better machine readability for knowledge

March 26, 2021 9:12 AM

Wikidata is "Wikipedia's not so little sister": "a free and open knowledge base that can be read and edited by both humans and machines." Wikidata aims to be "accessible to everyone in their language without privileging any particular language by design". Also: "A lot of technology today is trying to simplify the world by hiding necessary complexity and nuance. Conflicting worldviews need to be surfaced. Otherwise we take away people’s ability to talk about, understand, and ultimately resolve their differences. Wikidata is striving to change that by not trying to force one truth but by collecting different points of view with their sources and context intact. This additional context can, for example, include which official body disputes or supports which view on a territorial dispute."

Disclaimer: I know the author of this explanation.

posted by brainwane (5 comments total)

This is really cool. Thanks!

For example, it was not possible to easily find an answer to the question of what are the biggest cities with a female mayor because the necessary data was distributed over many articles and not machine-readable.

I would have a very hard time coming up with a universal definition of a mayor, or a city, that doesn't encode many regional and language-based assumptions. But, that doesn't mean we shouldn't try. (And, the author clearly knows this and has thought about it a lot more than me.)
posted by eotvos at 10:18 AM on March 26, 2021

Oh hey this is helpful. I was just yammering about Wikidata the other day when I was at an edit-a-thon for medical librarians talking about how it was another way people could help Wikipedia be better, by making the sure the metadata was good. I used to be really confused (and, I'll be honest, kind of put off) by Wikidata but I've been using it more with the Women in Red project specifically as this article mentions, to find women in categories who are famous ("notable" in Wikipedia parlance) and who didn't have pages. It's so helpful. Thanks for this.
posted by jessamyn at 3:39 PM on March 26, 2021

Inexplicably, my federal agency employer blocks access to wikidata, claiming it is a source of malware. Super frustrating, since it would be a really useful tool to in the linked data space.
posted by rockindata at 7:23 PM on March 26, 2021 [1 favorite]

This is a personally interesting connection, because I am literally skimming this wikipedia article, "Homophobia in ethnic minority communities", and the top contains a warning that "the examples and perspective in this article deal primarily with the United Kingdom and the United States and do not represent a worldwide view of the subject".

That doesn't jive with my intuition. I think intersectionality is a pretty universally applicable idea, and so there's a fine line in editorship that acknowledges, okay these examples and studies are about the US and the UK, versus claiming on a conceptual level that they are not representative of other views. So computerizing citations and sources can help make information authoritative, but in the end do we want an encyclopedia where one page says intersectionality exists and another page that denies it and is equally valid and has "supporting data with sources", because it is just a conflicting worldview? It's a moral task that editors and authors cannot punt indefinitely, so what is the limit here, etc. Another name for not trying to force one truth is both-sidesism.
posted by polymodus at 12:37 AM on March 27, 2021 [1 favorite]

This additional context can, for example, include which official body disputes or supports which view on a territorial dispute.

Another counterexample, since Taiwan is not recognized as an official body in the WHO, they're effectively excluded from much of international health politics. So Wikidata could do the same, by disrecognizing real bodies as "non-official", they could use this policy to exclude voices. So who gets to make the final call? Etc.
posted by polymodus at 12:49 AM on March 27, 2021 [2 favorites]