As well as attendees from the project we had more than twenty others from companies in Sweden, Denmark, the UK and the Netherlands and universities in Denmark, Sweden, the UK and Brazil. The day went very well, with project members like myself being pleased to see how the separate strands of the work are coming together, and that the interactions we're having informally are turning into practical tools.
More importantly, we got positive reports back from the external attendees too, who were able to give us really helpful suggestions about how we could extend and build on what we had done. The prize for the coolest talk has to go to Benjamin Vedder, whose demo of fault injection for a quadcopter using a QuickCheck model stole the show [OK, the fault injection only happened in a simulation – to the relief of the front row – but still fun to see the use of QuickCheck in Erlang used to inject faults in to C++ embedded systems!]
Benjamin's quadcopter |
The morning programme gave an overview of the project and property-based testing in QuickCheck, as well as introducing the VoDKATV platform from Interoud, which provided a case study running through the more detailed presentations in the afternoon
- Overview of the project – John Derrick, University of Sheffield
- Case study: a web service for administrating the VoDKATV platform – Miguel Á. Francisco, Interoud
- Property based testing for web services: an introduction – John Hughes, Quviq and Chalmers University.
- Inference of state machines from QuickCheck traces – Kirill Bogdanov, University of Sheffield
- Automating Property-based Testing of Evolving Web Services – Huiqing Li, University of Kent and Laura Castro, University of A Coruña.
- Fault injection – Benjamin Vedder, SP.
- More-bugs -- how to not find the same bug over and over again – Ulf Norell, Quviq.
- A Property-based Load Testing Framework – Diana Corbacho, Erlang Solutions Ltd and Clara Benac Earle, UPM, Madrid.
- Smother: Extended code coverage metrics for Erlang – Ramsay Taylor. University of Sheffield
- Automatic complexity analysis – Nick Smallbone, Chalmers University
I'd recommend this kind of open day event to anyone planning a complex research project, as it gives you all a chance to get some invaluable external insights into the work that you're doing. It's also a real fillip to the project to see the tangible progress that has been made, and also to enthuse project members to do even better in the second half of the project.