They mention their source for notable people was Freebase, which iirc use DBpedia which in turn uses English Wikipedia. This is the source of the bias. I do not

They mention their source for notable people was Freebase, which iirc use DBpedia which in turn uses English Wikipedia. This is the source of the bias. I do not know if information from the other Wikipedias is available via Freebase. But I think you're right that it would be useful to note at least somewhere close to the title that the provenance of this data skews it quite a bit. Edit: the paper itself is centered around providing this framework for looking at cultural history. The author of the blog post for Nature is the one who made the over-reach in the title.