Topic on User talk:GerardM

Hi :-)

I was skimming through the "families who’ve had the same occupation for five generations" WDQS example from the wikidata newsletter, and saw this odd entry.

I took a look at the articles/references, and those parent attributes for Buffie Johnson (Q21648341) were incorrect - both names are of her teachers, not relatives - Stanley William Hayter (Q641304) and Francis Picabia (Q157321).

I'm just letting you know, in case this indicates a problem with either the tool, or the dataset it was using in that instance. Hopefully you know where to best followup. Thanks!

There is a difference between there being a problem with data and problems with a tool. The fact that some data is wrong is to be expected. At that Wikidata is no better than Freebase. The problem is that we do not accept our own failure and insist on the perfection of others.

I am not OK with the discrimination of Freebase data. I am not OK with the way the Primary Sources tool works. It is in my opinion a disgrace. Thanks,