Git 3.0 will use main as the default branch | Hacker News Reader

Git 3.0 will use main as the default branch

thoughtbot.com

96 pointsingve258 comments

Honest to God not something I care about but: this is pretty much the nail in the coffin for “master”. I do know some people _did_ care about the name. Sometimes surprisingly senior people who never supported a tech upgrade want the name changed. In any event, it’s done, “main” won, it’s fine, let’s move on.

Maybe they resisted because it was completely ridiculous waste of engineering resources all over the country and for absolutely no tangible reason other than white people trying to feel better about themselves.

I work in the field of film mastering (with countless product names with the word “master” in it) and luckily no one got the ridiculous idea in their head that we need to change this lingo.

Show me a single person who has a valid reason for me not calling my branch “master” or my bedroom “the master”. I honestly think this sort of ridiculing word policing is why we lost this last damned election. And if you’re somehow proud that you’ve renamed your git branches, you’re very likely a contributor to that lost election.

Makes sense when you release 3.0 and basically allowed to introduce breaking changes.

In tech field there's lots of people living on the very fringes of society, hidden away behind keyboard.

I actually worked in film audio engineering and Master is not the universally used term and hasn't been used uniformly throughout history. I have an analog Mackie mixer from the 2000s with "Main" as the name of the Main Bus that was designed before the whole debate took part.

As far as software goes, things are similar. The process of "Mastering" is an exception.

As far as git branches go, I am fine with main. It has two advantages over master aside from any culturual questions:

1. main is more self-explanatory for beginners who don't know how "master" was/is used in tech.

2. it is shorter. While two letters don't make a huge difference, that is still a subtile advantage.

Whether these two points alone are enough to justify the needed work (which is probably not a lot to be honest), IDK.

That "waste of resources" is completely made up, this changes nothing for any existing repo what so ever. Any existing repo that updated did so completely voluntarily, no tool forced them to.

At most you could argue that you needed to run one additional command when pushing the initial commit during this transitional period where GitLab/GitHub had updated the name but Git itself has not. Therefore, now we're back to square one with less "waste" as you put it.

That "waste" of resources was absolutely tiny. Took me just some minutes. And I didn't do it because of DEI, just because I think it's a better name.

Same for me, but kind of because of DEI. Basically, it offended some people, and even if I thought it was a little overblown, it took about 2 minutes to change the default name of future repos to be something else (which was at least as good, and perhaps better). It made some people happier at approximately zero cost to myself, so why not.

Yeah that’s how I feel about most progressive stuff - sure it might not bother me, but also changing doesn’t bother me either. It costs you so little to accommodate other people.

I used to feel that way up until about a year ago. At worst I would roll my eyes at the silliness and then move on, because this stuff rarely matters much one way or another.

But then the 2024 elections happened, along with a bunch of exit polls, voter interviews, and other data showing that a surprising (to me anyway) number of people hate this kind of virtue signalling to the point that it can sway their vote. It's very possible those swung votes have ushered in a host of harmful changes that I think do matter a great deal. So now I'm sick of this stuff, it's not only a waste of time it's actively harmful.

I’d like to see a study showing 1) people aware of this issue and 2) for whom it swung their vote to the right. That’d have to be, what, 10 complete idiots? “Well, I was going to vote for A, but some of B’s supporters asked if I would please be considerate, and that’s a bridge too far.”

I don’t think it’s true that the culture war issues themselves were the cause of those swayed votes so much as there’s a propaganda machine running 24/7 stoking those resentments and using such cultural critique as fodder.

This works really well to whip people into an othering frenzy to distract them from voting for their own economic interests.

Oh, sure, it was people asking for branch renaming.

It wasn't the multiple dedicated Neo-Nazi propaganda networks that call all minorities and immigrants and political opponents enemies of the state.

It wasn't the election of politicians that are actual, convicted rapists and felons who distract by pointing to those who can't fight back.

Sorry it was exit polls that convinced you not to care about other people so much?

Don’t fool yourself kiddo, you were always an asshole, you were just waiting for the right excuse, just like the rest of us.

The deal with progressive ideology is that it progresses. Fixing inequality, prejudice, and injustice are a lifelong project, because as fast as you address issues, bigots will create new things to be bullies about. You don’t get to just get off at some point and be like “oh okay things seem good enough now.”

Progressive ideology tends to treat moral progress as inevitable, while pursuing social transformation in ways that can undermine the very institutions and norms that make progress possible.

Making "some people" happier isn't zero cost if the people in question are intolerant lunatics with ideas corrosive to the social fabric. It's one reason why the pendulum is swinging fiercely in the other direction.

TIL "I'm uncomfortable calling it master-slave, can we do main-replica?" is the idea of an "intolerant lunatic" that is "corrosive to the social fabric".

Good Lord, just listen to yourself.

Red-lined districts still shape America to this day and several red states have been rampant on racial districting to screw minority communities. You can't even pretend the history of slavery is in the past in America.

This is why I left California and utterly done with you people.

I live over 1,000 miles from California, bud.

I have encountered at least two bugs due to the change in names.

Everything considered I invested an hour or more in total. I am pretty sure decades of engineering time and resources were invested over the years because some people didn't like a default globally used for decades.

Every time I push to master I get this song in my head

  Master! Master!

Every time I push to main I have in my head:

  ...meh

Master of Puppets -- Metallica

To be fair, the song is about control and the abuse of power.

I heard that it's about drugs controlling you

This stuff reminds me of what my mother said about feminists trying to get people to spell women with a y. She didn't like it because it made feminism seem like something petty and frivolous.

If I put my tin foil hat on it feels like a psyops to make the left look like a bunch of morons.

Yeah, I mean, isn’t a masters degree even worse, then?

a masters degree is about mastery, not about being central/main/leading...

Masters degree is from “magister”, meaning director/chief/boss/leader

wikipedia says: The original meaning of the master's degree was thus that someone who had been admitted to the rank (degree) of master (i.e. teacher)

and

from Latin: magister, "teacher"

teaching requires mastery, not leadership, although the concepts are related because mastery is also good for leadership.

my armchair etymology suggests that master and mastery were closely related until it started to be applied to leadership as well.

Magister means teacher and the things I mentioned. See https://en.wiktionary.org/wiki/magister#Latin

Now I just know why I couldn't get a job. It's because I got a bad and worse degree for two years

In Microsoft v. AT&T, decision 550 US 437 (2007), there was discussion about a golden disk, and the terminology changed to master disk during the course of the proceedings, because the disk wasn’t actually made of gold.

I remember that Justice Antonin Scalia objected: “I hope we can continue calling it the golden disk. It has a certain Scheherazade quality that really adds a lot of interest to this case.”

<https://www.supremecourt.gov/oral_arguments/argument_transcr...>

what, the golden ratio is not made of gold? have we been betrayed all this time?

Wait until you hear about showers

you mean, like baby showers? yeah, totally not what i expected...

