I wonder if this is moving away from master/slave terminology similar to e.g. https://www.vice.com/en/article/8x7akv/masterslave-terminolo...
And it wouldn’t be even that improbable, in my opinion at least. In the end, in many languages the word “figuratively” was so overused as a stand in for “literally”, that it became it informal synonym (I know for sure that it is a thing in both English and Polish, but it is probably more widespread). If words of completely opposite meaning can be blurred that way, it wouldn’t be so strange for that effect to emerge as a case of euphemism.
As in, to describe actual human slavery?
Most English-speaking countries don't have actual human slavery any more (or at least, they don't use the words master and slave to describe it.) So when using the words master and slave to refer to actual slavery, it's usually in a historical sense, and it seems unlikely that we'd change the words we use to refer to something in the past.
Sorry, but I do not understand the question. Do you mean as to distinguish slavery relationship in logical systems (as with databases) versus human relations? If so, that wasn’t my intention.
What I was talking about was more about a euphemism replacing an actual word, as with taboo avoidance:
> This terminology for the animal originated as a taboo avoidance term: proto-Germanic tribes replaced their original word for bear — arkto — with this euphemistic expression out of fear that speaking the animal's true name might cause it to appear. According to author Ralph Keyes, this is the oldest known euphemism. [0]
Here, there is nothing whatsoever wrong with the _thing_ being referred to (a password, in this case.) The issue is that the commonly used word _also_ has some other associations. So picking a different word means you lose those associations. Not the same thing at all.
It does make one wonder, if there was more diversity in CS when these things were being made whether "master-slave" analogies would have made it in or whether other analogies like "parent-child", "major-minor", "leader-follower" or similar would have made it in.
[0] https://about.gitlab.com/blog/2021/03/10/new-git-default-bra...
In the end it doesn't really matter once the word has gotten tainted. It's still ugly that incorrect deductions spread so quickly and corrections don't.