Scientific Discovery in the Information Era: The Role of Data, Databases and Categorization

• Report rights infringement
• published: 29 May 2016
• views: 9

10:00 AM, Monday, May 16th, 2016 John Rumble, R&R; Data Services, Gaithersburg MD USA, John.rumble@randrdata.com The Information Era has revolutionized scientific discovery. Not only has the relationship between scientist and information, especially data, fundamentally changed, but also scientific data themselves have become very different. A simple example is that as recently as fifty years ago, an individual scientist could read, digest, and exploit all the data related to a phenomena. Today no one person can use, much less read, a terabyte or more of data. In my presentation, I begin by giving a short history of scientific databases and their role, first in collecting scientific facts and then in predicting new facts and developing scientific knowledge. I then discuss scientific data today and the challenges the digital age brings in exploiting those data, especially in dealing with complex systems. I proceed to talk about the tyranny of independent variables and the multi-disciplinary Tower of Babel, illustrating how these issues affect the discovery of causality in complex systems. I then give my thoughts on the role of categorisation in creating scientific understanding – past, present and future. Here I focus on establishing goals for categorisation and metrics for assessing progress towards those goals. I discuss how overlapping and competing categorisation goals are affected by lack of knowledge about independent variables and sparse property matrices. Finally I try to relate categorisation to knowledge on synthetic science. Throughout I draw on examples from many disciplines, with special emphasis on nanomaterials. The second part of my presentation is designed to be a group discussion on “Categorisation as Risk or Knowledge: Is it either/or or both?” Website: https://unnaturalkinds.wordpress.com/