Summer 2013 panel discussion

From Earth Science Information Partners (ESIP)

A Panel Discussion on Survey Scope

It has been suggested that at the summer ESIP meeting we convene a panel discussion on the topic of scope for a data decadal survey. It would include members from agencies and organizations such as NASA, NOAA, USGS, NSF, ESIP, CODATA, RDA, etc. The goal would be to hear their perspectives on scope and use that to help identify possible statements of task.

Please add thoughts, comments here.



03.21.13: We are moving forward with this idea. Carol and Erin may place this as a plenary session in the meeting.

All that follows is a straw man for the sake of discussion. Please chime in.

We are thinking of two parts: 1) visionary, 2) panel discussion

Panel discussion abstract

The Potential Value and Impacts of a Data Decadal Survey

A Decadal Survey is a broad study of a topic or area that is coordinated by the National Research Council (NRC). It provides community consensus on research goals and priorities for moving forward in a particular focus area over the next ten years. The studies are requested and funded by government agencies and other organizations and are used to prioritize research areas and focus efforts and resources accordingly. The decadal survey process has been shown to be a robust method for developing goals and to support objectives in pursuit of high priority science questions. The NRC has done a number of these studies, including its first decadal survey for Earth science in January 2007 at the request of NASA, NOAA, and the USGS.

Scientific data collected or modeled by government agencies are a public investment and should be stewarded appropriately to maximize the return. Today's science is increasingly collaborative. Important research questions increasingly span projects, discipline domains, and other boundaries. There is a growing interest in the repurposing of data far from the point of collection. Data collected long ago can become important today. Current data management and stewardship practices are not sufficiently meeting these pressures. Scientists today regularly cite the 80/20 rule for working with data: 80% of their effort is spent finding, understanding, acquiring, and putting the research data in a usable format, and 20% doing actual science.

Members of ESIP and representatives from the NRC Board on Research Data and Information (BRDI) have organized a cluster to investigate the need for and feasibility of conducting a Data Decadal Survey (http://wiki.esipfed.org/index.php/Data_Decadal_Survey). The survey would address overarching issues and research priorities in scientific data management and stewardship. Improved practices in this area could ultimately enhance scientific knowledge by increasing the meaningful availability of data and redirecting resources previously required for data discovery, acquisition, and formatting to performing actual science. When data sets can be easily analyzed and combined in novel ways then new scientific insights are more likely to occur and more quickly. Such a survey could address at the broadest level gaps in data management knowledge and practices that hold back scientific progress.

At the ESIP summer meeting we are organizing a panel discussion around this idea. We are inviting panelists to give us their vision of future data developments, and to discuss data management and stewardship (DMAS) topics and issues.

Questions for the Panel

  • Do you have ideas for science that could be done with improved DMAS that cannot be done now? If so, what?
  • Do you know of serious gaps in DMAS that negatively impact science? If so, tell us about some.
  • Do you see a need for a Data Decadal Survey? If so, why? If not, why not?
  • What is your vision for the future regarding scientific data? Please be bold and include fanciful, idealistic, lofty, and even utopian ideas.

If in support of a survey:

  • What do you think are the big questions around DMAS that must be addressed? What are the highest priorities?
  • Is a Decadal Survey the right vehicle, or should it be something else?
  • What do you think should be the scope of the survey?
    • Given that either extreme of depth or breadth is of less general use, what exactly should we target?
    • Earth Science only or broader? For example, does it make sense to start "small" in the Earth Sciences and then generalize? Or, as data management problems across all domains are basically similar, should we start with the more general and perhaps go into greater detail later?
    • Data? Software? Methodologies?
  • What would be metrics for assessing survey success?
  • A Data Decadal Survey is risky because the topic is extremely broad, the community of data users is vast and heterogenous (and can include commercial interests), and the outcome will not be a focused mission, facility, or research initiative. How should that risk be managed?

Potential panelists

We seek a visionary perspective, in additional to agency and organizational perspectives.

We think 5 panelists would be a maximum.

