Does anyone know of any blogs that center around, or frequently discuss nix? My only subscription in freshRSS is NixOS weekly...and thats long since been abondoned.

It'd be really nice if there was some kind of news or ecosystem summary, so I could know about the newest nix projects as well as official sources.

Hi all, I've currently got a big folder full of a bunch of mishmashed sets of files in nested directories:

• model files downloaded from thangs/thingiverse/etc
• source files for models
• my own SCAD source files for my own modeling projects
• the stl exports of those files
• I don't store the cura projects, but maybe I could
• all the gcode. These get copied over to octoprint to print stuff.

Before I go organizing it all, I wanted to ask how the community sorts their modeling files, or if they use other tools to help keep it all sorted. Any best practices out there?

Additional question: do you just use a good directory structure, or do you use other tools that manage projects for you? I've never used any other CAD software besides OpenSCAD, so maybe those applications do some organization for you. Or maybe you self host something like manyfold

For those of you unaware: https://nushell.sh/

This is by far the most unique shell out there, since it doesn't use raw text as output/input to command line calls, but instead an actual data structure. It's like if every CLI call returned a database table, in a way.

My first is this silicon spatula. It's construction isnt just a silicon tip with wooden handle. Its the red silicon for much more of the handle, which I've felt makes it easier to clean and last longer, since gunk isnt getting wedged between the handle and tip. I like it so much I have two.

The second is probably just a spray bottle with water and dish soap. I clean up messes and the stove and countertops with it, and it's incredibly convenient.

Try it yourself: https://codeburg.org/

The actual site is https://codeberg.org/

Edit: sorry I was making a baseless claim in the title. I dont have proof that Microsoft did it. Here's the whois for what information is actually known.

In the process of updating one of my home manager configurations to 24.11, I got an error about a package being marked as broken. It was something like python3.12-libarcus. Let me tell you, that isn't listed in my configuration, and the stack trace didn't help me...so what do you do?

I discovered a tool called nix-tree: https://github.com/utdemir/nix-tree

It will scan a nix store and allow you to interactively look at all packages and their dependency tree. I decided to look at my current profile with nix-tree ~/.nix-profile/ and look around. (note, I think there's a better way to look at a configuration itself, not just a previous build.) The tool lets me search for dependencies, so I searched for python3.12-libarcus and it brought me to where it is in the tree...

I found the broken package! It was ultimaker cura, slicing software for 3d printers, which it turns out is very outdated in nixpkgs. There's discussions on the github about just wrapping the appimage instead

No surprise I use python, but I've recently started experimenting with polars instead of pandas. I've enjoyed it so far, but Im not sure if the benefits for my team's work will be enough to outweigh the cost of moving from our existing pandas/numpy code over to polars.

I've also started playing with grafana, as a quick dashboarding utility to make some basic visualizations on some live production databases.

I feel like Im dancing around perhaps the most fundamental piece of my operating system everytime I run and install software. Starting services with systemctl and checking logs with journalctl is the extent of my knowledge.

Do you know of good resources or tutorials for learning how systemd works and how to use it to run software on my desktop and servers? Thanks.

Typically when I'm working with photos, I'm doing graphic design type work. I've been using GIMP for this. GIMP is meant for raster graphics editing.

You could also use Inkscape for vector graphics, or Krita for more digital painting type work. But I know all these tools are very powerful and overlap on some use cases.

Do you use any AI-type tools? I use a image upscaler called Upscayl. It works really well and works entirely locally.

Do you know of any tools that can remove backgrounds? This would help with help with the type of graphic design I do.

What other tools do you like to use as it pertains to images?

I just setup my first automated and encrypted backup with borg. It's got me thinking about other chaotic events, and how to respond accordingly. I figured now is a good time to document my infrastructure: hardware, network, a files. This way if something bad happens, like my house burns down, I or a family member has instructions for how to quickly recover data and services. Examples:

1. If my website goes down, with my nextcloud on it, what steps do I need to take to recover the data and restore service?
2. If my harddrive fails, how do I access lost data and reimplement redundancy after a replacement is stood up?
3. If someone important to me needs to access encrypted files, how can that access that data and get access to the passwords/encryption keys?
4. If my phone bricks, how to recover 2fa codes?

So I'd like to have a physical printing copy that tries to cover these emergency scenarios. Of course, I'll have digital copy around as well.

