I thought I would explain how I have been marking up occurrence records in RDF in this blog post.
The quad store view of the different RDF entities mentioned in the blog can be seen here:
* Note that some of these links get change in email so you might want to use the
bit.ly bundle above
Although the exact format of these could be changed there are several existing advantages and some possible future advantages.
1) Each data provider could expose their own data to the web without a central authority.
This could be in the form of RDF, RDFa, or as gzipped RDF dump files. (records.rdf.gz)
They would be automatically discoverable on the LOD cloud.
A third party service could add value by pointing to the various repositories, perhaps with an annotation system like or as part of CKAN. (
http://ckan.net/)
2) Data validation in terms of being "well-formed" could be checked with the existing RDF and RDFa validators.
3) The next level of validation e.g:
Is this a valid geographic coordinate?
Does the coordinate match the geographical names listed *(You could even have a service that takes the geo area and determines the correct geographic names etc.)
Does the scientific name match a known species?
Are the other required fields present and well formed.
Is this a new record for that species in that specific geographic area
Would have to be done by a validator created for the community, but it would benefit from the existing tools.
4) All the third party LOD tools and services would be able to use these records.
Respectfully,
- Pete
--
---------------------------------------------------------------
Pete DeVries
Department of Entomology
University of Wisconsin - Madison
445 Russell Laboratories
1630 Linden Drive
Madison, WI 53706
TaxonConcept Knowledge Base / GeoSpecies Knowledge Base
About the GeoSpecies Knowledge Base
------------------------------------------------------------