Feature summary (what you would like to be able to do and where): Create a way to filter or prevent the "What came first" games from accidently creating editorial comparisons between different types of ethnic or social violence.

Use case(s) (list the steps that you performed to discover that problem, and describe the actual underlying problem which you want to solve. Do not describe only a solution):

I love the What Came First games: good place to learn things about history. Though "What came first" comparisons are normally relatively innocuous when focused on famous people or events,; however, I have encountered an unusually high rate of comparisons between violent events focused on ethnic minorities. This is in part a product of probability and how the lists on EnWiki are built: they include a lot of wars and battles, but the editorial intention was for an all-inclusive list of all events, not for small sets of events to be randomly called out and compared. If we were, for example, creating timelines of colonial violence or ethnic cleansing, where those were the main thing being compared: the on-wiki community would have had much more extensive discussion about appropriate comparison and historiography. In my volunteer capacity, I have been part of several deletion discussion on EnWiki where our consensus was to delete the article for this kind of historiographic problem (i.e. not having expert opinions supporting such a comparison).

The accidental comparison of these acts of violence, introduces real issues in historiography that could create very abrupt, and harmful, representations when you are randomly shown the events together. For example, if you are someone from the effected communities by this violence: you are being represented with a comparison your violent history with another minority's violent history. Or, if you are effected by something like ethnic cleansing or mass violence AND that event is compared to another event that happened in the last 50 years: there is very good chance that you that you are introducing interpretation by comparison for peoples effected by that violence. Normally, this kind of comparison would have sensitive contextualization if it were published in other Wikipedia pages, explaining how and why those choices of interpretation are being made, and laying out various historiographic framings for the comparisons. These comparisons are accidently editorializing in a way that is going to disproportionally effect underrepresented groups in our global audience.

Here are a couple examples of the problem:

I also encountered another which compared an ethnic cleansing campaign in New Zealand in the 19th century with one in the United States in the same time period that I didn't screenshot.

When I first saw this in June, I went through 3 days of the source pages on English Wikipedia and think that: if you are deploying a random algorithm to select events to compare historical events, you could apply a filter that excludes Wikidata items that have: battle ((Q2890384), armed conflict (Q1510144), execution (Q3966286) or mass murder (Q750215 ) in their "instance of" (P31) on Wikidata. We could probably look for a few more, or place some limits on it (i.e. most of our contemporary social experience of this violence, is in the last 400 years). This would reduce the likilhood of two violent events effecting current readers being compared to eachother, and allow the other events about this violence or colonial/racial/ethnic conflict to be a much smaller proportion of the pool, so would not have as many of the accidental comparisons.

Creating some boundaries on what are appropriate comparisons, and having the ability to exclude certain kinds of events from comparisons, would make sure we aren't accidently introducing historiographies (historical interpretive methods) of violence, coloniality, etc, through our algorithm.

Benefits (why should this be implemented?): We build a mechanism into the game that allows us to mitigate accidental harm, from comparison of historical events.