Ric Roberts: Data Platforms and the Data Value Chain

This deck supports Ric's talk about the wider perspective of data publishing, where the act of publishing itself sits within that, what it entails and the advantages it brings. For all talks from the day, head to: http://power-of-data-2017.swirrl.com/

Transcript

1. Data Platforms
   and the
   Data Value Chain
   CTO, @RicRoberts
   

   View Slide

2. data
   

   View Slide

3. data
   You are here.
   

   View Slide

4. profit!
   

   View Slide

5. profit!
   You want to be here.
   

   View Slide

6. profit!
   Examples
   • Deciding the best place to put a new school
   • Benchmarking hospitals
   • Working out the impact of poor air quality on health
   • Calculating the cost of increased flood risk
   

   View Slide

7. View Slide

8. data profit!
   ?
   

   View Slide

9. data
   collect
   clean
   curate
   

   View Slide

10. Examples
    • Surveys of various sorts
    • Administrative systems
    • Sensors e.g. detecting river flow
    • Social media
    data
    collect
    clean
    curate
    

    View Slide

11. data
    ?
    profit!
    

    View Slide

12. data
    use
    profit!
    

    View Slide

13. use
    • exploring
    • filtering
    • aggregating
    • downloading
    • exporting
    • analysing (data science!)
    • generating reports (xls, pdf, doc, ppt)
    • using it in interactive apps or visualisations
    • sharing results
    

    View Slide

14. data
    ?
    use
    profit!
    

    View Slide

15. data
    connect
    use
    profit!
    

    View Slide

16. connect
    • A common set of names for the things in the data.
    • A shared, documented and understood model of the data.
    • An agreed set of technologies for communicating and
    manipulating the data (standards!).
    • The data needs to be in a place people can get to it, in an
    relevant format (with a licence).
    

    View Slide

17. https://www.flickr.com/photos/kewl/7006904747
    

    View Slide

18. connect
    In computing, linked data is a method of publishing
    structured data so that it can be interlinked and become
    more useful through semantic queries. It builds upon
    standard Web technologies such as HTTP, RDF and URIs, but
    rather than using them to serve web pages for human
    readers, it extends them to share information in a way that
    can be read automatically by computers. This enables data
    from different sources to be connected and queried.
    — Wikipedia
    

    View Slide

19. connect
    • A common set of names for the things in the data.
    • A common set of names for the things in the data.
    • An agreed set of technologies for communicating and
    manipulating the data (standards!).
    • The data needs to be in a place people can get to it, in an
    relevant format (with a licence).
    

    View Slide

20. connect
    • A common set of names for the things in the data.
    • A shared, documented and understood model of the data.
    • An agreed set of technologies for communicating and
    manipulating the data (standards!).
    • The data needs to be in a place people can get to it, in an
    relevant format (with a licence).
    

    View Slide

21. connect
    • A common set of names for the things in the data.
    • A shared, documented and understood model of the data.
    • An agreed set of technologies for communicating and
    manipulating the data (standards!).
    • The data needs to be in a place people can get to it, in an
    relevant format (with a licence).
    

    View Slide

22. connect
    • A common set of names for the things in the data.
    • A shared, documented and understood model of the data.
    • An agreed set of technologies for communicating and
    manipulating the data (standards!).
    • The data needs to be in a place people can get to it, in an
    relevant format (with a licence).
    

    View Slide

23. https://www.flickr.com/photos/iwannt/8596885627
    

    View Slide

24. data
    ?
    connect
    use
    profit!
    

    View Slide

25. data
    publish
    connect
    use
    profit!
    

    View Slide

26. swirrl.com
    

    View Slide

27. publish An (RDF) Graph Store
    Apache Jena
    

    View Slide

28. publish Extract, Transform, Load (ETL)
    grafter.org github.com/swirrl/grafter
    

    View Slide

29. publish Drafting and publication workflow
    

    View Slide

30. publish A User Interface
    

    View Slide

31. publish APIs
    

    View Slide

32. View Slide

33. data
    publish
    connect
    use
    profit!
    collect
    clean
    curate
    What’s limiting the effectiveness of
    this value chain?
    • Cottage industry of skilled individuals
    • Data preparation is not always considering bigger picture
    • Those expending the costs != those reaping the benefits
    • Availability of skilled data analysts
    • Lack of guidance and standardisation
    

    View Slide

34. data
    publish
    connect
    use
    profit!
    collect
    clean
    curate
    

    View Slide

35. Data Platforms
    and the
    Data Value Chain
    CTO, @RicRoberts
    

    View Slide