Huh?? I’m using Kubuntu 24.04 right now and didn’t have to jump through these hoops. That’s weird.

I compile them because I want to use them with my system wine, and not with proton. Proton does that stuff for you for steam games. This is for like CAD software that needs accelerated graphics. I could probably use like wine-ge and let GE compile it for me, but I’m not sure they include all the Nvapi/cuda stuff that’s needed for CAD and not gaming. If there’s an easier way to do it, I’d love to hear! Right now I’m using https://github.com/SveSop/nvidia-libs

I’m a developer that’s been using Ubuntu distros for 20 years and never ran into such issues.

If you’re a developer that’s comfortable with desktop software toolchains, that makes sense. (And checkinstall is wonderful for not polluting your system with random unmanaged files). But I came at this knowing like embedded c++ and Python, and there was just a lot of tools I had to learn. Like what make was and how library files are linked/found, etc. And for someone who’s not a developer at all, I imagine that this would be even harder.

I’ve learned a lot, especially because of everyone in this thread

I’m glad!

Re the flatpak issue, what you linked is just saying that flatpak won’t be a default installed program and packages provided by flatpaks won’t be officially supported by Ubuntu support as of 23.03. I don’t think this effects your use of Ubuntu in any way. If you want to use flatpaks, just install the program. It will still be packaged in the Ubuntu repositories. 23.04 was over a year ago. I still use flatpak without a problem on my kubuntu 24.04 system. It’s just a one time thing to do sudo apt-get install flatpak and maybe a second package for KDE’s flatpak packagekit back end and it’s like canonical never made that decision.

The push of snaps instead of debs is a bit more concerning because it removes the deb as an option in the official repositories. But as of right now I think only Mozilla software has this happening? If your timeline is 5-10 years though, this may be more of an issue depending on how hard canonical pushes snaps and how large their downsides remain

All those patches seem like nice things to have, but are more focused on adding hardware support and working around bugs in software/other people’s implementations. If you have one of the effected GPUs/games/etc, then those patches probably make a huge difference, but I’d guess there won’t be noticeable frame rate differences on most systems. I have not tested this claim though, so maybe something on there makes a big difference. What’s nice is all the packaging stuff they’ve done to make setting things up correctly easily, not necessarily most of the changes themselves. Like on my system I compile dxvk and various wine nvidia libs myself since Ubuntu doesn’t package them. And it’s easy to screw that up/it requires some knowledge of compiling things

Reading your update, I’d still choose whatever distro packages the software you want with the versions/freshness you need. If you’re willing to tweak things, then the performance stuff can be done yourself pretty easily (unless you have broken hardware that isn’t well supported by the mainline kernel), but packaging things/compiling software that isn’t in the repositories is a huge pain. I think this is one of the reasons people choose arch even with its need to stay on top of updates. Is that the AUR means that you don’t have to figure out how to build software that the distribution managers didn’t package. Ubuntu’s PPAs aren’t great (though I don’t have personal arch experience to compare with)

I’m not sure what performance improvements you’re talking about. As far as I’m aware, the difference between distros on performance is extremely minimal. What does matter is how up to date the DE is in the distribution provided package. For example, I wanted some nvidia+Wayland improvements that were only in kwin 6.1, and so I switched from kubuntu to neon in order to get them (and also definitely sacrificed some stability since more broken packages/combinations get pushed to users than in base ubuntu). It’s also possible that the kernel version might matter in some cases, but I haven’t run into this personally.

I think the main differences between distros is how apps are packaged and the defaults provided, and if you’re most comfortable with apt based systems, I’m not sure what benefit there’s going to be to switching (other than the joy in tinkering and learning something new, which can be fun in its own right).

For some users less experienced with linux, the initial effort required to setup Ubuntu for gaming (installing graphics drivers/possibly setting kernel options, etc) might push someone toward a distribution that removes that barrier, but the end state is going to be basically identical to whatever you’ve setup yourself.

The choice between distributions is probably more ‘what do I want the process to getting to my desired end state to be like’ and less ‘how do I want the computer to run’.

Could you post the specific output of the commands that don’t work? It’s almost impossible to help with just ‘It doesn’t work’. Like when ping fails, what’s the error message. Is it a timeout or a resolution failure. What does the resolvectl command I shared show on the laptop. If you enable logging on the DNS server, do you see the requests coming in when you run the commands that don’t work.

I’m sure they’d welcome a pull improving the UX! https://invent.kde.org/network/kdeconnect-kde I think the implementation of the protocol is pretty well isolated from the UI, so pretty radical UI changes should be relatively easy

Yes

Does it resolve correctly from the laptop or the server. What about resolvectl query server.local on the laptop?

Isn’t .local a mdns auto configured domain? Usually I think you are supposed to choose a different domain for your local DNS zone. But that’s probably not the source of the problem?

You definitely use a firewall, but there’s no need for NAT in almost all cases with ipv6. But even with a firewall, p2p becomes easier even if you still have to do firewall hole punching

I’ve setup okular signing and it worked, but I believe it was with a mime certificate tied to my email (and not pgp keys). If you want I can try to figure out exactly what I did to make it work.

Briefly off the top of my head, I believe it was

1. Getting a mime certificate for my email from an authority that provides them. There’s one Italian company that will do this for any email for free.
2. Converting the mime certificate to some other format
3. Importing the certificate to Thunderbird’s (or maybe it was Firefox’s) certificate store (and as a sidequest setting up Thunderbird to sign email with that certificate
4. Telling Okular to use the Thunderbird/Firefox certificate store as the place to find certificates

I can’t remember if there was a way to do this with pgp certificates easily

From looking at the github, I think you don’t need to/want to host this publicly. It doesn’t automatically get and store your information. It’s more a tool for visualizing and cross referencing your takeout/exported data from a variety of tech platforms. It’s just developed as a web app for ease of UI/cross platform/ locally hostable.

Borg append only seems like the way to do this easily

I feel like this really depends on what hardware you have access too. What are you interested in doing?How long are you willing to wait for it to generate, and how good do you want it to be?

You can pull off like 0.5 word per second of one of the mistral models on the CPU with 32GB of RAM. The stabediffusion image models work okay with like 8-16GB of vram.

I’d be surprised if it was significantly less. A comparable 70 billion parameter model from llama requires about 120GB to store. Supposedly the largest current chatgpt goes up to 170 billion parameters, which would take a couple hundred GB to store. There are ways to tradeoff some accuracy in order to save a bunch of space, but you’re not going to get it under tens of GB.

These models really are going through that many Gb of parameters once for every word in the output. GPUs and tensor processors are crazy fast. For comparison, think about how much data a GPU generates for 4k60 video display. Its like 1GB per second. And the recommended memory speed required to generate that image is like 400GB per second. Crazy fast.

Chatgpt is also probably around 50-100GB at most

Second this router! It had the fastest CPU and antenna vs price when I last looked. I run zerotier as a VPN on it an it works great. Plenty of ram and flash for packages too.

I think pacreport --unowned-files might be able to help with that too. Showing you files that aren’t part of any installed package. Probably only does system files though, nothing in /home

I use qdirstat a lot to determine what files are eating all my space

Your ISP knows where you’re going anyway. They don’t need DNS for that. They see all the traffic.