User:PKM/bday5

Some random thoughts on the occasion of Wikidata's fifth birthday

I worry. It's part of who I am, and I suspect it is part of what made me a good project manager. (What might go wrong? How should we react to that? Can we prevent it?)

In Wikidata-land, I worry, will we build structures so full of layered qualifiers that people can't find things successfully? Will the Wiki way of working lead to so many different approaches to modeling that no one query can find all the relevant items? Will military units have parent organizations and subsidiaries, or have parts and be parts of? How would someone creating a query know which to look for? None of these are insurmountable problems. Perhaps future smart query engines will look at the modeling of representative cases and deduce how the queries should be structured. Perhaps future bots will ensure that parents and subsidiaries also have parts and are parts of.

We often talk about how young Wikidata is as a project, and it's true. I am frequently amazed by the basic things that are still missing. (We have "inventing" but not "invention"? What?) But I am also frequently delighted by the depth of modeling we do have, and the obvious thought and attention to detail that goes into such modeling.

So where might we go when we're not quite so young anymore?

The other day I was watching one of those new-fangled "videos" on a TV network news site that consists of a dozen or so facts in large display text, each laid over a relevant stock photo, and played as a slide show. And I thought - once we have structured Commons and Wikidata working together, can people build apps that will generate (or help curators generate) these sorts of videos, on health care and science for kids and art and history? Can we unlock that vast store of knowledge that will be in Wikidata in interactive experiences that will teach and inspire the world?

When I get frustrated disambiguating a dozen places called "James Island", I think about the things that will be possible when the interlinking and the modeling (and maybe the auto-describing?) catches up to our vision. And I remember that Wikidata has the potential to impact the world in ways we haven't though of yet, and maybe be even more significant in the next decade than Wikipedia has been in the last one. And working to make it better is a pretty good way to spend some time.