I don’t know the original author’s opinions on AI, but I think it’s still fair to say that it was clearly Copilot’s goal to steal all the code and be good, and they had all the code, and so ethics aside one would expect them to have succeeded with flying colour at whatever their goals were, even if those were bad goals.

But they failed instead, which is impressive.

I think you’re looking for the word “cue” my friend!

Ah, true, all fair points! It really is pretty apples to oranges comparing different lengths and averages and making up what might have happened.

I was curious about this, so I figured I’d just compare the 2024 Summer Olympics men and women’s races. Well, bad news upfront, the men’s race was 273km long, and the women’s race was only 157km long. Not a great sign out of the gate, but maybe that’s for dumb reasons.

The man who won gold finished 273km in 6 hours and 19 minutes, and the woman who won gold finished 157km in 3 hours and 59 minutes. A little tricky to compare…

So I figured I’d compare average speed across the whole race. Obviously this isn’t an ideal metric, because presumably a racer could go faster if they were racing a shorter distance, and so the women with the shorter race should be able to go faster than if they had to save more energy for a longer one. But I figure despite that, it would be at least the start of a comparison.

So the man who won gold had an average pace of 43.22 kmph over nearly 6.5h, and the woman who won gold had an average pace of 39.41 kmph over about 4 hours. So 43 kmph and 39 kmph aren’t that far off, but again over 6 hours that 4 kmph difference equates to 24 km of difference, and feels like a pretty big difference. And again, this is with the woman having the advantage of a shorter race.

Okay, not looking great. But that’s gold to gold. If we make some leaps and assume that the woman could sustain that same pace for the duration of the male race, or that the men’s race ended early but somehow they had the same average pace anyway, where would a woman with that pace rank in the rankings, if not gold?

Well, I can’t be sure, because at 77th place Charles Kagimu of Uganda went 273km in 6 hours and 50 minutes, which is 39.95 kmph, still faster than her 39.41. Everyone slower than 77th place seems to have gotten a Did Not Finish and thus no time was recorded.

So despite the fact that the women’s race is substantially shorter than the men’s, the gold medalist for the women’s race was slower than 77th place in average pace during her shorter race, and may not have even finished the race due to being too slow. So I don’t think it worked out.

Results I’m comparing: Men’s Race Women’s Race

I think we agree more than we disagree, but are at different points on the spectrum. For example:

Again, I don’t understand why a “line needs to be drawn” based on some imaginary attack on the English language. What’s the threat here? That someone submits a change while saying the word “aardvark” is offensive? Just reject that pr and move on.

This isn’t meant as a “gotcha”, but in this paragraph about not drawing a line, you drew a line. You decided aardvark was obviously too far, and that that PR should be rejected. How you feel about aardvark is how most of us already feel about the word “stupid”.

But more broadly:

Maybe if just a few people are hurt by something, and the choice is between doing nothing (and them being hurt) and saying “no worries, send the change” and not hurting a few people, we can just… Not hurt them? Seems straightforward.

I think most people (in this community, on this thread) are not pro hurting people. What I feel is more like: if you are hurt by the word “stupid”, or self-identify as stupid, you should not. No one is using it as a slur against your people. There are slurs! They exist, it’s just that this isn’t one of them, in the way people mean it. And so I feel like, in this case, at this point in the spectrum, these people should heal themselves rather than change software / the culture / the world to suit their insecurities.

If course it’s a squishy grey area, but if I found the word aardvark offensive because some kids called me aardvark at school growing up or something and bullied me, that’s tragic, and it’s very real for hypothetical me, but that’s something I should work through in therapy, rather than something I should make the concern of everyone around me. In my opinion. And I feel like being triggered by the word “stupid” is in the same category, also in my opinion.

If anything, and I’m stepping in bees again, it feels kind of egocentric to see someone write “replies are stupid” in their own code, in response to presumably their opinion about a standard or spec or something, and to see they’ve written that and think “this is about me”.

As an unfortunately pedantic person, it really bothers me that blacklist and whitelist get caught up in all this. Like, yeah, I can see why people think it’s related to skin colour, and I can see the argument that even if it wasn’t originally about skin colour, it leaves an impression of “white good, black bad” regardless of its original intentions. But fuck do I wish we didn’t call white people white and black people black. It’s not accurate, and would solve a whole bunch of these “colour-related phrases becoming racial” problems. We should just stop using colours to refer to people! But that ship has long sailed, and its harder to advocate in that direction, so I guess I’m fine with it. But I can dream 😛

Also “master” has other uses, like a Master Sculpter making a masterpiece, and more relatedly things like the “master tape” being the tape other tapes are copied from, a la “remastered”. But I concede it’s pretty hard to make that argument when DBs and BIOSes have “masters” versus “slaves” 😬😅

That said - is it really important to defend “stupid” as a word choice? Does rewording it, maybe to “senseless” or “ignorant”, create some huge negative impact for a user? It seems like kind of a minimal effort solution that can accommodate users, so why make it a big deal?

I know I’m wandering through a nest of bees here, but this cuts both ways, I think. No, this particular word isn’t important, and changing it is fine. Any one word can be fine. But similarly why did this user show up asking it to be changed? Is it a huge negative impact to leave it for the majority of users either? It feels like someone pulled a dictionary of newly bad words off a blog and grepped through the source with the perceived mission of contributing to the healing of the world, as a most charitable assumption on their intentions.

I think no one is worried about any one word, or any one PR. The concern is that the goalposts seem to change from words that 95% of people agree are bad, to words 60% of people agree are bad, to words like this that maybe 1% of people feel are bad, and there’s a grey area here on what level of badness is bad enough for all of us to change to accomodate one or two people’s sensitivities, and to what level those people should be responsible for their own sensitivities.

