From Botanical Knowledge
Jump to: navigation, search

The Portal will be a node in the web of data. It will ingest from FNA/CharaParse, multiple instances of Specify, multiple FP networks, and other data and image repositories. It will feed into FP networks, the automated workflows of its users, and the semantic web in general. It is a best practice to use HTTP URIs as identifiers, and to build community consensus around identifiers where possible.

Therefore, however the Portal reasons internally, data should be exported using canonical URIs, where possible, for taxa, authors, institutions, geolocations, and descriptive properties.

For example, we will have a page called "Alliaria petiolata". Its Special:ExportRDF page will, by default, include statements looking roughly like this:

 <swivt:Subject rdf:about="”>
   <swivt:page rdf:resource=”"/>
   <rdfs:isDefinedBy rdf:resource=""/>
   <property:invasiveIn rdf:resource=""/>

To interoperate with the rest of the semantic web, we will want to:

  1. use a different identifier, e.g., for the subject of the page;
  2. use a different identifier, e.g., for "property:invasiveIn"; and
  3. use a different identifier, e.g., for Canada.

For example:

 <swivt:Subject rdf:about="">
   <swivt:page rdf:resource=""/>
   <rdfs:isDefinedBy rdf:resource=""/>
   <ebio:invasiveIn rdf:resource=""/>

TO DO: Illustrate the use of Import Vocabulary to support this.