Digitalisèr.dk is a collaborative platform and repository for XML Schemas, datasets, software and technical standards shared by groups with a common interest in digitalising the Danish public sector.
It is a unique repository that supports digitisation in Denmark that brings together key resources, recommendations and guidelines on IT, communication and IT development. The main goal of the repository is to strengthen cooperation between public and private parties.
| Facts | Owner | Danish Agency for Digitisation |
|---|---|---|
| Target Audience | IT practitioners Public administrations Private sector |
|
| Assets | More than 17.000 XML schemes 169 software assets 750 data sources 236 standards |
|
| Federation Benefits | Expected Benefits | Raise awareness about what is going on in EU Member State Public Administrations. |
| Mapping to ADMS | Language | Danish |
| Licence | Licence metadata not yet available. Assumption that schemes on Digitaliser.dk are public domain. |
|
| Strengths | Mapping of the Digitaliser.dk artefacts on ADMS can be easily done. | |
| Export of Metadata | Format | ADMS RDF |
| Transmission to Joinup | Transmission Protocol | Metadata harvesting (solution 3) with the use of existing API |
Interview with Digitaliser.dk
Adam Arndt
After working as a web and database developer for 5 years, Adam Arndt changed domain to data standardization and IT architecture where he has been active since 2005 in improving IT and data use in the Danish public sector and in international trade. Digitaliser.dk is expected to join the ADMS-enabled federation of semantic asset repositories in June 2012.
Joinup team: Which benefits do you expect from the federation?
Adam Arndt: The main benefit that we see from joining the federation is that it can help raising awareness about what is going on the Public Sector within the Member States of the European Union. Besides sharing our assets with the federation, we also expect to use the export and search APIs of the central repository to give our users insight in what is going on in standardisation organisations and other Member States.
Joinup team: What are your experiences with mapping to ADMS?
Adam Arndt: The mappings can be done at two levels: At concept level and at the level of the controlled vocabularies.
: The metadata in Digitalisér.dk can be fairly easily mapped to the ADMS metadata. Resources on Digitalisér.dk can be mapped to the concept of “Asset” in ADMS. Every user on Digitalisér.dk can share “resources” within the context of a group to which he belongs. Every resource consists of one more or more artefacts (files), which maps to the concept of “Distribution” in ADMS. The scope of ADMS is limited to semantic assets, so only the resources of type XML schema on Digitaliserr.dk are relevant for the federation. Digitalisér.dk does not store any structured information for geographic coverage or language. As a solution, default classifiers like “Danmark” or “Danish” could be used. The main domain taxonomy used is FORM, the public sector reference model. This can be mapped to the EuroVoc domain concepts. One point of difficulty for us will be the licensing information, for which we currently do not offer any structured metadata. We assume that the XML schemas on Digitalisér.dk are public domain.
Currently, there are about 16000 XML schemas registered on Digitalisér.dk. This number is increasing every day. The large number of schemas may cause an imbalance in the search results of the central repository, especially if the metadata of these schemas are of insufficient quality. There are several possible strategies to mitigate this. One strategy could be to work with heuristics, for example by filtering out semantic assets that have a description length of less than 5 words. Another strategy could be to manually filter out the 10-15 semantic assets that are directly derived from Danish law.
Joinup team: How will you export the Digitalisèr.dk metadata?
Adam Arndt: On Digitalisér.dk, every resource has a metadata artefact. This is an XML file with metadata descriptions about the particular resource. The GET Resources API simply returns this XML file. We foresee to make use of our existing API to export the description metadata of our XML Schemas in a machine-readable format. Using an XSLT, the XML document can be converted into ADMS RDF.
Joinup team: How will the description metadata be exchanged with Joinup?
Adam Arndt: Given that we already have an open API in place that allows for metadata harvesting, we very much favour solution 3 (metadata harvesting) for transmitting our resources to the federation. This solution will not entail any additional work from our side. We have foreseen not to make any development to Digitalisér.dk in 2012. This means that it is currently unlikely that we will be able to adapt our API to produce ADMS XML in time to join the federation in June 2012. The conversion therefore has to take place after metadata is harvested from Digitalisèr.dk.
Nature of documentation: Other