This is a civilization and cultural level spectrum which has “change for your society” and “society bends to you without change” at its ends, and different people fall at different points on this spectrum, which will put that at different points on the “how bad does a word need to be for me to be a bad person for typing it in my own code” spectrum. And for me, I feel “stupid” is over my line and is a noisy change that might beget other more petty changes with no benefit to the vast majority, despite how simple it is. But you clearly feel more strongly, and I can tolerate that too.

All that having been said, I have no opinion or context about this particular user being banned from this particular chat, unrelated to the ethics of the PR.

Yeah, this one feels like an overreach to me. Calling a person stupid is bad, fair, but that’s because the word is negative, and calling a person a negative word isn’t nice.

But in this case, “replies_are_stupid” has nothing to do with a person. They’re inanimate. So calling replies stupid is labelling replies negatively, but that’s fine because they’re inanimate. I’m not sure “loweffort” is better or even applicable. I guarantee someone will have a problem with “loweffort” in the future. Maybe “ill-advised” would be better in this context?

And before I get strawmanned by someone saying “would it be the same if they called it replies_are_gay” or something, I think that is different, because that’s implying gay is bad, which is the actual problem with that usage. It’s inaccurate to the problem, and only makes sense if gay people catch strays.

If you read “replies_are_stupid” and felt attacked, you need a better therapist, because “stupid” isn’t an identity you should feel for yourself or those you love.

It’s nice of this poster to ignore the $181m spent on “other projects” and conclude this is some kind of scam. If we include the Linux Kernel with the other projects part, that’s about 67%, or two thirds, of their expenses are paying for various and assorted open source projects. Among them the kernel. So if you’re a “cash and cash alone” person, then 2/3rds of your money is still going as cash to software projects.

And if we include things like community tooling and project services, which may help a project in ways beyond just cash that becomes about 78% in total, or over three quarters.

That’s pretty good, I think, but to each their own.

Right, but what I’m saying is that git doesn’t store authorship information or line-by-line history, no matter how it’s done. Figuring out which line came from where is an algorithm the git blame command does every time you request it, and that algorithm can give different results depending on which options you give the blame command. And so what you’ve found here is a collection of commits that produces a situation the default blame algorithm can follow, without any optional flags, which is neat! Maybe not great for git history, but neat!

Interesting. Yeah, sounds like what git blame -C is for, so I’ve never made copies when splitting files, I’ve just moved lines between files naively. But I guess if one’s tools are limited and doesn’t have the ability to -C, then I guess I could respect the hack that is that solution?

I mean, I’m 99% sure git doesn’t store blame or authorship info in the pack files, even as a convenience cache, and just guesses by traversing the patch log with heuristics live when you run blame anyway, so the history mostly doesn’t matter there, but the way you’ve done it does seem to have tricked the heuristics into doing what you want without relying on an option, so that’s neat! It’s an interesting hack, and I like interesting hacks 😛

By the way, if there are down votes, they’re not from me!

I know words are squishy… but are these… films? It seems a bit strong to use that word to describe my glasses recording me peeing or whatever…

Yeah, I’m with you. I mean, git isn’t magic. You “can” squash anything, including a merge commit, by just being at the end result, running git reset <commit you want to be squashed off of> and then running a manual git add and commit there. That’s basically all a squash is.

But what you’ll be left with us a single commit that contains all of the code from the branch you’re squashing and also all the code pulled in from every branch you merged, all written as though it all came from this one commit. And maybe that’s what you want? But it feels like also maybe it’s not?

Right… for sure… but then if you don’t want to squash, then it doesn’t matter you can’t squash a merge commit.

Huh. I have never in my 19 year career using git, ever wanted to copy a file and pretend all of the history of that file is also the history of the new file. I mean, I don’t think I’ve ever even wanted to copy a file? Why are you copying a file?

Like, maybe I’m just too familiar with git to see the forest for the trees, but what the heck are you doing over there? 😅

And just in case it’s useful, a tip is that you can use git blame -C to have the blame algorithm use a heuristic to try and find a “source” line if it was moved, including from another file, during a commit, and then continue following the history of that line, to try and get the real commit where this was written, not just the last time it was moved around.

Yeah, 2 was just 1 again, with all the same jokes but in reverse. But 3 was legitimately different and better, so it saved the whole thing!

Lord of the Rings barely counts, because not only were all three books out and classics before the movies started (obviously), but the three movies were basically worked on at the same time. It’s nuts, but somehow they managed to do it.

So it’s not like they released the first, got crazy hype, and then phoned everyone up and said “electric Boogaloo, you in?”. They’d already shot most of the second and third by the time the first came out, as I recall.

Also I really liked Glass Onion 😛

Yup, basically. Systemd is “the first program” that runs, and then its job is to start all the other programs that make up a modern computer, most of which run in the background and a user will never see. It’s not the only such “init program”, though, and some people are grumpy that it does too much itself, rather than simply starting other programs to do those things.

But unless you’re involved in starting and stopping background processes, you can’t really tell which one you’ve got. Users aren’t “meant” to care which process was the one that started the power management daemon, or whatever.

Yeah, I 100% guarantee they aren’t thinking of that at all. It’s really clear both from the culture at large, and the contents of the message by “porn addict” they mean “horny often and into sex”, and by “they don’t want you” they mean you aren’t up to the standards of enjoyment they get from the porn they like.

Sure, but RAG has a Wikipedia article about the specifics of the process, history of its use, links to papers and articles about it and its advantages and drawbacks. It’s also useful as a feature on a matrix for comparing one tool or model’s capabilities to another. None of that is true of the sentence.

Virtually all of computing could be reduced to voltages across terminals changing over time, but it can still be useful to give specific terms to specific applications of this process, so we have something to talk about.