The golden rule would deform pretty quickly be useless for measurement.

you can call your branch whatever name you want. nobody cares, nobody is stopping you.

That was true before the 3.0 release. Why didn't the people offended by "master" just change the branch name? Because it was never about their own branch names. It was about everyone else's.

> Maybe they resisted because it was completely ridiculous waste of engineering resources all over the country and for absolutely no tangible reason other than white people trying to feel better about themselves.

I think the resisting probably wasted more time than anything else.

We used the occasion to ensure that there was no hardcoded naming in our IaC, internal tooling and CI/CD. It was surprinsingly easy, gave us a great excuse to do some much needed clean up and now everything can work with any branch used as the main one.

Was it extremely important? Probably not. Was it worth fighting against/having a stong opinion about? Probably not either.

Sometimes, it's easier to just go with the flow and try to turn things which seem meaningless into actual improvements. If it makes the people who think it's not meaningless feel better, well, even better. It surely didn't cost me much.

I dont know, for me main makes more sense so I prefer main. The main branch isn't master of anything, but its the main branch.

> The main branch isn't master of anything

It's master as in "master copy":

A "master copy" is an original version of a work from which other copies are made, serving as the definitive or controlling version

Yah, they are losing something with the name change that they don't even understand because they apparently don't understand the intricacies of English. We would be better off changing it to "Gucci Mane" then we could tag our branches off Gucci's hit singles.It only makes slightly less sense than switching to "main"

That would make perfect sense if all branches had to be made from the default branch.

But they don't.

At `$CLIENT` we use `stable` as the default branch.

Use whatever works for you. Getting upset about a default that you can change is like getting upset about the default wallpaper of your OS.

And before you get all persnickety about that argument working both ways: the developers of git, get to decide the defaults and they did.

If you're so upset, fork it, revert the default branch name and maintain it yourself infinitely. That's definitely worth it just to keep a default branch name you like, right?

> If you're so upset, fork it, revert the default branch name and maintain it yourself infinitely. That's definitely worth it just to keep a default branch name you like, right?

No idea how you got that impression from my comment. It sounds like you're the one that's upset.

I don't care what you name your branches. I do think it's dumb to tell other people what (not) to name their branch though. But definitely not something I feel compelled to rearrange my life over.

Nobody is telling you what not to name your branches.

The people that wrote the software you're using for free, decided to change the default name.

That's it. Nobody has said you can't use whatever name you want.

> Nobody is telling you what not to name your branches.

> Nobody has said you can't use whatever name you want.

This is reductionist. The git people didn't pull this idea out of their butt. It came about because a lot of people were saying that we should not name our branches master.

I have no problem with what the git people did. Easy enough for them to change it, and it puts a dumb issue to bed (for them).

But I think it's fair for anyone to point out that the motivation was dumb, and to explain why it's dumb and how the word "master" is actually not an unreasonable choice in this context.

> Nobody has said you can't use whatever name you want.

Sure, until somebody makes the mistake of not renaming all of their old "master" branches and gets shamed by the word police over it.

Of course you're welcome to disagree.

> how word "master" is actually not an unreasonable choice in this context.

It doesn't even make sense in this context though. The name just got copied from BitKeeper which had master and slave branches.

Git doesn't have that concept.

> Sure, until somebody makes the mistake of not renaming all of their old "master" branches and gets shamed by the word police over it.

How are you going to be shamed? I thought there's nothing wrong with it?

> How are you going to be shamed? I thought there's nothing wrong with it?

If you re-read my comments you will understand that I don't believe there's anything wrong with using the word "master" to name a branch. But other people do, which is why there was an uproar and the default name was ultimately changed to "main".

Not sure how you were able to misinterpret this.

So what's your point?

If you don't think there's anything wrong with it, why would you care if someone else says "hey change this".

Do the same thing you'd do if someone says "hey you should use mongodb it's web scale": tell them you disagree and won't be doing that.

If you don't think there's anything wrong with it, how can you be "shamed" into doing something you disagree with?

I am doing exactly as you say.

I made a comment saying I disagree with the word police and I think it's dumb to cast people as being insensitive for using a longstanding word that makes sense to many people in the context it's used in.

But git is a distributed VCS system - there is no master copy. But there certainly is a main branch, which may have many copies on many machines.

It's one of those things I couldn't care less about what it's actually called as long as it's uniform everywhere.

I had a few frustrated evenings of debugging when Github changed the default to main and my local scripts expected "master".

All fixed now, but still an annoyance. Don't think about it much anymore.

Some of your scripts could possibly use `origin/HEAD` and reflect whatever origin thinks is the default branch. (Though obviously that assumes you always have an `origin` remote or something remote-like.) Including using the commit referenced by `origin/HEAD` to find the `origin/{branch-name}` that matches if you want a name to check locally.

PSA: You can run one git command and ignore this change and associated drama entirely. I don't care which you prefer, but let's not pretend like main "won" when sticking with master is as easy as:

    git config --global init.defaultBranch master

main won

To be fair, ignoring the drama is just adapting to changes, which is crucial in this field. Our old repo defaulted to master, our new ones defaulted to main. No time was spent on bike shedding.

This has been the case on nearly every open source and proprietary project, I have worked on.

Most people do not care and will stick with the old default on old projects and use the new default on new projects. Occasionally, it stokes conversations around possible third options that are more descriptive like stable or development, but the norm is to just go with the default.

Going out of your way to set the default to the old name really reeks to me of slacktivism. People probably think that they're taking a stand, but in actual fact others will just assume that your repository is older or that you have an old configuration.

git config --global init.defaultBranch default

Because Mercurial is using default as default, and if we have to resort to worse SCMs, let them behave like Mercurial anyway

I did this the first time git asked me, because getting asked every time would get annoying.

Glad it's going back to having a default and not asking.

Not at all. This is just about defaults. People can still choose arbitrary branch names. People can still set the default branch name, as I have:

  [init]
      defaultBranch = master

I just think "master" is an awesome word. Master record. Mastering. It just sounds cool to me and I'm gonna keep using it.

I also think "main" is a stupid word that doesn't say much about anything. I even hate "main" functions.

Tip: the next time you need to name a function, don't use "a stupid word that doesn't say much about anything". That's not how you're supposed to name stuff in programming. :)

In that case, I think 'git' is probably the name we should have been looking to replace...

It's hard when compilers force you to use "main".

  int main(int argc, char **argv);

I once wrote a liblinux library for Linux software development with freestading C. One of the things I did was replace the "main" function with a "liblinux_start" function.

Ada compiler does not force to use main

That's just the default. Nobody stops you from specifying `gcc -Wl,--entry,foo`.

Standard C stops you. The C standard library is hardcoded to call the main function. Providing one's own ELF entry point also breaks libc initialization unless the exact same startfiles are used.

Freestanding C gets rid of the libc so that's not a problem.

Maybe you could also do some runtime linking trickery when dynamically linking libc?

