Meeting:Abstract 2013 CSDMS meeting-061: Difference between revisions
No edit summary |
m (WikiSysop moved page Temp:Abstract 2013 CSDMS meeting-061 to Meeting:Abstract 2013 CSDMS meeting-061 without leaving a redirect) |
(One intermediate revision by the same user not shown) | |
(No difference)
|
Latest revision as of 10:58, 22 July 2015
Browse abstracts
A vision for EarthCube from the perspective of solid Earth geophysics
[[Image:|300px|right|link=File:]]A major challenge of geophysics today is addressing the problems of general interest through intense collaboration that bridges disciplinary boundaries. Such collaborations are greatly complicated by the fact that Earth Sciences have steadily diverged and evolved to the point of the Tower of Babel. Scientific jargon makes it difficult to meaningfully explore ideas across disciplines, while lack of cyberinfrastructure for sharing causes poor reproducibility and code reuse.
My vision for an EarthCube frontend is that of a maximally simple API that could be run from any platform or in a browser. At it's core, it would support the following functionality:
- make it really simple for someone to submit their own data, models and software with provenance and descriptive metadata;
- support data discovery in 4D space, at a range of scales, through semantically-enabled metadata (and the data might - and will - be stored in one of the existing databases);
- have potential for elaborate visualization capabilities;
- build up upon a social network of some sort (so that there's a face behind each data component); and, finally,
- make it easy to create, modify and run workflows remotely through intelligent combination of software and data components.
The last point seems critical for long term useability of EarthCube, and requires upfront thinking and code coupling capabilities.
Specifically, the plug-and-play component programming approach used by CSDMS could be adapted by the larger solid Earth geophysics community with great long-term benefits, hopefully resulting in better scientific reproducibility, code reuse and, eventually, streamlined collaboration.