[LTER-im] Discussion with Open Knowledge folks

[John Porter]

This is a FYI about a videoconference I held with Daniel Fowler of "Open
Knowledge International" (https://okfn.org/) - a technology development
group originating in the UK, but with worldwide links (Dan was in
Ethiopia when I spoke with him).  We were bought together by Andrew
Sallans of the Center for Open Science (http://cos.io).

The discussion focused on information exchange about the organization
and objectives of our two organizations, with some more extensive
discussion of areas of common interest and potential opportunities for
interaction.

Open Knowledge is working on promoting and developing what they call
"Frictionless data" (http://data.okfn.org/) which they characterize as
"lightweight standards and tooling  to make it effortless to get and
share data".  Like us, they have the concept of a "data package" - which
in their case includes metadata in JSON (rather than XML) plus a CSV
file.  They are working on extending their metadata to include different
"profiles" that can be used to match up their metadata with different
domains.  They are also currently engaged in developing standards for
the descriptions of units (I pointed them to our unit dictionary).  

I told him about LTER and EML, most of which was new to him, and
discussed the similarities and differences between EML and their tabular
data package, the most conspicuous of which was that they don't
accommodate multi-line headers (an issue for VCR and some other sites
data).  In general, EML is more specific, so that writing a stylesheet
to go from EML to their package format might be relatively easy - apart
from the header issue.  We also discussed the possibility of an
ecological-data "profile" for their data package that might be
translatable into EML.

They have funds for software development and are developing tools
similar to our software generation tools among others.  We discussed the
relative advantages of crosswalking metadata, from the standpoint of
making our data more accessible via additional repositories and tools
(e.g., CKAN, BigQuery, datahub.io) and similarly for their data with
DataOne, code generation tools etc.  There may also be opportunities to
collaborate on the development of standards for units and other metadata
elements that would promote automation and interoperability.

I left him with the names of some additional information managers that
he might like to contact and an array of URLs
(http://portal.lternet.edu, http://im.lternet.edu,
http://unit.lternet.edu, https://knb.ecoinformatics.org/#tools/eml), so
I anticipate that at some point we will have further contacts with the
OK folks.   However, if you are interested, feel free to contact Dan
directly (he is CC'ed on this message) if you see some opportunities you
would like to discuss with him.

   -John Porter

-- 
John H. Porter
Dept. of Environmental Sciences
University of Virginia
291 McCormick Road
PO Box 400123
Charlottesville, VA 22904-4123