This hack is really just a programming excercise :-)
What data: The List event 2012
What I've done with it
- I've written a Haskell program that...
- Transforms the XML into RDF
- Defines a schema for The List specific fields
- Reuses ontologies for field e.g. location (wg84pos)
- Serializes the RDF into the Turtle file format
- The program goes out to to the The List web API to grab more info about locations
- Goes out to Linked Geo Data and discovers the nearby pubs to the venue (within 100 metres)
Why?! have I done it?
- Linked Open Data is really awesome.
- It'd great to have The List on the cloud of linked open data
- ... on this map: http://richard.cyganiak.de/2007/10/lod/
- (More exposure on the semantic web for The List events!!)
- Venue location info is disambiguated, so all kinds of interesting information can be sought from the semantic web (finding pubs is just a proof-of-concept).
- list_mappings.hs <- the Haskell code to generate the turtle file
- list.ttl <- the file generated
- README <- this file
- You can install a SPARQL endpoint... point it at the .ttl file, and you have a semantic search endpoint for free out-of-the-box.
<#fringe> rdf:type ja:RDFDataset ; rdfs:label "FringeData" ; ja:defaultGraph [ rdfs:label "" ; a ja:MemoryModel ; ja:content [ja:externalContent <file:/location/to/rdf_output.ttl> ] ; ] ; .
- Get in touch with me.. I'd happily add more (email@example.com).