Professor Malcolm Atkinson, UK e-Science Envoy, Director of the e-Science Institute (eSI) in Edinburgh
Thursday 5th February 2009, University of Leeds
Slides: http://wikis.nesc.ac.uk/_escienvoy/files/4/45/WRG-Leeds20090205MalcolmAtkinsonV1-reduced.pdf
e-Science exploits the rapid growth in the power of digital systems: e-Infrastructure, instruments, automation and computation to achieve new methods of discovery. Whether those discoveries are fundamentals underpinning a discipline or are interpretations of current circumstances on which immediate decisions are based, their dependability is of vital importance? The methods depend on composing many systems, themselves complex, to achieve results. No one person can comprehend all the aspects of the e-Infrastructure which they use. It changes rapidly and is globally interconnected. For many computational models submodels are added every year to include new phenomena. The collections of data, used for calibration and input, are multi-variate, multi-source and dependent on many methods of measurement and post-processing. Each of these complex systems depend on each other for their integrity. Is this reminiscent of the interlocking complexity which led to the collapse of the financial systems? What must we do to make e-Science dependable?
These are big questions and the lecturer admits he doesn't know the answers. By focusing on how we treat data he will try to motivate other researchers to start answering them.