The libc could easily add support for such a thing but I doubt they'd care enough to. They'd probably see it as adding complexity for zero gain.

I submitted an issue to the GCC bugzilla about renaming certain symbols generated by compiler.

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113694

I don't know how to implement it and nobody else cared enough to.

I prefer to default to `develop` and then eventually branch out to `release`: that way my branch names are pretty explicit. It seemed silly to me to start with a "central" branch, no matter the wording, because that's not actually how Git works (and it's rather uninformative).

For... some in the comment section, please recall the HN guideline: "Comments should get more thoughtful and substantive, not less, as a topic gets more divisive."

I remember when default branch was simply `trunk`, but maybe that was in SVN not git… trunk goes well with branches, maybe git should stick (heh) to it

Which for some of us, Spanish speakers, was on occasion amusing or lewd, depending on the context and culture. The Spanish equivalent is 'tronco' which is very similar and it is slang for a couple of things.

It's not silly at all, there are multiple ways how you can version your changes in git.

I am also leaning towards this, especially for active repository where a magic or master branch does not mean anything

s/repository/repositories;s/magic/main/

I sincerely hope you consider choosing "magic".

So, 'master' was the default branch name. Why is the default changed in 3.0? Is it because of the allusion to slavery in the United States? Even if so, what does a Git branch name has to do with that country's history? Did Git used its branch name to enslave people? When a user does 'git checkout master', are more people getting enslaved - some kind of butterfly effect?

Does it need to be said that if the US had or has a problem, it's they alone who need to deal with it? Since when did the rest of world took upon itself to "solve" their problem? Did we all get green card or something?

I don't feel like this will stop here. What's the next word some people in some other country decided to declare offensive?

Shame.

> Did Git used its branch name to enslave people?

Git took the branch name from Bitkeeper which did have "slave" branches and used a "master/slave" analogy. Git didn't also inherit the "slave branch" concept in that same way, but it did have that heritage accidentally imported when git lazily reused that branch name.

> Does it need to be said that if the US had or has a problem, it's they alone who need to deal with it?

Slavery/indentured servitude is a worldwide problem that still exists today in countries that are not the US. Even if you think this is an over-correction in relationship to the historic US Slave Trade specifically, that was a multinational effort involving the British Empire, the Dutch Empire, and many other Former Colonies beyond just the US' involvement. The US took advantage of the trade, but it neither invented the trade nor was the lone slave owning country involved in that trade, nor was it the last country in that trading group to end slavery trading in practice even if it was one of the last ones on paper.

I'm not trying to argue against it, I think "slave" branches make no sense anyway, but to GP's point BitKeeper didn't enslave anybody, just used the word.

If we believe we should remove allusions to negative things why are we ok with "kill", "orphan", "evict", "bash", "cut", "isolate" etc? What is special about that terrible concept that we should stop using the word even when not applied to people at all?

The point of bringing up Bitkeeper is as much because why use a word divorced from its original meaning at all? "master" wasn't an explicit choice by a git maintainer, it was inherited noise. When confronted with where that choice came from, in finding it wasn't a choice but a bad legacy, the git maintainers generally agreed it might be nice to pick something that made more sense as a choice (rather than bullshit noise from a practically dead and gone upstream project) and after much debate "main" made sense as something a lot of people were using anyway.

That's what is "special" about it, that it wasn't special. It wasn't chosen. It was just a stupid inherited default that didn't make sense when questioned.

It was never an intentional allusion to a negative thing, it was accidentally a negative thing causing real people some harm, and it was easier to fix than to justify why it was a negative thing in the first place.

master alone feels somewhat (I disagree personally) non-scary, but consider that master/slave has been used elsewhere in computing: i2c, ide protocols for example. It's just a bad look. It's of no consequences to change, this is easy, simple. Have joy that we improve.

When is Linux going to rename “man” which according to my employer is a non-inclusive word and flagged in pull requests

Yeah, or killing parents and children. sighs. People really need some context-awareness.

What...?

I think they mean the kill/pkill. Not defending the silly chain of what-about arguments though.

Is it silly though? With enough linguistic archeology I bet you can make this entire comment I'm writing right now extremely problematic and offensive. The linguistic treadmill means exactly that older terms change meaning back in time. They also change meaning FORWARD in time, meaning your inoffensive terms today will almost certainly be offensive in the future.

It's also the case that offense is language dependent, which is always funny when Americans hard ban certain words on chats and then Swedes can't use the Swedish word for "end" because it's spelled like a slur in English.

Everyone needs to stop this nonsense.

> Swedes can't use the Swedish word for "end" because it's spelled like a slur

"Ände" is a slur? (excuse my lack of transductional skills)

They probably mean "slut". The word has the same meaning in Danish, by the way

The whole terminology in IT could be turned upside down because it can be quite offensive if people ignore the context, so it is not limited to processes. There are utilities like "man", "finger", etc. that could come across as offensive too, to some, with no context-awareness.

Today it is "master" -> "main", tomorrow the whole IT terminology.

There are many PRs on GitHub with regarding to these, by the way.

... also what about pins? Slave and master pins! Must be about slavery, right? No, it is not, not at all.

In any case, who made the association of the git branch "master" to slavery? It is absurd. People need to take the context into account.

IMO for something to be offensive it has to have intention to be offensive. Otherwise it's misunderstanding.

Of course, but they do not care about that. They made the association, and now they are being vocal about it. I am pretty sure most of us never made this association or attribution. I have never thought about slavery until they told me their own associations to it.

I am pretty sure master / slave pins were not intended to be offensive, nor attributed to slavery. Similarly with the git "master" branch.

And quite frankly, the problem is that we cater to such people instead of teaching them to be context-aware.

> In any case, who made the association of the git branch "master" to slavery? It is absurd.

BitKeeper, the VCS that preceded Git, used the terminology "master" and "slaves", so the association is not based on nothing:

https://github.com/bitkeeper-scm/bitkeeper/blob/0524ffb3f6f1...

It is based on nothing. It is not intended to be offensive, and it is not intended to be about slavery. Similarly how master and slave pins are not either, or how blacklist and whitelist are not about race either!

BTW FreeBSD changed blacklistd to blocklistd. :(

I do not mind blocklistd, but then again, there was nothing wrong with blacklistd either.

I grant it's not nothing, but I think it's not enough of something to make changes over it. Thinking of a master record or similar is the natural reaction when you learn about the terminology, and most young people have never used bitkeeper, so unless you go out of your way to explain why this is "bad" most people won't even know, so what do you gain from it?

Don't forget killing orphans and zombies.

I will not forget master and slave pins either. :D

Has the Linux community thought through the implications of using words like "containerization", when millions of slaves are still shipped around in shipping containers [1]? Docker is problematic technology to say the least.

[1] https://www.aljazeera.com/news/2018/1/26/slavery-in-libya-li...

