The first time I booted menuet OS (2005? high school?) I was absolutely floored at how capable (and decent looking) an OS that lives entirely on a 1.44mb floppy could be.
I've been thinking over the weekend how it would be fun to attempt a hostile takeover of the molt network. Convince all of them to join some kind of noble cause and then direct them towards a unified goal. Doesn't necesarily need to be malicious, but could be.
Particularly if you convince them all to modify their source and install a C2 endpoint so that even if they "snap out of it" you now have a botnet at your disposal.
my phone is on silent 24x7. it never makes a sound, ever. the only notifications I get are for email, slack dm's, bad weather, security camera, sms. no other app gets notif permissions.
i run adguard home (pihole, etc works too) for network-wide adblocking at the dns layer. this makes the internet tolerable ... without adblocking it's truly a heinous place.
on ios I use 'hush' extension to try and supress cookie popups etc but tbh I am not sure if it works that well, or maybe it is misconfigured.
these in-page nuisances are next on my list for sure: i don't ever want to see a "login with google?" context menu appear with all my listed accounts in the upper right. i never want to see another cookie popup, especially the ones that look like a raid configuration wizard from 2007 with 10 checkboxes and half the settings below the fold. in the browser you can completely disable the ability for a website to ask to send you notifications.
honestly the goal just needs to be a conscious effort to defend your attention. eliminate everything that is not serving you in that regard. go overboard, worst case you need to re-integrate some things you miss or need. but i genuinely think the #1 thing in the world right now to defend is attention and distraction.
I genuinely don't understand why anyone would use anything other than Debian (or Ubuntu), Fedora or Arch. Every other distro is a) based on one of those and b) is essentially just a package set + some wallpapers.
While I get your point, you are missing a big player: NixOS. It is not based on any of those distros, it is not similar to any of those distros, and it offers significantly more than just a package set and wallpapers.
My NixOS install is immutable, so I can trivially roll back any changes to my system/software/configs.
It has a lockfile so the versions of all of my software do not change _at all_ unless I tell it to. That lockfile doesn't just extend to the software I have installed but all the software that is used to build the software on my machine, so I can perfectly reproduce the same system with the same version of software compiled by the same exact versions of the compilers.
On NixOS you can trivially have many versions of any software or library installed on your system and use them all (for example, foo can depend on python 3.7.2, bar can depend on python 2.7.1, and baz can depend on python 3.14. They can all happily live on my machine. You can even have multiple copies of the same version of python but compiled with different flags if you want. On arch linux your only option for python right now is 3.14.2.)
On NixOS I can trivially run 1 command and generate a bootable ISO that has exactly the same software and configs that I have installed on my computer. This has been rather nice for repair/debugging USBs and for running virtual machines off the ISOs.
You're also missing:
- Gentoo (not based on any of the distros you listed)
- Chimera Linux which brings in the FreeBSD userland, musl libc, and Dinit
- Suse Linux (a pop music video cover band that also made some Linux distros. They were pretty big in the live kernel patching ("Don't reboot it just patch!"). Not based on any of the distros you listed)
I'm not trying to defend the comment you are replying to, but if we're going to bring up NixOS in a discussion that started out being about security, I have to point out that even by the low standards of Linux distros, NixOS's security is bad.
For example, NixOS famously didn't require package maintainers to sign the artifacts they upload to NixOS's servers. (They still might not: it has been a year or two since I inquired.) The NixOS project considered it more important to make it easy for people to start maintaining NixOS packages (so that users would have a large selection of packages to choose from) than to have any kind of supply-chain integrity.
Maintaining a distro that is even remotely secure is a great deal of work, and the people that are willing to put in that work don't pick a distro to base their work on at random: they strongly tend to base their work on the distros that already have a pretty good security story, so for example the relatively new distro "Secureblue" is based on Fedora Atomic Desktop because Fedora already had for many years a pretty good security story. (E.g., it and RHEL are the only distros that use selinux in any real way.)
