EarthCube Update: Still More Charrette Impressions

EarthCube

EarthCube is a joint effort of the NSF Geosciences Directorate and Office of Cyberinfrastructure. Its goal is to use technology and knowledge management techniques to increase the productivity and capability of researchers and educators working at the frontiers of Earth system science. (If you're new to EarthCube, you may want to read these previous posts.)

Three members of the Unidata Program Center staff were among the attended the EarthCube charrette held in Washington D.C. on November 1-4, 2011. The following are Unidata Development and Outreach Manager Ben Domenico's notes from the meeting. Impressions from Unidata Program Center director Mohan Ramamurthy and developer Russ Rew are also available.

By Ben Domenico

The charrette was interesting, but not what I expected. The emphasis was on developing a "bold vision" of what the geosciences enterprise could be like in 10 years, then determining needed capabilities for achieving that vision along with barriers that have to be overcome. Very little attention was given to more than 100 whitepapers that had been submitted before the charrette. I came away frustrated, but in retrospect I wonder whether this isn't a natural at this phase. NSF is trying a radical new approach with EarthCube — having the scientific community work collaboratively to achieve a common set of objectives rather than on individual competitive grants and projects. Frustrations are inherently part of the process when the process is fundamentally new.

Since Mohan and Russ have already covered the salient points of the charrette in their earlier articles, I'll focus on one of the most interesting (to me) collaborative exercises I was involved in because it illustrates how, with some creative thought, groups with very diverse scientific goals and practices can overcome initial difficulties and chart a course where they might work together to develop interoperable data systems of use to others outside their individual disciplines.

On Thursday afternoon, there was a breakout session aimed at coming up with the "next step" we could take to begin the march toward the ultimate 10-year transformative EarthCube vision. The breakout group consisted of people from the seismic, oceanographic, and atmospheric science communities.

At first the group nearly broke into two groups because we could not see how the solid Earth (seismic/earthquake) community and the fluid Earth (atmosphere/oceans/climate) community had anything in common at the science layer. But one of the members came up with the unifying idea that the groups might have something in common in terms of developing systems for dealing with hazards — for example the earthquake/tsunami or the volcano/plume. With that goal in mind, the group came to agreement on a short list of practical objectives that we felt could be accomplished in the 3-6 month time frame. It involved a series of steps we could take to bring all the data systems represented by the group up to a common level for making their diverse datasets and the associated metadata available (in a common data discovery system).

One group (Southern California Earthquake Center) was searching for a means to make their data and metadata available at all. The thought was that they could use some of the technology from the other groups (CUAHSI HIS, Unidata THREDDS Data Server, OOI cloud-based system, Ocean Leadership RDBMS) to make their data and metadata available. That would bring all the diverse groups up to what we called

Level 1: Data and metadata exposed via web services

In parallel the other groups could begin working to have their metadata harvested for discovery within the CUAHSI HIS Central geosciences discovery system. This would most likely involve the use of a brokering service layer to mediate among all the different server interfaces and the discovery system client side software. Two of the groups already had experience using the GI-cat brokering software in that role. That would bring all the groups up to

Level 2: Metadata incorporated into one geosciences-wide discovery service

Those goals seemed to the group to be reasonable and practical for a 3-6 month time frame and would be a good illustration of how common infrastructure could be employed by a widely disparate set of geosciences scientific communities with a common goal of hazard response in mind.

But in case there was time left over, it was thought that the next phase objective could be in interoperability experiment with the datasets themselves. In other words determine whether the groups could successfully access one another's datasets via standards-based web services.

Level 3: Access and use of interdisciplinary datasets via web services

Unfortunately, after the group presented these proposed "next steps," with the ultimate goal of interoperability among the very different data and metadata systems represented by the group, the facilitator said that "this was not EarthCube" and that what we had in mind was far too limited, that we needed more steps, and so forth. It was getting toward the end of the day and the group was a bit dispirited by that response so we never followed up on it. As a result our proposal never got into the mix of two-page collaborative proposals. But, for at least one of us, it was a really practical set of objectives that we could have worked on as a collaborating team. I thought it was worth noting down what I could remember in case there is another context in which to work on this or another phase of EarthCube where it could be revisited.

Reflecting on my experience at the charrette, it occurs to me that the unstructured approach NSF is taking provides the community with an important opportunity to shape the future direction and form of EarthCube. That's how we should be channeling any frustrations we may be feeling with the process.

Comments:

Post a Comment:
Comments are closed for this entry.
News@Unidata
News and information from the Unidata Program Center
News@Unidata
News and information from the Unidata Program Center

Welcome

FAQs

Developers’ blog

Recent Entries:
Take a poll!

What if we had an ongoing user poll in here?

Browse By Topic
Browse by Topic
« September 2019
SunMonTueWedThuFriSat
2
5
6
7
8
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
     
       
Today