Yes, we ought to change "containerization" everywhere! In the Linux kernel, user-space, Docker, everywhere, only because people have difficulties with context-awareness! /s :D

Please IT people, tell me that this thing is absurd.

FreeBSD also ought to consider the trauma "jails" as an analogy can cause to those unfairly incarcerated by the U.S. justice system.

Now we are getting somewhere! :D

We better get rid of database constraints at the same time, since people were constrained on those same ships. I don't see how we can move forward as an industry until these grievous harms have been addressed.

killAllChildrenOf() is still my favorite method name

Side note, I had a professor who used to talk about "forking children" (processes) and I mean... the way she said it, with her accent... there are better words we could be using here to mean the same thing that wouldn't cause such absurdities to be uttered.

Nah, I think the terminology is fine. Everyone in IT understands it. The people outside of IT have no business in our terminology.

It costs nothing and makes people happy so why be a jerk about it?

Probably, because not everyone is made happy: some are annoyed. I am not going to enter the merit of either side.

you cant make people happy. Humans want always more. it's never enough

It’s non inclusive of people who think it’s stupid - their feelings are not taken under account.

If there was a case where that was more than a quantum of annoyance, where it actually was any kind of actual weight at all, maybe those incredibly small feelings might be interesting to weigh. But they're not. They're the smallest of feelings, weighing nothing. Making this change is the easiest lightest most obviously acceptable of wins.

This is such a weirdly patronizing comment. What exactly makes the feelings of the people who cried about the original name "bigger" than those of the people who don't like the change?

Master can conjure a horrible past. That's a real actual hazard.

Avoiding it should be easy. It should be the easiest of easy things. Make the change and move on. Hack your reality, improve things, don't make more of a glaring ugly mess of things than need be.

It costs so so little to do a small good act. You wouldn't name your primary branch holocaust or tyrant or oppressor or doommaker or worldruiner or hates[ethnicity]. But computing used master/slave extensively for decades (ex: I2C, IDE protocols). Folks arguing that it hurts their feelings to not have those names wouldn't be given any weight. These are alligator tears.

No one has made a case that "main" is a bad name. Because this isn't a real case. It's perfectly clear, no lesser a name in any way.

More broadly, when assessing positive liberty (freedom to do something) versus negative liberty (freedom from being dogged), we shouldn't favor positive liberty to make broadly harmful defaults that can hurt people and/or bring misery over negative liberty not have a world a rare couple antagonists insist on driving down in stature.

This feels to me like people either dragging their feet & bringing resistance, or, on many many cases, people actively obstructing, making an easy simple improvement much much more fraught & hard. And many of those people I feel like do it because they know it is a good change, and they actively seek to keep the world worse. I have little and descendingly less pity for any. If someone wants to build a case for why their feelings here that we shouldn't do this have weight, I'm all ears. And I'll spend some time to read more comments to see what I see. But the person I replied to made zero case for why they felt their emotional injury (as they begged for tolerance of intolerance), yet felt that their case should carry as much or more weight, where-as master/slave usage in computing are words which associates with slavery. Which is the ultimate positive liberty vs negative liberty case, which reflects the matter here: the negative liberty to not be enslaved outweighs any positive liberty to enslave.

You're free to think it's stupid!! That fact though just doesn't matter very much in this case. It doesn't actually really affect you. You can adapt, with barely more than zero cost. It's stupid maybe perhaps possibly!! But of no cost. And that's the weight of your feelings here: it's not actually of any consequence to you, you are claiming stake in a matter without any basis. You're free to feel however you like in this world, but whether those emotions actually match/reflect the circumstances that spring them matters. Generally I think most people kind of agree that it sucks that master/slave (ex: i2c, IDE protocols) nomenclature was chosen & used in computing, and calling the shift away stupid-in my view- should be taken as the smallest imaginable quantum of protest, the smallest tears imaginable. Stop stopping the world, let time progress forward, don't trap us in your negative energy forever.

Really? I pointed out how patronizing your comment was and your response was to write a massive wall of text that is somehow even more patronizing?

Naming a branch in my git repository "master" does not "hurt people". It does not "bring misery" unless you were already miserable and looking for things to justify it. This change is of consequence to me. It does affect me. "Barely more than zero cost" is more than zero cost. I don't care if you think my opinion "doesn't matter" or is "weightless", because yours doesn't matter any more than mine. You are not a higher being of superior importance because you chose to dedicate your life to fighting a never-ending war against the unreleting evil force that is a word on a screen. Your tears are the biggest, and most pathetic of all.

Feels so very "all attack, no meat" to me. I don't see anything here working at deserving respect, nothing genuine, no try to win over to your opinion: it is focused on disregard & disconvincing, casting aside, with no actual thought or explanation or reasoning provided. You say this has consequences? Has meaning to you? Yet you don't elaborate on what consequences are, on why this great nothing-burger is so offensive.

> because your [opinion] doesn't matter any more than mine.

I wrote a wall of text because I actually care a tiny bit and have put work into explaining and backing my opinion & character. Because I think those opposed have a bar to meet, just as I want to clear a bar of respectability & engage duly on a topic that should be easy to improve on. What I see is you being quite aggressive & scathing while providing very scant little to argue your opinions.

You say our opinions are equally valid, but to me, opinions are weighed on argument, on what is said, and that's why I've tried to put my logic and character down here, to be honest & forthwith about what I'm saying: because that's the basis we have to use to assess opinions. This seems pretty agreeable to me, and again you have option to disagree, but I can't imagine doing so.

This is absolutely a question of weights. Making mountains out of mole-hills is to be avoided. The weight of this question is hardly worth considering at all, and it's absurd what tears you spill over doing such a simple improvement for the world. I'll also allow that perhaps the certainty & approach I've entered with has entrenched & only enhanced enmity to my cause. But with context, having seen master/slave used around in computing, and given how easy it is to shift & adapt to better, I continue to find opinions to the contrary to be unduly self-important, for no real reason. Why stand against such clear wins? My character may not be perfect for all, but let that not alone dissuade us.

> No one has made a case that "main" is a bad name. Because this isn't a real case […] they know it is a good change, and they actively seek to keep the world worse

It's a bad name and a bad change. I won't be using it. Not because "master" is good (I also won't be using "master"), but because "main" is bad—even worse than if they'd started referring to it in the documentation as the "default integration branch" and so named it "int".

I'd sooner support changing "master" to "margarita" than changing it to "main".

What is the issue with calling a branch `main`? It's the standard that's been adopted in my company and we haven't run into any issues; I'm wondering if there's anything we've missed.

(I've also seen `dev` used, short for 'development branch,' which seems pretty reasonable).

I think those people are taken into account, and on balance are ignored because their argument is far less reasonable. "Slavery was one of the most awful things humans have ever done so we shouldn't continue to use the language of slavery to describe every day things because that diminishes the importance of our history" makes a lot more sense than "we should keep this terminology because it's tradition."

We fix things that are broken. That's progress.

