Impressive.

In the linked article the author mentions that there are multiple definitions of racism and people often aren't clear about which one they use; and then decides to use the one without "..., but only when white people do it" as a default. And says that it is okay if white authors decide to write only white characters, but if they write also non-white characters they should describe their experiences realistically. (Then in the comments someone asks whether saying that every human being is racist doesn't render the word meaningless, and there is no outrage afterwards. Other people mention that calling someone racist is usually used just to silence or insult them.)

I am not sure whether this even should be called "social justice". It just seems like a common sense to me. (This specific article; I haven't read more from the same author yet.)

Somewhat related -- writing this comment I realized that I am kinda judging the sanity of the author by how much I agree with her. When I put it this way, it seems horrible. ("You are sane if and only if you agree with me.") But I admit it is a part of the algorithm I use. Is that a reason to worry? But then I remembered the parable that all correct maps of the same city are necessarily similar to each, although finding a set of similar maps does not guarantee their correctness (they could be copies of the same original wrong map). So, if you spend some time trying to make a map that reflects the territory better, and you believe you are sane enough, you should expect the maps of other sane people to be similar to yours. Of course this shouldn't be your only criterium. But, uhm, extraordinary maps require extraordinary evidence; or at least some evidence.