Yes, there are legitimate reasons here, and the presentation is, indeed, very nice. There's some momentum enabling various hybrid models for wikibase, and allow

Yes, there are legitimate reasons here, and the presentation is, indeed, very nice. There's some momentum enabling various hybrid models for wikibase, and allowing projects such as this to enjoy the freedom of an individual platform, while still using Wikidata as the backend for some or most or all of the data seems promising. I disagree with the idea that this data is inferior in some way to what is expected for inclusion in WD. Their sourcing seems to be impeccable, better than maybe 80 % of WD at the statement-level ("Imported from cebwiki") and certainly adequate on the item-level. Data on individual slaves also strikes me as far more useful for the possibilities that Wikidata offers than academic papers. The latter are important in their own right, but metadata on journal articles is of little use unless you're studying impact factors, academic careers, or the rise and fall of academic disciplines. Structured data on slavery neatly connects with geography, economics, culture, and really any aspect society today. I imagine there were changes in regions of origin over time, for example, and it'd be interesting to see if these can still be found in, say, regional culinary preferences, language, or music.