This is actually THE thing that people miss when they complain about labelers: every system of labeling is a function of bias (and I'm going to define that in a moment), and this includes larger systems like databases, archives, libraries, and (whenever federation happens) specific instances.
What people are calling "bias" here, me included (because it's shorter), is actually better termed "viewpoint". Moderation is a function of viewpoint. You choose a viewpoint lens through which to moderate and apply it to your policies and actions.
Because I am a good pragmatist, I'm going to follow Dewey's definition of bias, which is a disposition towards something. I'm defining "bias" in this way because the mere act of labeling something is a disposition towards describing some content as objectionable, or some users as problematic.
Bias as a habit seeks conditions favorable to its execution. To this end, we all have biases, or dispositions towards understanding the world in particular ways. So, in some respects, "bias" is coextensive with "viewpoint," except insofar as (philosophically) a bias disposes someone to action.
A "viewpoint," at least pragmatically speaking, doesn't necessarily entail taking action. It might be the way that the world unfolds such that it becomes a field for action (bias) that seeks to reshape the world in line with that viewpoint. Trust me, keeping this in mind will make sense in a moment.
Now, to get into the tech of it all, a database, if we're getting down to brass tacks, is a product of bias. It requires the deployment of particular dispositions towards understanding what is and is not relevant for inclusion within the database such that is reflects our viewpoint on the world.
A database, following Benjamin and Annany, is also an exercise in world building. This is important given the rapid deployment of bullshit engines, but that's not what we're talking about here. If I wanted to create a database of "scientists" and I excluded social scientists from it, that's bias.
That's also the creation of a world wherein "science" and "scientist" consists of a singular way of doing things. I have used my bias in line with my viewpoint to create a world that determines that "social science" is not science, and anything that database is used for will reproduce that world.
