Datavindplaats (literally, the place to find data) aims to be the most complete offering of datasets and APIs provided within the region of Flanders.
Historically, the data landscape in Flanders was a collection of many catalogues; each one trying to provide the best service to its audience with often overlapping information. Among the most notable were the catalogue for geospatial data, the catalogue for open data, and the catalogue of non-public APIs. Because the catalogues were not properly synced and were showing similar information in different ways, searching for a dataset or API required having knowledge of all these portals.
Another observed roadblock in the use of these existing portals was the implicit assumption that visitors had “technical” background, causing the user interfaces to not appeal to a broader public.
Moving forward, instead of providing a portal for each individual data or information domain, there is now the possibility to gather all domains into one user-friendly portal, the Datavindplaats. The information shown is based on the umbrella concept, leaving out many domain-specific concepts, so the user interface can remain simple, clear, easy to use and lovable. To achieve this and unify the catalogues, the metadata descriptions had to be aligned and harmonised.
In the past when describing datasets or APIs, multiple metadata standards were used. For Geospatial datasets and services, ISO standards were used. For Generic Open data, the DCAT-AP (1.2) standard was used. For the so-called ‘closed’ datasets and their services, there was no standard yet. To allow Datavindplaats to present all these in a user-friendly and similar manner, all metadata has now been bridged semantically into the consolidated DCAT (2.0) standard using SEMIC specifications. Bridging the gap between standards improves exchange, reduces and simplifies maintenance, increases efficiency and consistency… and keeps it as simple as possible for data providers in describing their data once-only, while reaching the widest possible audience.
The harmonisation of the metadata and the presence of a user-friendly interface, allows the Datavindplaats to start a journey into a new area: to allow privately-owned datasets in the open catalogue. Any successful data solution is a combination of firm and trusted data from the government with privately-owned data. By bringing the two together, the Datavindplaats becomes a facilitator of the next generation of data solutions.
If you have questions about SEMIC specifications or would like to take a similar journey, please contact us.
Future outlook
Next steps in our regional use case, is the migration to DCAT 3.0, which will allow adding versions and series to Datavindplaats, and aligning its implementation across the different domains. Alignment on data quality is continuous, for example by creating agreements on the use of specific thesauri or code lists so that exchange with the Belgian and European data portals is simplified as well. And certainly the uniqueness of metadata descriptions has to be handled in the federated ecosystems when harvesting catalogues into others, to avoid duplicates and keep proper references to the original source metadata descriptions and catalogues too.
Finally, there is still a lot of work to increase the usability and make the Datavindplaats even more lovable. By continuing to increase the amount of datasets and APIs in the catalogue, there is a constant need to improve the search experience of the catalogue, to guide the searching user to find what they are looking for.
|
|