The sentiment here seems pretty anti-surveillance -- but why the false dichotomy? Both are important issues. It just comes off as demeaning to the people that died in 9/11.

So much vitriol here today, jeez.

I had the same thought about the death count when I was 11 years old. Then you grow up and realize why an attack of that nature on American soil was and still is the big deal. That's how they were able to push the PATRIOT Act and normalize surveillance, the death count numbers weren't scary enough to get that job done.

Okay now you are just being demeaning to victims of drunk driving

> Fixing this broken language and attitude is a step towards ending these deaths.

Is it? The person you responded to sees car crashes as an important issue to solve yet still calls it "car accidents". So maybe it's not that important after all.

Nothing makes me saltier than language policing; I'd say fixing _that_ broken language is more of a step towards fixing issues than the actual language itself.

It's not about language, it's about culture. If you saw someone running through a crowd with a knife you'd expect them to be thrown in prison. But someone speeding through a neighborhood? Or dangerously overtaking a cyclist? No bother.

If the person running with the knife hurt someone they'd be a murderer. But the car? Oh, it was just an accident.

You think they were doing it on purpose every time someone is in an accident, or you think every time someone says "accident", they're referring to or you're purposely missing the nuance to make a point nobody agrees with, or you think that people in general are not annoyed when someone drives recklessly?

If it's the first, I dunno - you're wrong. People aren't. If it's the second, aside from insurance fraud, nobody is trying to fuck up their day and lose a functional car in the process, maybe getting seriously injured. If it's the third, you're just being annoying and nobody will take your point seriously if you can't make it more saliently. If it's the last, go check out reddit. There are at least half a dozen subs where people post bad drivers and everyone in the comments is collectively grumbling.

Are they plane crashes or plane accidents? (Replace "(air)plane" with "car".)

Accidents. The aviation industry is the worst example you could use. Every single time, the goal is to find the issue, NOT the blame. The industry is insanely safe in large part BECAUSE they refuse to frame it as a crash, and instead frame it like an accident.

But if there is blame, pilots will never fly again. They don't receive a slap on the wrist and an "oh, you".

Well yeah, it’s not a faultless thing. If you’re being reckless and insane they obviously fire you, but literally nobody does that. It’s not an easy job to get, you’d be throwing away a great career. Closest you’ll come is someone getting fired for drinking and piloting.

There is a strong push in many sectors to stop using the word accident, so your assertion that this is a point no one agrees with is false.

Oh, you think I was being literal

Funnily enough, this is more readable to me than most Clayde jargon.

the person you responded to has been a user since 2019. why do you think they artificially push it?

I also use jj as my daily driver now. I like it a lot. I have no skin in the game either.

Which written language has the most history of terse, succinct writing? If Claude doesn't improve I'm ready to learn a new language just to avoid its prose. I'm only half-joking.

Better start chinamaxxing

Probably Mongolian

This was the feature I wanted the most on Github... until I started using jj.

I know that jj isn't for everyone, but for me it's been trivial to see the state of and update all the stacks of branches I have on the go.

Not that I don't welcome this feature; I just think I don't need it anymore.

I am confused - I love JJ and stacked diffs with Gerrit and Phabricator, but trying to port that same "each commit gets reviewed individually, and things don't break when a half-reviewed stack gets rebased" workflow to GitHub has been a nightmare, how does JJ help?

Here's the simplest form I could distill it to. The other comments I find quite verbose or confusing, no offense intended.

    x: Ann's current age
    y: The age difference between Ann and Mary

    24 = x + y      (ie. their current ages)
    x = (24/2) + y  (ie. their age in the past)

    Solve for y in one equation and plug it into the other. You'll get x = 18.

That's not how I read it. I read it as:

  Now : Mary is 24 = 2y,  Ann is x
  Past: Mary is x,        Ann is 12 = y

Moreover, we have x + delta = 24, and 12 + delta = x (they get older at same rate), so delta = 6 and x = 18.

Uhhh... what? The OP is right; even if you do it intuitively it's still algebra. If you need a proverb to justify it: Just because your hammer is made of stone doesn't make it less of a hammer.

This works really well for me but, when I get out of the car or off the subway, then I get a bit of vertigo. I’m really afraid that using the feature too much would make the vertigo worse so I stopped.

One thing I don't quite get is how trusted publishing is supposed to be more secure. It still allows publishing if they pwn your workflow.

Is it purely more secure because they can't exfiltrate your secret keys to publish again?

I feel like if your workflow gets pwned you'd be rotating your keys anyways, so I'm not sure if the vendor lock-in is worth it.

Trusted staged publishing helps a lot: you have to independently pwn the workflow _and_ then complete a separate 2FA flow as a maintainer. The workflow never sees any keys that can publish independently.

Keys can be reused from anywhere. Trusted publishing means the attacker must trigger the specific workflow on GitHub, which is more difficult and leaves trace of actions on GitHub itself.

Trusted Publishing pins the publish to the workflow. The next step is to isolate "run untrusted code (`npm build`)" from "publish" in your workflows. Now you have two properties - the publish can only happen from one place and the publish is isolated from untrusted code execution.