European Commission logo
INSPIRE Community Forum

What makes INSPIRE difficult - Top 10

If you have responsibility for implementing INSPIRE compliant datasets and services in the topographic and cadastral reference data communities, what barriers have you come across? We would like to put together a Top 10 of these problems and see if we can solve them.

 

Feel free to expose the results of your experience.

  • Anja HOPFSTOCK

    By Anja HOPFSTOCK

    1. Data transformation (Spain, Netherlands, France)

    • data transformation is a big challenge and will need lot of resources
    • GML (complex, heavy)
    • transforming simple attributes into feature types
    • handling of transport property types
  • Alexander RAMAGE

    By Alexander RAMAGE

    The biggest worry I have for the implementation of INSPIRE is Data Transformation.

    The data specifications that are used day to day are "flat" and the INSPIRE DS are hierarcical!

    Alex Ramage

  • Ulrich DÜREN

    Of course, data transformation is a big challenge. But it is done by experts and only once. What worries me more, is the lack of usability of the resulting complex data. "Standard" GIS are not able to handle all those relations, features without geometry, properties that have become features with identity, linear referencing, coverages, etc. etc.

    E.g., one address feature in the source data set becomes up to five features including complex data types in the target data set.

    In the end, you need a special application for every INSPIRE-Theme...

    Ulrich Düren

  • Dominique LAURENT

    By Dominique LAURENT

    I completely agree with Ulrich.

    In IGN France, we are struggling to make the data transformation: this is hard work but feasible.

    Our main concern is also about usability of the INSPIRE data; we had launched some studies about this topic (testing integration of INSPIRE data into GIS, collecting feed-back from first users) and the conclusions are rather negative: it is not easy at all to use INSPIRE data.

    This is especially true with themes AD and TN where the geometry and the semantics are in separate feature types (semantics as AddressComponent or as TransportNetworkProperties); this makes the resulting data very difficult to use, even for simple consultation.

    IGN France would encourage the MIG to provide common simple, flatten logical models based on the INSPIRE conceptual models (possibly with some minor lost of content) that might be easily used by current client applications.