Typically the debate is between running two competing functions. One group's essentially takes the outcomes produced by two different populations, and takes an

Typically the debate is between running two competing functions. One group's essentially takes the outcomes produced by two different populations, and takes any differences as sufficient evidence that the systems themselves are biased towards one population or another. Implicit in their argument is that the populations serve as identical inputs. This is the argument of the Ibram X. Kendi/Robin Diangelo, collectivist/CRT types of the left. The other group's essentially says that outcomes produced should be proportionately relative to inputs, and consistent between individuals such that comparable inputs produce comparable outputs. Implicit in this argument is that race and gender are almost never relevant inputs to any outputs, such that two individuals with comparable inputs aside from race or gender should be expected to net comparable outputs in most circumstances. This is the argument of the individualist/liberal types of the right. How I personally classify these problems is that the first solution is "bad code," it's a bad diagnosis and produces bad solutions. The second solution is "good code," it would produce a just and stable system. However, the second solution ignores "bad data" which is already in place, whereas the first solution implicitly picks up and addresses the bad data as the bad data will largely contribute to the different outputs. And what I mean by good code vs bad data is that previous "bad code" (i.e. procedural systems producing outputs from inputs) have put individuals from different populations in unequal circumstances. So while the system I described as being the better of the two for producing just society is better at actually serving as the basis for our procedural systems, it is insufficient in addressing the "bad data" which is already in place, such that people from different populations, and individuals in general, are given such drastically different starting positions, that it doesn't matter that their inputs are going to net them proportional outputs so long as their inputs were so dramatically hampered from the get go.