0
Facebook
Volume 90 Issue 47 | pp. 4-5 | Letters
Issue Date: November 19, 2012

Deliberating Data

Department: Letters

The article “Globalizing Data Infrastructure” gives the impression that concern over interoperability and sharing of data are recent phenomena (C&EN, Sept. 24, page 26). This is hardly the case. Numerous organizations, both national and international, have addressed this issue over the past four decades. One important example is CODATA, the Committee on Data for Science & Technology of the International Council for Science (ICSU), founded in 1966, whose constitution includes the following objective: “Enhance coordination of data projects in order to provide consistent formats, maintain standards, or facilitate conduct of the work and encourage standard interchange formats for data.”

CODATA has carried out numerous projects in the physical, bio-, and geosciences aimed at adopting common data formats and standards by a consensus process. It has worked with the ICSU unions and other international scientific bodies to achieve this goal. Many of the ICSU unions and other discipline-based organizations have adopted exchange formats to handle data in their own field.

An example from chemistry is ThermoML, an XML-based standard adopted by IUPAC in 2006 for capture and storage of experimental, predicted, and critically evaluated thermodynamic properties. IUPAC has also developed InChI, the International Chemical Identifier, a unique identifier of chemical substances that can be used in printed and electronic data sources, thus enabling easier linking of diverse data compilations. InChI has already been adopted by dozens of database providers, publishers, and government agencies.

My own 40-year experience with data management shows that bottom-up endeavors of this type, led by scientists who understand the data and how they will be used, are more successful than a top-down approach. Additional funding for these activities is certainly needed, but I hope the leaders of the Research Data Alliance will carefully examine and build upon the existing data infrastructure rather than setting out to reinvent the wheel.

David R. Lide
North Potomac, Md.

 
Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society