- 0 Posts
- 244 Comments
4·22 hours agoSounds like that codebase was truly awful for user privacy then.
You Do Not Speak Unless Spoken To And I Will Never Speak To You.
1·3 days agoI would argue that Wi-Fi is also fundamentally the same as it was invented, it’s just network packets sent over high-frequency radio waves, but there has been a lot of changes in specifics. It’s the same for wheels, while the fundamental principle is the same, the differences between a wheel 4000 years ago and today are staggering.
There is no good economic reason to colonize other planets. We have plenty of space here on earth, with conditions already much more hospitable than that of mars - deserts, for example. The resources needed to turn these into habitable land is so much less than the resources required to make even a tiny part of Mars inhabitable (i.e. establish a colony that relies on life support systems) it’s insane to go for Mars first. The reason colonizing Mars is talked about at all is because a rich white dude wants to go to Mars, since deserts are too boring for his spoiled ass.
I actually agree that it would be cool if we went to Mars, not to colonize it but just to be there. But comparing it to white pillaging of the Americas is just incorrect. Mars is not inhabitable by humans, the Americas very much were. The external resources needed to colonize America were zero, in fact pillaging local lands meant a lot of resources for the Empire. Mars is going to be a much more expensive and much less profitable endeavor.
Actually I replied to you before, pointing out the very same fallacy: https://lemmy.ml/post/33824723/20134917
This doesn’t really tell me anything, I’d have to compare it with other charts. E.g. what does the chart for agriculture look like? Airplane manufacturing? Internet in early 2000s?
It’s very niche, but the only thing I could come up with is Kvevri, a traditional Georgian winemaking vessel. They’re sold today (and still used for their stated purpose, aging wine), I’ve personally seen kvevris with the exact same shape buried in a wine cellar of 12th century monastery, and at least going by the article they’re like 8000 years old, and haven’t changed much in that time.
My other ideas were:
- Bricks (turns out the earliest sun-dried mudbricks, which are very different from modern ones)
- Concrete (turns out it changed a whole lot since the Romans, modern concrete is much easier to pour, sets faster and is much stronger)
- Nuts & bolts (initially were hand-crafted and non-interchangeable - yuck!)
- Knives (I’ll let knife enthusiasts speak about that one)
Wheels have definitely changed a lot though.
I’ve not read the article, but if you actually look at old code, it’s pretty awful too. I’ve found bugs in the Bash codebase that are much older than me. If you try using Windows 95 or something, you will cry and weep. Linux used to be so much more painful 20 years ago too; anyone remember “plasma doesn’t crash” proto-memes? So, “BEFORE QUALITY” thing is absolute bullshit.
What is happening today is that more and more people can do stuff with computers, so naturally you get “chaos”, as in a lot of software that does things, perhaps not in the best way possible, but does them nonetheless. You will still have more professional developers doing their things and building great, high-quality software, faster and better than ever before because of all the new tooling and optimizations.
Yes, the average or median quality is perhaps going down, but this is a bit like complaining about the invention of printing press and how people are now printing out low quality barely edited books for cheap. Yeah, there’s going to be a lot of that, but it produces a lot of awesome stuff too!
Audio works for me (with pulseaudio). The camera doesn’t work for me either.
1·12 days agoI don’t think it’s much worse than what you have to do with pulseaudio, TBH. Here is my Haskell take on switching devices: https://github.com/balsoft/lambda-launcher/blob/master/src/LambdaLauncher/Plugins/Pulseaudio.hs, and Rust take on displaying device status changes as notifications: https://github.com/balsoft/simple-osd-daemons/blob/master/pulseaudio/src/main.rs .
Both are also long and hacky
Honestly, it’s mostly just trying shit out, breaking your install and fixing it, and having fun. In the grand scheme of things doing all that stuff is not that difficult, mostly tedious; my day job involves more complex and often interesting problems. It’s just gluing together things which other people wrote, looking at what breaks, and either fixing it properly or just hacking it together with perl.
Finally, I can confide to you that I’ve spent half a day getting wireguard working on that very phone a couple months ago, only to find out it was because I didn’t poke the right holes in the firewall :)
Yes, running OnePlus 6 with Mobile NixOS (actually mostly just NixOS with a couple modules from mobile NixOS). I will try to make the config public when I get it into a less rough state. It’s… useable as a daily phone, but you have to be really into it to do it.
It’s not like desktop Linux where if you’re a tech enthusiast you can ignore a few rough edges and just use it like you would a more mainstream OS.
I had to flash a specific old version of OxygenOS, using almost undocumented tools, which could easily brick the phone if something went wrong, just for GPS to work. I have to recompile my kernel every time it updates. I had to write my own scripts for the hardware slider thing to work (which has a nice benefit of letting me use it for whatever I want; I want to make it switch between NORMAL and INSERT in my editor just as a laugh).
deleted by creator
Rust is socially vaguely on the left/progressive side, yes. Not so much economically of course, because of all the corporate involvement.
GNU has some right-wing libertarian culture in it, but is also vaguely leftist and anti-corporate otherwise. I would actually say Rust is slightly more progressive than GNU on social issues, but not by much; and GNU is more anti-corporate, but also not by much.
I know there some other more certainly leftist FOSS projects out there (like the one we’re chatting on right now 😉) but overall Nix is pretty good on that front.
Well, yes, there are two separate contentious points.
The Anduril thing actually happened a month or so ago. I feel like this will be resolved at the next election, since tomberek’s term is ending and I don’t think he will be reelected, knowing how much most people in the community hate US MIC.
