reminds me of the debate in programming about the central branch yor all changes in git being called master. people wanted the terminology changed because they thought it was about slavery, like master/slave hardware devices. but it’s not; it’s your known-good version, like a mastered mix of a music track. that word has a different root and was never connected to slavery.
…but if you have to explain that every time, then changing it may be for the better. not because the people using it are bad, but because it makes new folks less likely to join. so the new default is main.
My issue with the ‘master’ debate is that the overwhelming number of people I saw fighting against its usage had only “added inclusive language to readme/CoC” as their GitHub history. At that point you’re not a programmer, you’re an idealogue injecting your preconceptions into a world that you don’t understand. Frankly we need less intervention and presumption from this single viewpoint, the western white imperialist, and more people asking questions as their default mode of interacting with things they don’t understand.
Tangentially related is whitelist/blacklist. Nowadays its more common to hear allowlist/denylist or blocklist. Though I’m moderately sure the dated term is coded in something racist, unlike your example.
Colour based terms are super cultural too from what I’ve been told, stuff like red being bad and green being good isn’t universal so imo it’s not a bad idea to use more explicit terminology.
Beyond that, if you go into reporting and the like, red/green colour coding for indicators isn’t accessible (colour blindness isn’t uncommon, last job I had a few colleagues with red/green and one with blue/yellow, I was told that making them very distinct shades helps a lot), people also print stuff out on monochrome printers (there’s old data viz wisdom that suggested designing for this) so I prefered symbols when I did more of that work, still suggest it when I get asked to review things.
How many people had to be explained what a master copy is? Do these same people need to be explained what a master degree is?
As for the ‘master and slave’ terminology, yeah it should go away, if anything because it was rarely even accurate. For example, the IDE interface refered the primary drive as master and the secondary as the slave. But the primary drive has absolutely no control of over the secondary. It just defines who gets the bus id of 0 and 1…
The change still annoys me at times as the default to git is still master. Its the hosted git (Github, Gitlab, etc) repos that default to main. And on occasion I get reminded this when I push a local initialized repo to them…
it happens enough that it became worth it to go with it. people new to any field will come into it with only the “common” understanding of words, and it’s such a small thing.
reminds me of the debate in programming about the central branch yor all changes in git being called
master. people wanted the terminology changed because they thought it was about slavery, like master/slave hardware devices. but it’s not; it’s your known-good version, like a mastered mix of a music track. that word has a different root and was never connected to slavery.…but if you have to explain that every time, then changing it may be for the better. not because the people using it are bad, but because it makes new folks less likely to join. so the new default is
main.My issue with the ‘master’ debate is that the overwhelming number of people I saw fighting against its usage had only “added inclusive language to readme/CoC” as their GitHub history. At that point you’re not a programmer, you’re an idealogue injecting your preconceptions into a world that you don’t understand. Frankly we need less intervention and presumption from this single viewpoint, the western white imperialist, and more people asking questions as their default mode of interacting with things they don’t understand.
yeah that was an issue.
Tangentially related is whitelist/blacklist. Nowadays its more common to hear allowlist/denylist or blocklist. Though I’m moderately sure the dated term is coded in something racist, unlike your example.
Colour based terms are super cultural too from what I’ve been told, stuff like red being bad and green being good isn’t universal so imo it’s not a bad idea to use more explicit terminology.
Beyond that, if you go into reporting and the like, red/green colour coding for indicators isn’t accessible (colour blindness isn’t uncommon, last job I had a few colleagues with red/green and one with blue/yellow, I was told that making them very distinct shades helps a lot), people also print stuff out on monochrome printers (there’s old data viz wisdom that suggested designing for this) so I prefered symbols when I did more of that work, still suggest it when I get asked to review things.
How many people had to be explained what a master copy is? Do these same people need to be explained what a master degree is?
As for the ‘master and slave’ terminology, yeah it should go away, if anything because it was rarely even accurate. For example, the IDE interface refered the primary drive as master and the secondary as the slave. But the primary drive has absolutely no control of over the secondary. It just defines who gets the bus id of 0 and 1…
At some point, if enough people are annoyed and the alternative is just as valid, the annoyance of getting used to the new term is worthwhile.
I was annoyed by the change like 5ish years ago, when the movement started, but now idc anymore.
The change still annoys me at times as the default to git is still master. Its the hosted git (Github, Gitlab, etc) repos that default to main. And on occasion I get reminded this when I push a local initialized repo to them…
That’s configurable since Git 2.28. You can change
init.defaultBranchto main or trunk or whatever you want.it happens enough that it became worth it to go with it. people new to any field will come into it with only the “common” understanding of words, and it’s such a small thing.