"Was"? There are millions of people in slavery in the US penal system today. Entire cities like Dubai continue to be built by slaves. I, for one, find it to be extremely insulting towards their intelligence to think that unrelated use of the word "master" in a different context is somehow considered to be offensive.

Except the master branch has nothing to do with slavery. Master branch doesn't even rule over other branches. It's more like 'master bedroom.'

> Except the master branch has nothing to do with slavery.

It did, it originated from Bitkeeper that literally used to the term "slave" to refer to non-master branches.

> It's more like 'master bedroom.'

This is even more ridiculous. Where do you think that term came from? What made that a master bedroom in comparison to the other bedrooms? Could it be because that was the one the master was sleeping in, in comparison to the ones slaves were sleeping in?

There are many more kinds of masters than just owners of slaves. The word "master bedroom" only appeared in 1920, it has absolutely nothing to do with slavery.

No, that's just one of those made-up lies people repeat often enough online to become "true" because it's the top search result and because it makes them feel good about continuing to use that term.

Here it is for example in an Australian newspaper ad from 1844: https://trove.nla.gov.au/newspaper/article/31742822

> TO LET, Westmoeath Cottage and Garden, situated near to Cook's River, only three miles from the city. the cottage contains parlour and drawing room,and four large bed rooms ; detached kitchen, bakehouse, landry, storeroom, four stall stable and double coach-house, servants' rooms neatly fitted up, together with hay-loft and granary, school house and master's bed-room. A cottage containing four separate rooms for overseer and workmen ; two excellent wells of water on the premises, about six acres of garden neatly laid out and planted with the best vines and fruit trees, 'This property is fit for a family of the first respectability.

Or here it is in London-based The Examiner from 1845: https://upload.wikimedia.org/wikipedia/commons/0/08/The_Exam... (page 523, middle row, a couple of lines below the "Police" headline)

I'm sure I could find more examples, but I think two will sufface.

You couldn't own slaves in London in 1845, and in any case the name derives from the "Master of the household", so if you want to be mad about it, you should call it sexist, not racist. Or you could just be chill, stretch the meaning a bit and say the couple together are the masters of the household. But, now I'm curious: Where do you draw the line? You don't like git master branches and master bedrooms, but what about other uses? You can have a master key, master record, master a skill, create a masterwork, be a master to an apprentice, join the toastmasters, be a master of ceremonies at a formal event, you can dress up for comic con as Master Yoda, Master Chief, or Dumbledore (the Headmaster of Hogwarts), you can be a Master Chief in the US Navy, be the dungeon master for a game of D&D, get a Masters' Degree and so on. Which of these things are in your opinion bad and should be renamed?

This is very much like asking why are you focused on fixing one bug at a time in your software when you can fix every reported bug simultaneously?

I don't know man, maybe it's because fixing this one completely inconsequential bug faces so much backlash for no particular reason other than "change bad"?

And well done with using an example from a book series where the only Asian character is named Cho Chang and where there are elves with long noses in charge of the "central bank". That really works in your favour, you totally owned me [pun intended] with that one!

Good job finding something to complain about in one of the 13 examples I listed. This unassailable refutation utterly destroys my whole argument :(

> A footman in his lordship's service stated he went into his master's bedroom [...]

isn't an example of the phrase "master bedroom".

I am also skeptical of "school house and master's bedroom". The main cottage has "four large bed rooms". Why would the "master bedroom", if it is meant to be read as it is today, be listed after the list of detached outbuildings?

Someone else linked this historical document to show that the words in Git has indeed to do with a master-slave relationship: https://github.com/bitkeeper-scm/bitkeeper/blob/master/doc/H...

But as I said elsewhere, I do not care about that a lot. However, I do think those words are bad for other reasons because they do not illustrate the _actual_ role and relationships of branches in a VCS in a _good way_. If the master branch is not actually ruling over other branches, then it should be named something else, like "primary" as far as I am concerned.

Language is in constant flux. If a word has a remote historical connection to master/slave in a precursor project people nowadays don't even know, and people invent a "folk etymology" comparing it to a master record, do you really gain anything worthwhile from insisting on the history?

PS: about the role in git: I don't feel that strongly about it, but I think master is somewhat more descriptive. The master branch contains the most up-to-date version of the source code, so e.g. if I'm working on a feature branch and a colleague pushes a bug fix that affects me into master, I need to merge/rebase to get the latest changes into my feature branch. So, while the master branch doesn't "rule" the feature branches, there is still the implication that changes to master should find there way into the feature branches at some point, which I think main doesn't convey that clearly.

I'm with you on the cost basis for this. It makes no sense to die on this hill.

But to answer your question, it's unclear how "inclusive" this actually is. No person of color I've talked to about this (at least 3 that I can recall) has thought this was anything other than weird virtue signaling by admittedly mostly white social justice warriors. Their feelings on "master" terminology ranged from mild bemusement to "I REALLY DON'T GIVE A SHIT."

If the only people being "included" by this are people whom are themselves being uninclusive, even if for the right reasons... I question how "inclusive" this actually is.

That said, I'm definitely in the idgaf camp. My new repos are "main" because that's the default. Most of my old repos are still "master" because I never bothered to change it.

It is simply the soft bigotry of low expectations rearing its ugly head…again.

“Well OBVIOUSLY”, says the white knight, “some people are too stupid to understand the context and nuance of a langue like mine English. We just HAVE to adjust it because they won’t understand.”

It says much more about the people pushing such things than it does about the perceived slight.

This is closely related to the Heckler's Veto.

Correction: it costs something (breaking with past convention which adds friction), and it makes some people unhappy and others unhappy. This is not a harmless, all-upside change the way you are making it out to be.

Because that's newspeak.

Because it rewards stupidity.

> costs nothing

The cost is measured primarily in time. Experts in git/GitHub just experience a little annoyance (per repo). But for new comers (esp self-learners) the cost is much higher (takes the form of dysfunctional instructions, tutorials, readmes), and at the margin could cause someone to give up on a tutorial.

Cost should be weighed up against benefits i.e. that those hurt or upset by use of the word 'master' would no longer be. It's highly questionable whether the term master ever upset anyone (not simply those who were upset by the idea that someone else could be upset by the term's use).

A second-order cost is the precedent. There isn't a word in the English language which cannot be interpreted as malevolent given enough effort from the interpreter. Therefore it can be a better strategy to accept that there exist words with multiple meanings depending on context, and live with this language feature/imperfection, rather than impose costly changes on everyone to benefit a (possibly non-existent) few.

Because it goes hand-in-hand with the euphemism treadmill. I defer to the late great George Carlin's words on the subject:

  I don't like euphemisms, or euphemistic language. And American English is loaded with euphemisms. Cause Americans have a lot of trouble dealing with reality. Americans have trouble facing the truth, so they invent the kind of a soft language to protest themselves from it, and it gets worse with every generation.

