It seems like a mistake to invent racial terms out of whole cloth for no reason. It shouldn't be done.

> It shouldn't be done.

That's linguistic prescription. "This word means what it means and if you change it you're incorrect".

I'm not saying that. My position is it shouldn't be done because creating new slurs for no reason is stupid. A prescriptivist and a descriptivist could both agree to that, though they'd disagree with each other on whether the idea of the change is legitimate.

Which is probably a similar position to ccmcarey's original comment. No position was taken on whether the change breaks the rules of language or not, the argument was that either way the change is being bought on by ignorance of both normal usage and the lineage of the word (ie, potentially in defiance of both prescriptivist and descriptivist logic).

This isn't really an issue of prescriptive vs descriptive philosophy. Although the descriptivists will be hopping mad.

Not directly no. But they enforce the unconscious belief that "black == bad" and "white == good". Read up a bit on unconscious/implicit bias [0]. There was an online test floating around that measured your response rate to a white face or a black face and those results were eye-opening.

Also consider how other industries have approached this problem[1] and how things have changed there.

[0] https://www.sciencemag.org/news/2020/03/meet-psychologist-ex...

[1] https://www.theguardian.com/women-in-leadership/2013/oct/14/...

No, they don't. What your sources did was bring to light the current stereotypes in people's head, not how they were created. What you want to do is change those stereotypes, not mingling with the words that only communicate the stereotypes. It's about the concepts in the heads, not the words.

If you want to change the stereotypes (which you can't remove, just change), you have to provide different pictures: Showcase black business men/women, talk about the performance of Obama, have high achievers with diverse background do talks in schools, or sports stars do shows with kids, do shows with female arab DJs and so on. In short: Give the brains of people input that forces them to adapt their stereotypes.

This differentiation of signified and signifier often is hard to get right if your own moral system already conforms to the social goal, because in that case the word seems to be equal to the concept/meaning. So we just have to transport to word to transport the meaning, right? But this isn't how language works. Words can only trigger the frame they belong to if the other person already has the concept in her mind.

Words can also be used to put those concepts in people's minds.

So you can use words to describe things (like we do right now), and thus hope to invoke mutual understanding, but you can't put a new concept into another person's head by inventing a word. You can trigger a concept already there if the other person already associates a specific meaning with a specific word, though.

So if you want to better the situation for e.g. African Americans in the US, replacing "master branch" with "main branch" has no effect, because a) this master is not the master/slave master - the words may have had identical meaning (I don't know, perhaps both meanings have a common ancestor), but today the word "master" in the context (=frame) of source code respositories means something completely different than master/slave. Just as "slave" in the US doesn't mean "person of slavic origin" anymore.

But more importantly you don't change the stereotype of African Americans this way. That you'll only achieve by constantly pushing different images of African Americans in the relevant contexts, like, off the top of my head, a collective day where every github user with darker skin starts to use a real profile picture on github.

You can justify pretty much anything with that reasoning. If you follow through with that, you'd have to ban the words "black" and "white" from the language completely, except in a racial context.

> There was an online test floating around that measured your response rate to a white face or a black face and those results were eye-opening.

This investigation was about seeing faces, not hearing words though.

Interestingly enough, there was one paper that concluded that some of the responses to the test was driven by the prior associations between the colours themselves, rather than the racial part.

Why would you do that? I can wear a black shirt without "badness" being implied.

Not everything is about skin color.

It depends. Am I trying to sleep or stay awake?

Also, what do blind auditions/interviews (a real change that would make sense in the Software industry as well) have to do with language policing?

You were thinking of Good Friday.

Still, apparently the origin of the name seems to have more to do with the idea of a "black day" (a day when a disaster occurs), according to Wikipedia:

> The earliest evidence of the phrase Black Friday originated in Philadelphia, where it was used by police to describe the heavy pedestrian and vehicular traffic that would occur on the day after Thanksgiving. This usage dates to at least 1961.

Why would you do that? Wearing black does not imply that a person is bad. "Blacklist/whitelist" implies that whatever is in the list is bad/good.

While this association is not in any way "natural" or necessary, and it's not even universal, it is still extraordinarily old - dating all the way to Ancient Egypt and influencing European culture from there to now. And thus, it is extraordinarily hard to remove by playing language games with one word.

I should also mention that I am aware that in Japanese culture (and I believe others in that area, but don't know for sure), the traditional color for mourning is white, not black.

This is a pretty huge jump in reasoning. These things have nothing to do with each other. You might as well be saying "red means stop and green means go, so subconsciously people hate Native Americans". Which is nonsense for many reasons.

Perhaps principally that people can discern racial features from faces without using colors at all.

That's pretty strange reasoning. Unless you mean to imply that "red man" is in common usage for a male Native American?

Let me paraphrase this argument, but correct me if you have a different understanding:

  - 'black' and 'white' have some racial associative strength, x, which is sufficient to cause bias in other contexts.  I.e. x > r for some threshold r.
  - 'red' has racial associative strength kx, for some k < 1.    I think we agree that k < 1, since the association is less strong.  Where we differ is I am (hypothetically) saying kx > r, still, whereas you (seemingly) think kx < r.

In fact, we _haven't_ demonstrated the first point anyway, so it's just compounding an already hand-wavey explanation of how things work. If someone can assume x > r with little evidence, why can't I assume kx > r? You might have priors on the size of k because you think 'red' is less strongly associated with race, but we know nothing about x or r, so it's pretty irrelevant. If you can hand wave the first point, you can hand wave the second. As I did.

I'd rather there was no hand-waving. But if that's the game we're playing...

But unconscious bias testing has been shown not to test anything in a consistent manner, being largely unreproducible, and unconscious bias training has been shown not to impact anything much or even consistently impact test results.

Such things appear to be pseudoscience and bordering on a scam.