Potential panel members (note: order is irrelevant, and only some affiliations/associations are listed, "LOCAL" means local to Chapel Hill)

  • John Wilbanks
  • Stan Aholt, RENCI, LOCAL (and already participating on another panel)
  • Steven Friend (from genetics)
  • Berrien Moore, College of Atmospheric and Geographic Sciences, University of Oklahoma
  • Chris Greer, Associate Director for Progrma Implemention in Information Technology Laboratory, NIST
  • Piers Sellers, astronaut, meteorology, computer modeling
  • The (2 hour) video "Earth from Space", Earth from Space has comments from visionary-types:
    • (From Netflix website) cast: Waleed Abdalati, David Adamec, Charlie Bristow, Carl Feldman, Holly Gilbert, Jeffrey B. Halverson, Jay O. Sanders, Piers Sellers, Emily Shuckburgh
  • Martha Maiden, Program executive for Earth Science Data Systems, NASA
  • Jeff Newmark, Executive Secretary, NASA Heliophysics Subcommittee
  • Dan Baker, Director Laboratory for Amospheric and Space Physics (LASP), Chair of Heliophysics Decadal Survey
  • "Lessons Learned in Decadal Planning in Space Science, A Workshop", Report from Nov 12, 2012 workshop, includes names of workshop presenters and moderators, including several mentioned here
  • Thomas Karl, Director, National Climatic Data Center
  • Mark Parsons, RDA
  • Fran Berman, RDA, BRDI
  • Peter Fox, RPI, Ambassador at Large
  • Cliff Jacobs, EarthCube (retired)
  • Bob Chen, Director CIESIN, mmanager SEDAC, prior secretary-general CODATA
  • Chris Mentzel, Program Officer, Gordon and Betty Moore Foundation
  • Sara Graves, BRDI, CODATA Executive Committee member, Director of IT and Systems Center, UAH
  • Peter Colohan, Senior Policy Analyst, White House OSTP
  • OSTP/NITRD Big Data Senior Steering Group members (chairs are NSF program managers, ie. control $$)
    • Mark Suskin, NSF, Co-Chair
    • Robert Chadduck, NSF, Co-Chair
    • Laura Biven, DOE Science
    • Sky Bristol, USGS (Chief, Science Information Systems, "developing an information science research and development program within the USGS Geospatial Information Office")
    • Allen Dearry, NIH
    • Emily Grumbling, NSF
    • Alan Hall, NOAA
    • Tsengdar Lee, NASA headquarters, High End Computing Manager, manages supercomputing facilities at NASA Ames and Goddard
    • Peter Lyster, NIH (climate modeling, computing, now at NIH in computational biology, crosses domains!)
    • Sastry Pantula, NSF
    • Jonathan Petters, DOE Science
    • Don Preuss, NIH
    • Tom Statler, NSF
    • Wendy Wigen, NCO/NITRD
  • Otis Brown, Vice Char of Panel on Land-Use Change, Ecosystem Dynamics, and Biodiversity, ~2008, now at NC State, Raleigh, and NOAA NCDC (Asheville) affiliate LOCAL
  • Ana Barros, Duke University LOCAL, member of National Academies Research Committee (CRC), has served on ~10 NRC Boards and Committees

Draft letter of invitation to serve on panel

Dear X,

I write to you as a representative of the Earth Science Information Partnership (ESIP) Data Decadal Survey cluster. If you are not familiar with ESIP, the page http://www.esipfed.org/ provides some information about us.

With National Research Council (NRC) representatives, our cluster is exploring the feasibility of conducting a Data Decadal Survey. As part of that effort, at the upcoming ESIP summer meeting we are holding a panel discussion on the topic. The abstract is below.

You have been identified as an expert/visionary in the field of Y. Would you be interested and available to serve on our panel? It will occur on the morning of Wednesday, July 10, at the University of North Carolina in Chapel Hill.

Thank you for your consideration of this invitation. If you have any questions please contact me. Please let me know if you are available at your earliest convenience so that we can plan accordingly.

Sincerely,

Anne Wilson ESIP Data Decadal Survey cluster coordinator/representative/??

Outcome, deliverable

Based on the ideas from the panel discussion, develop a short white paper to describe a potential workshop to be held in the fall of 2013, which would include representatives from beyond ESIP, and which would provide options for the nature, scope, tasks, and constituencies for a Data Decadal Survey.

Discussion, ideas, thoughts, brainstorming

Some cluster members are uncomfortable expanding the scope of the survey beyond Earth sciences. Does it make sense to start "small", with Earth sciences, and then generalize? Or, as data management problems across domains are all basically similar, start with the more general and go into greater detail later?