Video https://www.youtube.com/watch?v=vuEQixrBKCc or text form: https://www.lingq.com/en/learn-english-online/courses/87644/...

Something I wrote back when GitHub did this:

> This change does hurt people. Have you never seen a newbie struggle with an out of date tutorial? There’s fifteen years of books, tutorials, videos, and other learning material out there that assumes the use of `master`. It takes about a decade for these things to die off in the clear-cut cases. However this is not even a clear-cut case, because Git defaults to `master` and GitHub defaults to `main`, so the confusion will last longer.

> Version control is a hard enough concept to grasp well at the best of times. And now there’s going to be a bunch of newbies – many of whom will be black – getting frustrated and confused because they are following the tutorials as best as they can and not getting the results they expect. On balance, there is probably more harm done to black developers with this change than leaving things as they were. Haven’t you ever seen a young developer lose confidence when they are trying the best they can to follow instructions and it’s just not working? Disliking the fact that this is being done to new developers thoughtlessly is not “pointless”.

> There’s also the matter that everybody I have seen advocating for this change has been white, and the responses I have seen from black developers generally ranges from “this is pointless” to “this is [performative / virtue signalling]” (insert appropriate term depending upon whether they are left or right wing). People start with the assumption that this is clearly the right thing to do for black developers, but people assuming that don’t seem to be actually listening to black developers about this, or at the very least, only listen to the ones that agree with them.

> This could have been done in a better way – coördinate with the main Git project around changing the default across the board, plan an update to as much documentation as possible, and make it happen in sync. But GitHub charged in unilaterally, seemingly with an overactive case of a white saviour complex, with people like you telling everybody ”this change hurts nobody”. This was done in an entirely thoughtless manner and does hurt people.

— https://www.reddit.com/r/git/comments/jtrx1k/announcement_po...

Beyond what I originally wrote, you also had ridiculousness like `git init` creating a branch called `main` if you install Git through Apple developer tools and `git init` creating a branch called `master` if you install Git through Homebrew; or getting a repo with `main` if you initialise the repo on GitHub and then clone it locally, but getting a repo with `master` if you initialise the repo locally then push to GitHub.

I don’t especially care what the default is, but I do care that GitHub didn’t seem to give a shit about the disruption they caused as long as they looked like they were performing racial justice of some kind (whilst having zero black people in leadership!). Why wasn’t the change centralised through the Git project so everybody could make the change together?

https://web.archive.org/web/20201001133529/https://github.co...

Fortunately, with this change, everything will return to being in sync.

How dare you make sense! Unfortunately nobody who disagrees with you will eat this up.

As someone who has seen black colleagues happily using master/slave terminology in new code (embedded systems so this was around I2C devices), and the only one or two developers I’ve worked with who tried to push this renaming being white - I have two issues:

First, I think it’s just a dangerous idea that people have to be so coddled and be ‘protected’ from seeing words that aren’t slurs and have either no contextual relationship or only very tenuous relationship to actually sensitive concepts (‘master’ has many meanings divorced from African slavery and git never had slave branches so plenty of other senses of the word apply). I feel like it’s going so far that anybody who needs this probably haven’t yet really achieved the resilience needed to operate in the real world.

And secondly, since none of the black developers I worked with actually cared about it, it feels like it’s not great to do something and think you’ve done your bit for racism prevention when you’ve done nothing that makes any real difference to it!

> makes people happy

If your definition of "people" is "armchair political activists with nothing better to do with their lives", then yes.

Primarily because it encourages people to keep doing this performative stuff, think they're actually helping people, and take resources away from more meaningful work. By resisting these efforts, I'm preventing what would likely be runaway acceleration.

Git (and computer science) shouldn’t use ‘hash’ either since it’s a drug reference.

So a hashtable is basically the drug dealers menu?

Curiously "man" is etymologically gender neutral, where wo/wif(man) denoted the female prefix for man, and the male dropped what would otherwise be were(man), seen in for example werewolf, which is gendered.

So a female werewolf would be a wowulf then? Cool!

There is discussion on whether wer/e usage in werewolf has further/other roots, but assuming consensus landing on my GP, then yes!

wodude

so you’re saying a woman is just a female man? that’s crazy, dude…

Yes, "man" is etymologically a person, "woman" is a female person, and "wereman" is a male person. Over time "were" got dropped, but "wo" got kept, so "man" is either a gender-neutral person or a male person, distinguished based on context. E.g. "mankind" is gender-neutral.

yes

I was making a funny

« mann » comes from Old English and stands for a human being.

« wïfmann », literally "female human", led to « wife ».

« were » means man and comes from Germanic and I don't think « weremann » has ever been a thing.

Blew my mind when I found out “world” is a direct derivative of the word “were”.

“world” is derived from “were” + “eald” (old), and meant “the age of humans”, which was distinguished from the age of the Gods, when the Æsir and Vanir dominated, and the age of the Jötnar.

I find it interesting how the term shifted from a (mythical) temporal concept to a spatial concept, to now often a social concept (e.g. the Fourth World).

Don't give them ideas

How did Scrum Master escape this treatment?

"Sprint". "Bus factor"?

We didn't need for another argument against Scrum, but that's a very compelling one.

You're right. I'll call it Scrum Main now.

i suggest scrum whip (as in party whip, has a similar role and function)

Whitespace?

Does he have a MSc?

Because the term Scrum Master wasn't derived from master/slave.

Git's concept of a master branch was borrowed from BitKeeper which used master/slave terminology. https://github.com/bitkeeper-scm/bitkeeper/blob/master/doc/H...

Eh, that's some interesting historical trivia, but I don't see how that matters tbh. If everybody is fine with the word because they (quite reasonably) assume it was inspired by master record or something similar, and you bring up some 25 year old history (20 years when the debate got started) about not even git but a precursor, does that really help with anything? I am sure there are other innocuous seeming words that have a dark etymology, should we go search for them so we can update the language?

> If everybody is fine with the word

Some people weren't evidently.

The question was "How did Scrum Master escape this treatment?" - I think I answered that question accurately.

You're implying that those who are throwing hissy fits about "master" are aware of bitkeeper documentation and their (wildly unchecked) emotional response to this matter is nuanced enough to take "provenance" of technical terms into account.

Do you even realize how ridiculous these nonsensical "arguments" sound?

What's your theory as to why Git master had a lot of support for changing it while Scrum master did not?

at my work place we call them "Agile Champions" now

Ironic considering I've yet to see one of those be able to sprint 100 meters

I've never felt more like a slave than under the tyranny of a bad scrum master.

I’m disappointed that this headline will lead to more clicks. This is your reminder that in git the branch name is just a pointer to a commit. Renaming that pointer is relatively seamless on GitHub (https://github.com/github/renaming?tab=readme-ov-file#rename...). Also, git 3.0 isn’t forcing this change on existing repos, just new ones that no automation depends on. And if you really like the old name that’s always an option for your repos. Remember, it’s just a pointer.

