Today's interview is with Prof. Mark Plumbley, of the Department of Electronic Engineering & Computer Science at Queen Mary University of London.
Tell us a bit about your organization and its goals.
We are a new project "Sustainable Software for Digital Music and Audio Research", funded by the UK Engineering and Physical Sciences Research Council (EPSRC), and based at the Centre for Digital Music (C4DM) at Queen Mary University of London. The aim of the project is to provide a Service to support the development and use of software, data and metadata to enable high quality research in the Audio and Digital Music research community. It's really about getting research results—including robust software to implement that research—out to the people who should be able to use them, and then keep it working.
Tell us a bit about the software your group uses.
We use some generic signal processing development tools like Matlab, as well as C/C++, Python, Prolog, etc. We also use some music-specific software, like Max/MSP, SuperCollider, and Ableton Live. Some people use Subversion for version control (even occasionally for writing joint research papers in LaTeX).
Tell us a bit about what software your group develops.
Some of the software we've already developed is for our own research, but we are increasinly making more and more available for others to use: see http://www.isophonics.net/ for a selection. Some of this includes:
- Sonic Visualiser: an application for viewing and analysing the contents of music audio files.
- SoundBite: an iTunes plugin to create great-sounding playlists.
- BeatFx: a suite of real-time musical audio effects which are automatically synchronized to the beat.
Who are you hoping Software Carpentry will help?
The new project includes software developers whose job it is to take flaky research software and turn it into robust software usable by other researchers. We hope that the Software Carpentry course will train up a next generation of PhD students and researchers that can create robust research software for themselves, so they don't have to rely on some other developers to come along later and clear up the mess.
How do you hope the course will help them?
We hope that the course will help researchers to think about robust software development from the outset of their research, not as an afterthought. Therefore when the paper is published, or the thesis is finished, the software that implements that research will be available for others to use in a robust and sustainable form. This should also benefit the researcher themselves, since people using their software will cite their research papers when acknowledging their software.
How will you tell what impact the course has had?
We would like to follow up the students who attended the course, and see how well they are producing well-written and well-documented research code, in comparison to what they would have done without it. Some of the impact will be more nebulous, about changing attitudes to the role of software in research. But if we get this right we should be pushing at an open door. "Research Impact" is the name of the game at the moment, and what better way to help your audio research have impact than make software for it available in as sustainable software which is usable by the people who need to use it!