Thursday, December 18, 2008

Informatics at AGU

Earth and Space Science Informatics is still the fastest growing group in the American Geophysical Union, with 3082 affiliations this year, up from about 2000 a year ago. We had 20 technical sessions during this week’s Fall meeting in San Francisco, with oral talks and posters everyday. Tickets for the ESSI reception on Tuesday were sold out. [right, AZGS geoinformatics chief Steve Richard (center) discusses the public release of GeoSciML v2 at AGU, with Flip Dibner, (left) OGC Interoperability Inst, and Boyan Broderic, Geol. Survey of Canada. Boyan and Steve unveiled the release during the Wed. session on interoperability.]

The interplay of computer scientists, domain scientists, and this year, a big influx of librarians, made for especially robust exchanges.

A few observations –
  • Provenance of data (and software) is maturing quickly as a field
  • Interoperability is accepted as a goal by everyone
  • Search and discovery tools and mechanisms are proliferating
  • Some components have become or are becoming institutionalized, such as Digital Object Identifiers or DOIs
  • Virtual Observatories and observation networks are a huge area with almost unimaginable flows of real-time data on the cusp of becoming a reality
  • Cloud computing offers tremendous flexibility for data flows that vary over time
  • Social networking is emerging as core aspect of scientific communication and collaboration
  • There seems to be a mark up language for every topic
  • Some folks continue to re-invent the wheel
  • Half of the scientists at NASA appear to be developing sophisticated data systems for the other half – with each of them independent and incompatible with the others

The NASA adoption and certification procedures for informatics mirror the military aerospace process – massively complicated and bureaucratic. There is no way the academic or private sector will adopt this. And a colleague lamented that this is symptomatic of NASA lack of nimbleness and innovation.

1 comment:

  1. Great to read something on ESSI in your blog! I could not attend, but so I get at least some impression on what's going on there;)

    Btw: What exactly do they mean with provenance? Has this to do with trustworthyness of data?