International Workshop on Improving the Usability of Geospatial Data

Output from International Workshop on Improving the Usability of Geospatial Data held at Ordnance Survey (UK) on 14 June 2017, sponsored by EuroGeographics and EuroSDR is reported below.

This is what the workshop found:

  • Quality metadata is often insufficient to communicate how usable or relevant a dataset is to a user, because it is either incomplete, not easily obtained, not understandable to the non-geospatial professional or it is ignored.
  • Other industries have developed easy to understand visual metrics to communicate relevant information about the product (examples include: food, energy efficiency in white goods, and water consumption in dish washers).
  • Geospatial information is increasingly being used by non-experts in many different domains. (sat-navs are a good example here).
  • The experts identified that different use cases require different quality of data. What is essential for one may be irrelevant for another.
  • What appears to be required are easily understandable quality metrics that relate to specific use cases to enable the user to assess whether the dataset is relevant to their use. Included in the information should be data regarding provenance and lineage.
  • Visual metrics, including colour coding, are easily understood by all.
  • There was some doubt amongst those present on whether the users (and sometimes producers) were able to receive and understand this information.
  • It might be useful to allow users to rate a dataset so that others could see the feedback. For example. 5* - Easy to use, easy to download and free as well or 1* - Expensive and didn’t meet my need.
  • There is a need for quality metadata to be machine readable

The outcome from the workshop was a proposal for combined activity between experts and research with feedback from users. It should be noted that this is not an academic exercise but a practical one

Actions include:

• Identification of a number of datasets and identification of use cases that make use of these datasets.
• Definition of data quality metrics (EuroGeographics QKEN to select)
• Identify and test various ways of presenting this information to potential users including machines

The request for funding would be around asking researchers to identify

• Whether this is the right information for potential users.
• The best way of presenting the information.
• Whether this adds value to the potential user.