Wikidata for Research – presenting the team and a first sketch of the workplan

Two weeks ago, I wrote about bringing Wikidata and research more closely together on both the technical and the community level through the Wikidata for Research project. Since then, seven institutions and about 20 Wikimedians have come together and sketched out a skeleton for a work plan that they want to expand into a full-fledged EU research proposal over the following four weeks. The proposal shall address both an increased use of research data in Wikidata and of Wikidata in research workflows, including in citizen science contexts. The proposal is being drafted in the open, and anyone can join in.

The workplan currently consists of six workpackages, which will briefly be introduced here:

Workpackage 1 in such projects traditionally deals with management and coordination of the project. We follow this pattern, but have added communication on top of that – both between the partners and with the wider research and Wikimedia communities.

Workpackage 2 is concerned with software development at the interface between Wikidata and research. This will include tools to enrich Wikidata content with information from scholarly resources, and to harvest content from Wikidata. Code hardening of existing tools, speed-up of work on research-relevant features that Wikidata is still missing are also being considered in this context, as are improved documentation and research-related work around Wikibase, the semantic backbone of Wikidata.

Workpackage 3 is concerned with identifiers of research-relevant concepts and Wikidata as a relay station that maps those identifiers and their relationships between different scholarly resources. This part follows most closely in the footsteps of the Wikidata activities of the Gene Wiki project, which were described on this blog in November.

Workpackage 4 focuses on enriching Wikidata with research-related information. This includes the identification of suitable kinds of data and facilitating their porting to Wikidata, along with mechanisms for verification and handling provenance.

Workpackage 5 is concerned with uses of Wikidata in research contexts. Here, some specific use cases will be explored, and workflows set up that facilitate handling relevant information through Wikidata. This includes, for instance, the usage of Wikidata identifiers (e.g. Q153 for ethanol) in the context of scholarly communications, or visualizations of relationships between scholarly concepts or scholars (e.g. this network of Ancient Greek philosophers).

Workpackage 6 will produce tutorials, courses and other open educational resources and organize training events related to Wikidata in research contexts.

Coordinated by the Museum für Naturkunde (which was briefly presented in the previous blog post), the group of partners also comprises

• Wikimedia Deutschland (which will handle the technical development related to Wikidata),
• Maastricht University’s (whose Department of Bioinformatics will oversee the work on identifiers ranging from small molecules to meteorites to paintings),
• Open University of Catalonia (which is a virtual university and will handle the education and training part),
• The European Library (which is an umbrella organization of national libraries in Europe and will provide data and use cases across all scholarly domains of knowledge),
• Université Paris-Sud (which operates a cluster of laboratories ranging from particle physics to neuroimaging and will provide data and use cases on that basis)
• Universidad Politécnica de Madrid (which operates the Spanish DBpedia node and will focus on interactions between DBpedia and Wikidata).

The proposal is being developed in response to a funding call by the European Commission, and if it is successful, the project partners will receive funding to carry out the work described in the proposal.

In addition to that, the project is looking for associated partners. These would not receive funding from the European Commission but can be involved in the project in an advisory capacity. A number of institutions have already signalled their interest to serve in this role, including the York Museum Trust, which oversees a range of collections spanning from arts to humanities to media and was presented in last week’s Signpost.

The details for all of this are still in flux, and will remain so for the rest of the year. You can help turn this idea into a useful whole that is good enough to actually get a chance of being funded to improve the interaction between Wikidata and research.

Text by Daniel Mietchen