The reconciliation algorithm that merges data from multiple sources (= multiple RDF graphs) needs some further work:
[x] Detect and remove duplicate people in authors/maintainers/contributors
[x] Merging blank nodes from multiple graphs (#12) may still lead to some incorrect results: maybe we can tackle this and the above one by assigning some temporary IDs which we strip later
[x] Expand the list of properties that may only take a single value
[ ] Handle merging of dates (favouring the earlier/latest date in case of a conflict)
The reconciliation algorithm that merges data from multiple sources (= multiple RDF graphs) needs some further work: