... personal wiki, blog and notes
Bryan's Blog 2017/11
Science and the Digital Revolution: Data, Standards, and Integration
I was asked to give a talk at this CODATA meeting which was aimed at developing a roadmap for:
Mobilising community support and advice for discipline-based initiatives to develop online data capacities and services;
Priorities for work on interdisciplinary data integration and flagship projects;
Approaches to funding and coordination; and
Issues of international data governance.
I gave a talk on Data Interoperability and Integration: A climate modelling perspective. For this talk I was asked to address an example from the WMO research community on what we have accomplished in standardising a range of things, and reflecting on what has worked/failed and why. I wasn't given much time to prepare, so this is what they got: pdf (11.5 MB).
There were some other interesting talks, some of which resonated particularly with me, including
Bill Michener talking about Dataone, and the importance of governance activities where all lines of responsibility are clear: no committees or panels whose remit is unclear or whose reporting lines are non-existent.
Jeremy Fry talking about digital infrastructure in support of (physical) chemistry, who had some thought provoking comments on creativity, "engrooved" bad habits (harking back to a talk by Paul Trowler, which I have yet to get a copy of) and the potential role of AI in scientific discovery:
Also, human creativity seems to depend increasingly on the stochasticity of previous experiences - particular life events that allow a researcher to notice something others do not. Although chance has always been a factor in scientific discovery, it is currently playing a much larger role than it should. (The Atlantic, April 2017).
Robert Hanisch talking about experience of the IAU in developing technical infrastructure, and in particular the governance of technical components for the International Virtual Observatory Alliance. The IVOA attempts to follows the W3C governance model:
Data access protocols require two independent implementations.
Notes are promoted to Working Drafts, which become Proposed Recommendations, which get endorsed to become Recommendations, which eventually become standards although the did note that the formal endorsement step wasn't (yet?) in practice.
All communities reported difficulties in funding infrastructure at the sorts of levels necessary, and little or no investment in usability ... funders like prototypes ... but not boring maintenance and improvement ... which means that sometimes scientific take-up falls far below what was envisaged.
Unfortunately I was not able to go to days one and three of this three day meeting.
by Bryan Lawrence : 2017/11/15 (permalink)