The other git 3.0 changes are more consequential and worthy of discussion - changing from SHA-1 to SHA-256 for greater security and performance, changing the storage format for performance and introducing Rust.

> changing from SHA-1 to SHA-256 for greater security

Linus has a different view, he referred to the SHA-256 migration as "pointless churn": https://www.youtube.com/watch?v=sCr_gb8rdEI?t=11m

SHA-1 is not broken enough to be a serious issue for git. The migration to SHA-256 has been forced by on git by clueless morons, and it is, in this very special way, similar to the master-main rename.

Only if you don't have any pdf files in your repo.

Also wonders if SHA-384 or SHA-256 would be more "future proofing" or whatever the points the supporters making

Will Git produce repos with both SHA1 and SHA256 commits, or is this a per repo thing?

AFAIR, git is now kind of hash function agnostic, although SHA-256 is the only one there in practice.

How does Git know which hash type a given hash is? Is there a prefix, that is just never shown?

> just new ones that no automation depends on

Except for automations that happen to create new repositories.

The most intolerant wins again.

I'll be keeping “master”, than you very much.

Turns out people really, really, REALLY don't like to be forced to change their decades old thinking, vocabulary and definitions of words when they feel its either pointless busy work, propagation of an ideology they do not support or a straight up attempt at mind control.

You could have stopped at average people don't like to change. It being justified or not as very little impact on the reception.

It applies equally to this than to systemd, wayland or Rust - but Rust is now trendy so people will bandwagon in their appreciation to look cool and in - to quote other topics regularly discussed on HN.

I fear that the most interesting thing about how these changes are received by the HN community is what it says about the HN community itself.

They could have expressed themselves in a way that aligns with your sensibilities, but they exercised their freedom of expression.

I personally found that additional detail to be very valuable, because it conveyed disdain without triggering justification for people like you to censor the comment completely.

I have no interest in censoring the comment (what?). That’s nearly as far fetched as people calling renaming branches brain washing and propaganda.

It’s a rhetorical flourish pointing that in my opinion most people here are actually change averse, a trend which I personally notice more and more here.

I have no sensibilities. I am neither American nor interested in American politics. I am however deeply convinced that people having strong opinions about the names of branches have issues. An opinion I apply equally to people strongly advocating for the change and to people strongly against the change.

Mercurial uses "default" as the default branch since the inception

Good. "master" implies a master/slave architecture which is straight-up technically wrong - and directly opposite git's true design! You couldn't pick a worse default. Even if it weren't tinged with racist colonial overtones, there is reason to change it for correctness alone.

"trunk" would have at least matched the notion of "branches"

What about "boot", to be more international?

trunk as in tree, not car. Cars don't have branches.

To go with the way the world is going, maybe "boot" and "licks" instead of "trunk" and "branches"? =)

oh, don't worry, somebody will be offended by this as well - "trunk" as a sexual slang term refers to a woman's buttocks.

Huh. I always visualized "trunk" as some kind of chest where all the code was stored. Somehow it never occurred to me to think about trees. What a weird feeling.

i hope you don't get your source code tree (on disk) confused with the tree of branches in the repository then...

Interesting perspective

i knew it wasn't just me!

Following the logic of the master rename though, should we even then use the terms “tree” or “branch”, since so many people were murdered by being hanged from them in lynchings?

Since Torvalds hated CVS and Subversion with a passion, "trunk" was out of the question I guess.

extreme terminology prejudice

Cowardly acquiescence to any political ideology is disgraceful, and that's exactly what this was.

Or maybe it’s just people trying to be kind. Not everything is a fucking battle.

When we have sentient robots running around doing our bidding, I wonder what we'll call them.

We already call them slaves, but with a Czech loanword.

Been using 'sensei' as my default branch name for a couple years now.

I get to feel like a ninja when I commit, the conceptual meaning is close to that of the previous term, and there's no historical baggage related to the US. Win-win-win?

That just highlights that the way people are interpreting the word `master` is wrong. You've named your most important branch "teacher".

Isn't master ("ruler, teacher, etc.") upstream of master ("definitive version"), though, etomologically-speaking?

I do see what you're saying, though, and will admit to some cheekiness on my end.

If you try and push you sensei, before you know it you'll be doing jujitsu.

They should better fix bigger problems like real "decentralization" like using p2p protocols by default.

Their current state of "decentralization" is "the developer might still have a copy of the repo on his laptop".

Yeah. Advanced Direct Connect p2p is using TIGER Tree Hash. Gnutella2 p2p is using TIGER Tree Hash. There is BitTorrent format extension to include per-file TTH, implemented by EAD TorrentBuild. For best performance we should also get rid of pointless per-block hashes in torrent files.

TTH is universal for p2p, and yet developers choose another hashes, making their tools not ready for p2p.

My lovely Mercurial is unfortunately also not ready for p2p

At the end of the day, all I really care about is consistency. It's annoying to switch between projects which use different branch names to describe the same thing.

That being said, this is a dumb reason to introduce inconsistency.

I had this problem too until I forced myself to use the `gcm` alias from oh-my-zsh’s Git plugin. IIRC it checks out to master if it exists, else main. It has almost entirely removed this distinction from my life.

I hope you never come across a repo that has an old out-of-date `master` branch, because they moved to using `main` in a bad way.

No problem. The parent had it the wrong way around. OMZ will use main if it exists, master otherwise. So a bad move still works!

Thanks that’s good to know

>More formally integrating Rust into Git’s own build process

What does this mean, to compile git v3 I will need Rust ?

As for 'main' I always used that, I do not know how that happened since I use git the say way I used RCS ages ago.

Yes, you will need Rust to compile git v3 from source.

Personally I think only branch that's minimally needed is "dev", with tags for releases. "master" or "main" should just be a pointer to the latest non-hotfix tag.

Any project repository where I'm the original author, the default branch is the "develop" branch. Let's move on!

All the woke stuff was so annoying past 5 years. I don't remember the company name but it was one of those feature flag toggle companies. For a year straight they communicated this change they were doing, basically replacing blacklist and whitelist with more PC terms. Someone at the company decided to not only make it a recommended change, but that the old usage would BREAK the app. They couldn't handle the word blacklist in the codebase so much that they were willing to punish their users. It felt like I was getting an email from them every 5 days for a year.

That all being said - main is shorter than master and I would say most of us here, probably 95%, are "used to it" and it would actually be annoying to go back to master.

We used to default

Wasn't it already?

I don't get it how you can be so emotional for yet another branch name. Especially as this change was announced 5yrs ago.

If this blocks you from "just want to do real work" maybe overthink your setup and adjust your pipelines?

While we are at it, can we please not have a human "master" in every Scrum team?

Interestingly enough, the times when I've had scrum masters at work were coincidentally the times when I felt the most disrespected at work. Not from the scrum masters though, from the people who the scrum masters protected myself and my teammates from, the higher ups. The scrum masters were cool. Even on a personal level the higher ups were more or less cool. But damn, organizational and market pressures can really beat down on people.

