Last week was the IPY GeoNorth Conference (http://ess.nrcan.gc.ca/ipygeonorth/). I used this opportunity to host an informal 1.5 hour discussion (a town hall meeting) on the evolution of the IPYDIS. Here are the results of that discussion.
Attendees
Mark Parsons, National Snow and Ice Data Center (Co-Chair IPY Data Policy and Management Subcommittee)
Ellsworth LeDrew, Univ. Waterloo (Co-Chair, Canadian IPY Data Committee)
Peter Pulsifer, Carlton Univ.
Jagtar Bhath, Canadian Forest Service
Bea Csatho, Univ. of Buffalo
Tony Schenk, Ohio State Univ.
Peter Ng, Government of the North West Territories
Karen Edwards, Canadian IPY Secretariat
Jean Brodeur, Natural Resources Canada (involved with North American Profile)
Yves Crevier, Canadian Space Agency
Fraser Taylor, Carlton Univ.
Doug Nebert, US Geological Survey (involved with GEOSS interoperability efforts)
Background
I presented a general overview on what the IPYDIS is and what it is trying to accomplish. I also provided background in the following topics, all of which elicited discussion. Particular topics discussed included
- the IPY Union Catalog
- the IPY metadata profile (http://ipydis.org/metadata.html)
- Communication mechanisms
- a possible IPYDIS directory
- the role of the Data Committee and data coordinators
- upcoming events
Union Catalog
Everybody generally agreed with the general approach to sharing metadata through a union catalog. There was some discussion on the protocol for sharing (e.g. OAI-PMH), but it was agreed that simpler is better. Note: I currently have a project that shares metadata between four data centers by simply transmitting FGDC formatted XML. This will link with the IPY portal at the GCMD.
If you are interested in harvesting or serving metadata, please let me know.
Metadata Profile
Again this was generally accepted. The need to validate and provide immediate feedback after entering a record was emphasized
Most discussion centered around controlled vocabulary, especially for parameters and location. We are currently using GCMD vocabulary for these fields, but I suspect that will evolve. On suggestion was to take a “folksonomy” approach and mine free test keywords.
There was also some discussion about registering the vocabulary, linking to the North American Profile, and working with “organically grown commons-based” approaches such as WordNet (http://wordnet.princeton.edu/). Unfortunately, my notes are rather sketchy here. Can anyone provide more details?
The Marine Metadata Interoperability (MMI) project (http://mmi.org) might also be useful here.
Communication
I presented options of email, this forum, a wiki, and social networks. None were received with great acclaim. This forum was agreed as a simple method that maintained a record.
IPYDIS Directory
I presented the possibility of a directory of IPYDIS participants and this was well received if designed as a “directory of experts”. Stay tuned for developments in this area.
Data Committee
To date the Data Committee has been emphasizing the data policy of open access and working to identify what data are to be collected. Town hall participants felt that there is now a need to focus on the preservation of the data collected as part of IPY. (This was brought up in the general session at GeoNorth too). The Data Committee should develop a “digital archiving strategy” and work with the JC and our sponsors to encourage broad adoption and support. The general idea is that data archiving should be mandatory for IPY participants. We need to identify what projects are currently doing for data archive and what the should be doing. The concept of a consistent publishing framework was emphasized. This will be a topic at the next Data Committee meeting.
A related topic was to develop a small brochure or flyer that demonstrates what projects need to do to ensure their data are preserved and accessible. One participant described this as a brief cookbook or recipe projects could follow. It should not only describe what projects should do, but how they benefit.
Priorities
It was agreed that the top priorities for the near term are to identify what data are being collected and to enable basic discovery of these data using the IPY profile BUT to gain buy in we need to emphasize effective data access mechanisms and capture best practices.
I think the new IPYDIS web site (coming soon) will be an excellent means to highlight some of the cool data tools and access mechanisms. For example, I will highlight NSIDC’s Atlas of the Cryosphere (http://nsidc.org/data/atlas/index.html). PolarView (http://www.polarview.org/) is another good example. What else is out there?
