@lnl@screaminginsi.de while I broadly agree with your point, there's some important addendums in my opinion:

a) another issue with "moral" licensing is that you're quickly going to open the can of worms of interpretation and incompatibilities if you try to respect it. we, as a sum of all humans participating in society, can't even agree on what "racism" means and doesn't mean. for another example, some "moral" licenses ban government institutions broadly, others limit themselves to military contractors and law enforcement in specific. but even with the OQL as brought up, am I allowed to use the work as an university employee, or potentially an academic? most universities are in bed with the MIC to some extent, after all. as these proliferate, the issue of "you need to respect the sum total of all licenses that comprise a work" can only get trickier.

b) while it's important to remember the de facto outcomes of choosing a given license - they are legal documents - i think it's also important to point out that a shocking number of people don't care about their work being used widely, and in their case there's no real downside to using a license as a political statement.

c) this isn't even particularly new discourse. Douglas Crockford was doing "moral licensing" as early as 2002, as shown in this somewhat humorous talk: https://www.youtube.com/watch?v=-hCimLnIsDA ; but in the end, JSHint - a fork of JSLint - put in great effort to relicense to MIT/Expat in 2020.