Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Government Data Janitors and OSLO

Government Data Janitors and OSLO

This presentation introduces Linked Data for administrators at government services

http://bit.ly/lov-able

Pieter Colpaert

March 26, 2014
Tweet

More Decks by Pieter Colpaert

Other Decks in Technology

Transcript

  1. "Saluting the Flag" by Israel Defense Forces Data janitors, we

    should feel the same about our data as the army does about our country
  2. XKCD: Devotion to Duty As data janitors we should feel

    as close to the data as a sysadmin to a server’s uptime Killing for the data
  3. ELIS - Multimedia Lab Linked Data what? Because it’s impossible

    to store all the world’s knowledge on 1 machine…
  4. ELIS - Multimedia Lab An example Earth is the same

    as The world Earth is a Planet The world has population 7.152 billion Machine 1 Machine 2 Machine 3 World Wide Web
  5. World Wide Web ELIS - Multimedia Lab An example Earth

    is the same as The world Earth is a Planet The world has population 7.152 billion Machine 1 Machine 2 Machine 3
  6. ELIS - Multimedia Lab name type same as Population Earth

    Planet The World 7.125*109 The World Planet Earth 7.125*109 { “Earth” : { “type” : “Planet”, “same as” : “The World”, “population” : “7.125*109” } } <Earth> <type>Planet</type> <sameas>The World</sameas> <population>7.125*109</population> </Earth> <Earth> <type> <Planet> ; <sameas> “The World” ; <population> “7.125*109” . Table / CSV / Spreadsheet JSON XML triples
  7. ELIS - Multimedia Lab Problem Earth is a Planet How

    do we make sure “Earth” means planet Earth on another machine as well? semantic interoperability and the relation “is a”? and “Planet”?
  8. ELIS - Multimedia Lab For example DBPedia: wikipedia in triples

    Earth on dbpedia: http://dbpedia.org/resource/Earth What other planets are there on dbpedia? http://data.linkeddatafragments.org/dbpedia?subject=&predicate=%3Chttp%3A%2F%2Fwww.w3.org%2F1999%2F02% 2F22-rdf-syntax-ns%23type%3E&object=%3Chttp%3A%2F%2Fdbpedia.org%2Fontology%2FPlanet%3E By asking different servers on the Web, different basic questions, really difficult questions can be answered
  9. POTENTIAL REUSERS Data CONSUMERS GOVERNMENT Data OWNER/PUBLISHER RML processor r&wbase

    The DataTank as a SEMANTIFIER SPARQL template CKAN data metadata RDF dumps Cloned Linked Data Suite of Visualizations SPARQL endpoint Triplestore Map Data Publish Linked Data Load Data Publishing Linked Open Data Publish Open Data Meta data catalog Publishing Open Data Application REST API The DataTank as an ADAPTER