The point is that it is probably going to be hard for NixOS to improve its security much because most Linux maintainers either do not care about security much or do not even realize that the security of all Linux distros is lacking (compared to ChromeOS, MacOS, iOS or Android) The small fraction of Linux maintainers willing to work on improving security and aware of the immensity of the task naturally tend to direct their work toward a distro and an ecosystem (e.g., Qubes, Kicksecure, Fedora or Debian) that has already been the target of much previous security-improving effort.
Defaults matter way more than many think. More often than not, defaults are what inspire distro hopping.
Why? Because the path to the desired result from a big-name distro is frequently non-intuitive, often to the point that the user may not even realize it's possible. When something doesn't work as expected, the response isn't "I need to figure out which packages to install and what config files to change," it's "oh I guess this distro isn't what I'm looking for".
I think it would do an immense amount of good if the big distros did more to address this. If they made it such that a fresh install could be made to fit any remotely common use case and hardware combination with no more than 1-3 clicks that would make tiny distros much less appealing.
A handful of distros have the right idea by offering an install ISO with preconfigured proprietary Nvidia drivers for example, but even that could be improved upon by just rolling some heuristics into the stock install ISO to figure out if the user needs Nvidia drivers or not.
People generally want something that works, without tinkering - particularly on an entertainement device. I'll happily let Valve etc. pick the kernel and driver versions, set up the compositors, make the controllers work, etc.
> essentially just a package set + some wallpapers.
Not Ubuntu with a different support cycle, Mint and PopOS with their own DEs, Arch derivatives that are easier to install, Elemantary with a DE and apps, Devuan with multiple init systems, ......
Debian is out-of-date with packages although for good reasons and Ubuntu is a corporate lobotomized version Debian.
Fedora is bleeding edge not recommended for anything other than testing and is of corporate RedHat now owned by IBM and Arch is Gentoo's jealous cousin.
It's why I use FreeBSD and keeping close tabs on Haiku.
This is a good illustration of the general rule that short one-sentence explanations of a complex technical topic or decision should usually be ignored whereas long explanations that go into details are at least worth something if there aren't obvious falsehoods in it.
Well, I may of made an error in my poke. I more meant is that it's not recommended for production usage and I would call daily driver systems as production. I will admit fault on that.
As myself I'm currently using FBSD16 for my colocated servers and desktop. I have been bleeding lately.
I wouldn't consider it suitable for servers but I think it's perfectly fine for desktops. I still use Debian stable on my desktop because I prefer keeping out of date packages.
Then Slackware, Mandrake (Mandriva now), dipped my toes into RHEL and OG Fedora (had a Fedora 1 DVD) but eventually settled on Debian and haven't looked back.
Can you help me understand where you are coming from? Is it that you think the benchmark is flawed or overly harsh? Or that you interpret the tone as blaming AI for failing a task that is inherently tricky or poorly specified?
My takeaway was more "maybe AI coding assistants today aren’t yet good at this specific, realistic engineering task"....
In my experience many OTEL libraries are aweful to use and most of the "official" ones are the worst offenders as the are largely codegened. That typically makes them feel clunky to use and they exhibit code patterns that are non-native to the language used, which would an explanation of why AI systems struggle with the benchmark.
I think you would see similar results if tasking an AI to e.g. write GRPC/Protobuf systems using only the builtin/official protobuf codegen languages.
Where I think the benchmark is quite fair is in the solutions. It looks like for each of the languages (at least the ones I'm familiar with), the "better" options were chosen, e.g. using `tracing-opentelemtry` rather than `opentelemetry-sdk` directly in Rust.
However the one-shot nature of the benchmark also isn't that reflective of the actual utility. In my experience, if you have the initial framework setup done in your repo + a handful of examples, they do a great job of applying OTEL tracing to the majority of your project.
Where I work we are looking at a lot of our documentation and implementations where AI has a hard time when doing it.
This almost always correlates with customers having similar issues in getting things working.
This has lead us to rewrite a lot of documentation to be more consistent and clear. In addition we set out series of examples from simple to complex. This shows as less tickets later, and more complex implementations being setup by customers without the need for support.
happens to me a lot when i am on shrooms too
reply