Geraint Lewis and Chris Power have a thought-provoking post today in The Conversation.
They point out that one of the key driving forces in science has been ‘spectacular growth in power and storage’. They add that ‘when the Square Kilometre Array starts observing the sky in 2020, it will generate more data on its first day than will have existed on the internet at that time‘.
However, they note that ‘many scientists-in-training are ill-equipped to write software (or code, in the everyday language of a researcher) that is fit-for-purpose’. They continue:
For this reason, it is probably unsurprising that many fields are awash with poor, inefficient codes, and data-sets too extensive to be properly explored.
This is exacerbated by a system that values the publication of scientific results. Geraint and Chris conclude:
- Science needs to make a cultural change in understanding on what makes a good modern scientist.
- Fertilise links with computer science colleagues.
- Develop a career structure that rewards those who make the tools that allow Big Science to happen.
I really enjoyed Geraint and Chris’s paper. Their ideas resonate with me at a time when I am thinking about meta-activities in learning organisations. They resonate too with my interest in dynamic forms of accreditation generally and in performance analysis in particular.
I am very attracted to Jason Lear and Darrell Cobner‘s ideas and framework for accreditation of performance analysts linked to continuing professional development. As a result of Geraint and Chris’s thoughts I will be keen to include learning to code in any accreditation framework.
A recent post by Tony Hirst gave me a sense of how we might encourage coding. I thought his presentation to journalism students at the University of Lincoln was exactly the kind of open resource that might be included in emerging Vocational Open Online Courses (VOOCs). In these contexts, paper publication is peripheral to personal learning and development. They have a clear industry focus.
Donald Clark writes:
We’ve gone for a solution that taps directly into subject matter expertise – experienced practitioners, experienced course designers and a delivery mechanism that goes straight to potential learners. That’s really what the ‘Napsterisation’ of learning is all about, the democritisation, decentralisation and disintermediation of learning.
I think an industry wide open online course in coding for performance analysis would harvest to rich and diverse experiences of the community that go beyond a single institution … so that many of us are able to deal with the oncoming flood of data in our own practice. This practice itself we be iterated through open sharing.
In doing so we might be mindful of Alan Alda’s advice about communicating science: ‘Communication is not something you add on to science, it is of the essence of science’.