Over the last two years, we’ve gathered data on the top-level politicians of almost every country in the world, and made it accessible to developers everywhere through our project EveryPolitician.
Now we’d like to take a step that we believe will benefit more people, and further extend the usefulness of this extensive dataset. We’re proposing to integrate more deeply with Wikidata, to fill the gaps in their coverage and provide consistent, linked data to their global community.
Wikidata is the central storage for the structured data each of its Wikimedia sister projects including Wikipedia, Wikivoyage, Wikisource, and others. Wikidata also provides support to many other sites and services beyond just Wikimedia projects so the combination of EveryPolitician’s data with the reach of Wikidata’s community is pretty compelling.
So in many places, the aims of the EveryPolitician and Wikidata projects are already aligned. We already synchronise EveryPolitician data with the good quality data available in Wikidata where we find it, and we feed back our own additions. As our datasets improve, it seems prudent to combine efforts, and resources, in one place.
You can see our proposal to make this happen here.
If you play an active part within the Wikidata community, or are someone who would benefit from this initiative, we’d very much appreciate your support. Please do add your endorsements or thoughts at the foot of that page if you’d like to see the project go ahead.
Image: Opensource.com (CC by-sa/2.0)
Earlier, the EveryPolitician bot described one way the data benefits from Wikidata’s global network, specifically getting transliterations of politicians’ names.
Interesting idea. What kinds of data are you talking about? Does it include donations?
You can see the kind of data we’re collecting at everypolitician.org. To date, we’ve focussed on data about who the politicians are, but in the future we’re planning on gathering data about what they’ve done too: in particular, quantitative things such as voting records and attendance records. Registers of interest and expenses would also be great. Some of that might never belong in Wikidata, but that’s not to say EveryPolitician won’t collect it. A baseline test for us is not just “is the data available?” but also, crucially, “can it be presented in a consistent, general format?”. That second test is key: if we can’t represent it in a way that’s general enough to apply to different legislatures and different terms, we won’t include it.
A deliberate consequence of this approach is that our data encourages things like comparative research, and (from a civic tech point of view, which is especially important to us) code re-use.