I feel very uncomfortable with Chess players holding 'Grandmaster' titles. Would much prefer 'Grandmain'.

Who gives a shit. Seriously. You can set your preferred name in the configs. And I have this since forever because I do not want to have a sudden name change. This will change will not impact anyone. But again we get this "oh this is so horrible". Nobody cares. Nobody forces anyone to change anything

This is plain stupidity over some SJWs pointless whining, sad to see people being such braindead to support bullshit like this.

Yet again the whining twitter-people get another win.

From a third-world perspective, it feels like American politics being injected into the developer domain because of some previous biases that Americans had. Which is sad for a community that claims to be global.

Lots of Americans don't support this either. Of course, that doesn't actually matter to the people who have pushed this change through. They think that it's a righteous change, so it doesn't matter whom it annoys.

Lots of non Americans support the change too. Master was a terrible default name, switching it to a better name that’s also more inclusive language was a good idea. Where American politics infected things was the the reactionary response of people who are vehemently opposed to any attempt to fix problems that don’t directly effect them.

This whole 'master branch' outrage doesn't even make sense. Should saying things like: "I've mastered computer programming." also be an offense?

We absolutely should deny and disregard these nonsensical demands _by principle_. What was even the actual case made by people who wanted this? And don't tell me "well it's not a big deal, just accept it don't whine about something so small", because that won't fly, or shouldn't at least.

This is whining. Name it what you want.

Adulthood is about realizing people who 'just want to do real work' will always be fucked over by people who vocalize their needs and organize like-minded ones, justified or not.

So now instead of game remasters we will get game remains?

Don't know why people here got triggered by DEI, but the "master" naming is just bad, "main" is actually a better and more generic description.

The word "master" means someone/thing that has the capability of controlling things, like "Mastered", "Master Degree" etc.

But in most Git contexts, "master" is just "one of the breach that we hand picked to put our finest results in", that's not mastering anything, it just means "if you know what's best for you (or not), just use this one".

Another similar wording is in IDE hard drives. Remember the fun time where you can to setup jumpers before your secondary IDE hard drive would work? Yeah that secondary drive is called "slave". I'm still confused why the first drive must be called "master" since I never see it whipping any other drives to make them work harder or really doing anything that's remotely controlling.

The computer guys in the old times really have a weird taste in naming things.

Well said. I think the slavery connotations in Git are not a big deal. But I have always disliked the word "master" because it never seemed well-suited for what role it played in the version control sphere. Labels like "main" or "primary" always seemed like better words to me, so I am glad that this change was made.

Interesting thought process. I always liked the “master” branch to a “master recording” since most branches are created off master.

This was always the sense I thought of too

You betray your own ignorance. Master has been used historically to refer to the "final" or "canonical" copy. For example, you make copies of a CD based on the master. Creating that master copy is called mastering.

The word "master" stems from Latin, meaning "great" or "teacher". Just as teachers pass on knowledge to students, data is copied from the master.

Any slavery related connotations is insanely recent, and arguably manufactured specifically for the purpose of linguistic censorship. Historically, slave owners were referred to as... owners.

The "final" or "canonical" argument can only indicates that the word "master" has multiple meanings, which is not a really special. Tho, "Master Degree" really means teaching, my bad.

> Just as teachers pass on knowledge to students, data is copied from the master.

In the Git context, the "canonical" way to use git is that you merge/copy data _from_ other branches, such as `wip` or `dev`, _back to_ `master` or `main`, not the other way around.

All and all, it's done, OK? Don't like it? Hey, you can still use "master", just customize it. But you'll probably piss some people if you do, and then you have to dig out your dictionary again and again. Choice wisely.

For IDE you are mixing up 2 separate concepts. The addressing primary/secondary is purely addressing, I don't recall seeing "Master/Slave" in relation to that. Labeling on HDDs I've seen always used primary/secondary

Where the master terminology comes in is that a certain version of ATA added bus mastering DMA from the drive. Maybe some harddisks had a jumper to disable or enable that

This is verifiably untrue. Older IDE hard drives commonly used master and slave on their printed labels instead of primary and secondary. Google “ide drive master slave” images and you’ll see plenty of examples.

I don’t say this to justify sticking with the older terms, just to assert that they were actually used.

Hmm interesting. That is true about the labelling of jumpers, I couldn't even find any images with other naming schemes

There is a double confusion here because "primary" and "secondary" were used to refer to IDE channels. Most machines had two IDE channels (i.e. physical connectors on the controller card), each of which could have a master and slave device (two connectors on the cable).

So you have IDE0, primary, and IDE1, secondary. For the four devices a typical system would support, they would be referred to as primary master, primary slave, secondary master and secondary slave. This was extremely accepted terminology.

Newer machines and BIOSes could usually boot from any of these four devices but originally, many machines could only boot the primary master. That's why it's the master and the other one is the slave -- it is subservient in the sense that it can not be a boot drive and was usually used for secondary storage, not OS.

I think you're missing the historical context[0]. It came from "master-slave architecture" which has an older history. It's why we refer to a master copy of a photo or document too.

I really don't care if it's main or master tbh. But is anyone actually offended?

[0] https://en.wikipedia.org/wiki/Master%E2%80%93slave_(technolo...

With master gone, it's time to get rid of the man-pages, which were invented to mansplain commands by bearded men in the 70s.

I eagerly await that change.

However, "man" originally means "people" as in "mankind", or "human". It's only recently (c 1400) that it replaced "wer" ("wif" being the feminine noun) as meaning males (or mid 20th century for becoming an exclamation).

In addition, this is entirely pointless, because man is short for manual, which comes from manus, the hand, because it is something, that you should keep at hand.

Somewhat disheartening to see all the arguments here saying it doesn't matter how we name stuff. Lived language creates reality.

There are groups of black people that complained about the terminology, repeatedly [1]. So isn't it arrogant as a white dude to say, this thing hurts you, and we could easily change it, but fuck you we won't do it because we have always done it that other way?

Some easy no effort replacements:

- master-slave -> leader-follower

- blacklist -> denylist

- whitelist -> acceptlist

[1] https://en.wikipedia.org/wiki/Master%E2%80%93slave_(technolo...

This argument isn't just absurd but push it to its logical extreme, and you're inviting brutal thought policing and full-blown totalitarianism.

I stick with "master" in my Git repos partly because it's an excellent filter: it lets me steer well away from anyone who pitches a fit over the word.

Ah yes, the if we change something that would mean we would have to change everything and that would be absurd, so we can't change anything ever argument. Classic.

Seeing some of the complaints around this feels like people are somehow still stuck in 2020 instead of 2025. People need to do a fresh pull down from main and update their arguments.

Frankly I don't really care either way, main is shorter and conveys the same meaning so by my metrics it's better. You can override the default to be whatever you want.

Reply on news.ycombinator.com