Implementation of the disco-spec, the DDI Discovery Vocabulary.
Switch branches/tags
Nothing to show
Clone or download
Latest commit 5314d7e Apr 16, 2018
Failed to load latest commit information.
UML class diagrams
src fix all warnings and errors regarding javadoc generation Feb 19, 2018
.gitlab-ci.yml Update .gitlab-ci.yml Feb 19, 2018 add pipeline badge Apr 16, 2018
pom.xml pom.xml -> decoupled from parent project; added compiler plugin and Feb 19, 2018

pipeline status

This package contains the implementation of DISCO, the DDI Discovery Vocabulary Specification.

DDI-RDF Discovery Vocabulary (disco)

The Data Documentation Initiative (DDI) is an acknowledged international standard for the documentation and management of data from the social, behavioral, and economic sciences. Statistical domain experts, i.e. representatives of national statistical institutes and national data archives, and Linked Open Data community members have developed the DDI-RDF Discovery Vocabulary – based on a subset of the DDI - in order to support the discovery of statistical data as well as metadata. This vocabulary supports dentifying programmatically the relevant data sets for a specific research purpose.


All classes, datatype properties and object properties are modelled according to turtle file, UML models, and the specification itself. Please note that the javasdoc in this package is taken from the specification texts.


In order to make the persistence of the model easier, all classes are enriched with JPA-annotations, which annotate all fields and relations discribed in the disco-spec. It was payed attention to just use JPA annotations, hence not being dependent to a specific implementation, e.g. Hibernate.

@Inheritance( strategy = InheritanceType.JOINED )
public class Variable extends Concept
	// properties
	protected LangString description;
	// relations
	@JoinColumn( name = "analysisUnit_id" )
	protected AnalysisUnit analysisUnit;

Please have a look at the disco-persistence-api for an API for persisting the data model.


There exists a persistence implementation for this model according to the API called disco-persistence-relational. It uses Hibernate as a object-relational-mapper.

Further physical storage types may be implemented.


It is assumed to have the validation NOT coupled to the model, thus the existence of datatype and object properties is expected to be managed by the business logic.



License & Copyright

Apache License, Version 2.0

© GESIS – Leibniz Institute for the Social Sciences.