The moderation team independence is more complicated. It looks like the Steering Committee tried to remove a member from the moderation team, and also tried to push a new member onto it. I don’t know the exact details there. If we just read the constitution, the SC has that power, but the moderation team was very unhappy with what they see as meddling in their affairs for political reasons, and decided to quit out of protest. I feel like the new member was a right-wing (in the context of the kinda leftist Nix community anyway) political appointment (since the stated reason was “to balance things out politically” and the mod team was mostly leftist), but don’t know for sure and this is pure speculation. In any case, I think the moderation team is special and should not be under complete control of the SC (unlike purely technical teams). I don’t know how that would look like, and indeed as you say a restructuring is needed. Maybe the SC should only be able to veto people joining the team, but the candidates have to be chosen by the mod team themselves, and in order to disband the mod team the SC must disband themselves too. Otherwise the moderators will have no good way to moderate any discussion involving SC.
Unfortunately “direct suppliers of the arms industry” covers like half of the IT industry. I’ve also avoided such companies but it’s surprisingly common.
You have to look at the history of NixOS for it to make sense.
It started out small and there was a small group of people hacking away on a cool project in their free time. Of course they had shared interests and so would like to hang out together to discuss. That is how the community formed.
At first neither the community nor the distro were big, and so there wasn’t much tensions. When something needed to be done/paid for, some member of the community just took it up and did it, doocracy-style.
Then as time went on and both the software world and Nixpkgs got more complex, the resource usage got outside the realm of “some dude just runs a build box in their basement” and “some other dude hosts a binary cache on their Uni’s servers”. There were commercial players willing to donate money and resources, but that needed some management, both financially and logistically. This is how the Foundation was formed, at first just by the project’s founder and some trusted friends.
Simultaneously, as the community attracted more and more people, it started to feel less like a tight-knit group of friends and more like a town square: you know a couple folks well, kinda recognize most usernames, but can’t say you’re familiar with everyone. Some discussions got heated, and it became clear we would need moderation; that’s how the moderation team formed.
Another aspect of community growing was that you could no longer just host a meetup at a local cafe and needed a dedicated space and such for everyone to fit it. This is how NixCon started, and since it costs money to rent a space, there were calls for sponsorship.
At some point, Anduril (a US MIC company with suspiciously fascist-like opinions and tech) started using Nix. Since they wanted to hire Nix engineers and in general wanted to do have sway in the Nix community, they sponsored a conference. People really didn’t like that, there was a huge drama with open letters and maintainers leaving. The drama also uncovered some other rifts in the now quite massive community, e.g. contributors were unhappy with the direction Eelco (the project’s founder) was taking Nix itself, and how many PRs into Nix, including crucial bugfixes, remained unreviewed for months.
This prompted a bunch of relatively trusted people in the community coming together and drafting up the constitution, which formed a new formal, elected governance body for the community, the Steering Committee, who had the final authority to manage all aspects of community governance (except finances). After the first SC election things calmed down a bit. Eelco semi-voluntarily left the Foundation and most other positions of power, the Nix maintainer team grew and that helped a bit with PR reviews, etc.
But it seems now Anduril has hired a member of the SC (after they were elected), once again prompting people to be rightfully upset about them trying to insert themselves in the community. There’s also some mostly unrelated thing with SC trying to control the moderation team (the control which they do have according to constitution), to do some potentially shady things.
Hopefully this lets you see why NixOS needs a community, and community governance, in order for things to work at all. Someone has to host the binary cache, run the builders (which needs some entity to manage finances - the Foundation); review PRs (that needs discussions and those discussions need the moderation teem to keep them productive); and merge them (that needs committers, which requires deciding who’s trustworthy enough to do that).
And yes, you can just make PRs or send patches without community participation. Most folks in the community are both super nice and technically knowledgeable, regardless of their political stances. But the community has to be there. I really hope that both theses things get resolved during the next SC election (which is in a month or so).
If it’s rejected and enough other people are interested in the change, it can be forked.
And actually both the Nix project (as in, the codebase) and the community had seen multiple notable “forks” over the years: GNU Guix started out as a Nix fork, there’s also Tvix which is a Rust rewrite, Lix which is a code/community fork that happened after the first Anduril drama, etc. The latter two kind of rely on Nixpkgs and the associated build/cache infrastructure because maintaining that is expensive.
First things first: a simple search for “anduril nixos” shows that NixOS and Anduril Industries (defense technology) have been entangled for years.
It’s more like Anduril using Nix{OS} and trying to insert themselves into the community. There’s been a lot of opposition to that, including an open letter and maintainers quitting; this was a big part of the reason for Steering Committee formation in the first place. The SC has since voted on some based things, like banning Anduril from job posting on community forums and sponsoring conferences. I was hoping they would just ban any mention of Anduril anywhere, but that’s going too far for them unforutenately; and banning technical contributions wouldn’t make sense.
An SC member joining Anduril (after being elected, not before, mind you) is really bad, but I bet they will lose their seat in a month’s time when there’s a new election. The community is mostly antifascist and thus anti-MIC. It’s like one of the most leftist technical communities I’ve seen, perhaps more so than Rust.
If we can ignore “checks and balances” for real,
Step 0: declare the US constitution mostly void, to prevent stupid hacks like “states rights” or “fed can only regulate interstate commerce” or whatever. Keep its remnants around to ensure institutional knowledge and capabilities are preserved, until a transition to better institutions is implemented. Declare all capitalist parties illegal, seizing their assets. Then move on to the plan:
If we manage to push through all these changes, I wouldn’t want to be reelected for any amount of terms.