talk)

In this talk, I will discuss the use of software engineering research results in industrial practice, based on two projects I have been involved with. The first project addressed the challenge that manipulation of financial market data had to be expressed precisely for a large number of different financial markets. The challenge was addressed by defining a functional Domain Specific Language (DSL) that was geared towards expressing these manipulations at a high level of abstraction. An environment that implements the DSL was built using the Eclipse platform together with a compiler that generates a Java-based reference implementation of these manipulations. The implementation is used as a test oracle to generate test cases, which are in turn used to validate a soft real-time system that implements these manipulations. In another project that is still ongoing, I have proposed the use of software product line research to engineer a family of mobile banking applications. I will reflect on the experience of integrating software product line principles and modern Agile development practices. I will then discuss a few areas of software engineering research, that I have personally been involved in, that I have found not to be very useful in practice. I will conclude by outlining some topics where novel research results would be very beneficial from an industrial point of view.