I'm focusing more on digital assets, like

I just got a drawing tablet, and have been wanting some software that would allow me to work out math problems, draw architecture diagrams, etc. I've seen some tools like Excalidraw, which look handy for the sharing capabilities. I also have just used plain krita, which has great feedback for the pen sensitivity, but obviously is overkill for whiteboarding.

Are there any tools you use or recommend for handwriting or picture drawing? Pen or mouse?

I love coffee, but have a surplus of tea bags that I want to experiment with. Does anyone have suggestions for how to get started with tea? Or a simple recipe to use as a baseline? I'm only working with tea bags at this time, which appear to be 2g. I would also love to know how much agitation you are supposed to do with the tea bag itself.

I'm in desparate need of setting up borgmatic for borg backup. I would like to encrypt my backups. (I suppose, an unencrypted backup is better than none in my case, so I should get it done today regardless.)

How do I save those keys? Is there a directory structure I follow? Do you backup the keys as well? Are there keys that I need to write down by hand? Should I use a cloud service like bitwarden secrets manager? Could I host something?

Im ignorant on this matter. The most I've done is add ssh keys to git forges and use ssh-copyid. But I've always been able to access what I need to without keeping those (I login to the web interface.) Can you share with me best practices or what you do to manage non-password secrets?

If given the option, which route do you go? I have services running in both, and I'll often just do whats easier. I dont really notice a different in performance the configuration for containers is simple enough I don't mind it.

I also wish there was a nix function that parsed a docker compose and used it for the oci-container config. Then I could use my existing compose files or the ones I find in docs online.

This idea is inspired by nixos-mailserver. It was so easy to spin up the mailserver after changing some DNS records and putting in some settings. I thought it might be a good idea to do the same for services that need public, decentralized infrastructure to support. Some ideas include

• Tor relay, or exit node
• Encrypted messaging nodes. It looks like SimpleX chat relies on SMP servers to relay communication
• Crypto miners (I know, I know, but you understand how it fits the “public contribution” usecase)
• Search engines like searxng (I currently use a public instance)
• Libredirect services, like proxy clients for social media

Maybe federated services, but those require more than just the software running on the public internet. Those require moderation and long term maintenance. Ideally, the services in this config would be ephemeral.

Does this sound like a good idea? Would you spin one of these up on a $10 VPS? I understand that this is the NixOS community, not necessarily the

This idea is inspired by nixos-mailserver. It was so easy to spin up the mailserver after changing some DNS records and putting in some settings. I thought it might be a good idea to do the same for services that need public, decentralized infrastructure to support. Some ideas include

• Tor relay, or exit node
• Encrypted messaging nodes. It looks like SimpleX chat relies on SMP servers to relay communication
• Crypto miners (I know, I know, but you understand how it fits the "public contribution" usecase)
• Search engines like searxng (I currently use a public instance)
• Libredirect services, like proxy clients for social media

Maybe federated services, but those require more than just the software running on the public internet. Those require moderation and long term maintenance. Ideally, the services in this config would be ephemeral.

Does this sound like a good idea? Would you spin one of these up on a $10 VPS? I understand that this is the NixOS community, not necessarily the p

TabbyML is a self-hosted code assistant. I have been unsuccessful at running it using my Nvidia GPU. There's two ways I've tried to deploy this.

As a docker container

Following the docs, it states I run the following docker run command. Below is what I run, modified to use the correct port:

 undefined

    
docker run -it --gpus all \
  -p 11029:8080 -v $HOME/.tabby:/data \
  tabbyml/tabby serve --model StarCoder-1B --device cuda


  

Then I get the following error:

 undefined

    
docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].


  

So this would appear that I don't have the "nvidia-container-toolkit" installed on my machine. So I go ahead and enable this in nixos:

 undefined

    
hardware.nvidia-container-toolkit.enable = true;


  

To validate that this works, I should be able to run nvidia-smi from within a container. I can run this from the host without issue:

 undefined

    
$ nvidia-smi
Wed Jun  5 08:14:50 2024
+--------------
  

I've been seeing easy ways to store and view tabular data. I'm aware of tools like nocodb, baserow, and mathesar. I'm currently playtesting nocodb. But I wanted to start a discussion on what everyone uses for easily storing tabular data, and if anyone uses these tools.

I've also tried nextcloud tables but it still is very early in development from what I can tell.