cause they're honestly pretty shit and known to be unreliable.
Can you elaborate here? I've had very few issues with Flatpaks and the documentation is pretty thorough. I'm curious what wider issues it has to make the whole ecosystem "pretty shit" and unreliable.
Most AI models need at least 24 but preferably 32.
Where are you getting this information from? Most models that are less than 16B params will run just fine with less than 24 GB of VRAM. This github discussion thread for open-webui (a frontend for Ollama) has a decent reference for VRAM requirements.
Nah. When I'm using Zed it's typically for Elixir/Erlang and I'm usually run debugging tools outside of Zed in a separate shell. When I'm using iex and/or observer I like to use a full screen terminal on a separate workspace/tab than the editor itself
Depends on what device I'm using. On my tower(s), I'm typically reaching for Rider, Pycharm, or Zed. On my laptop(s) it's pretty much always Helix or Zed. On servers it's vim 100% baby. I've gotten pretty comfortable working with theses tools, so I haven't really needed to look into alternatives at all.
Sounds like the perfect use case for devenv. I use it in a handful of personal projects and it's proved to be very useful when swapping projects especially when they require multiple services (eg. postgres, redis, nginx, etc.)
It can be setup as a flake that you can use with nix develop.
There's options to start services and you can use scripts if you want some easy ways to tear down environments while in the devenv shell.
Hope this helps.
Technically stable, polished gameplay, great aesthetics; but there's no heart. No passion. Everyone involved was just there for the paycheck. It's routine.
This is an almost perfect description of modern cinema as well.
In my work organization, we don't allow pushes from users that have not signed their commits. We also frequently make use of git blame along with git verify-commit. For this reason, we have most new developers at any level create a GPG key and add it to their GitHub profile shortly after they join or organization. We're a medium-sized FinTech organization though, so it's very important we keep track of who is touching what.
That said, I can't see it being all that important to an individual unless they're very security-focused. For me personally, I have multiple yubikeys and one is meant specifically for SSH authentication and GPG operations including signing commits. Since I use NixOS and home-manager, I use the programs.git module to setup automatic signing and key selection. I really haven't touched it at all in years now. It was very "set it and forget it" for me.
First result of a search:
Gitorious was a free and open source web application for hosting collaborative free and open-source software development projects using Git revision control. Although it was freely available to be downloaded and installed, it was written primarily as the basis for the Gitorious shared web hosting service at gitorious.org, until it was acquired by GitLab in 2015.
No particular order to these.
Interesting read. Wish I would've found it years ago when I started my first DevOps gig. The company used AWS and CloudFormation (YAML, not JSON) quite a bit along with Ansible. The things I saw in that hellscape were brutal.
I don't believe this to be the case with Flathub, only the Fedora repo. I'm asking about the wider flatpak ecosystem, not the fedora-specific repo or how it's setup.
I'd agree that seems like a needless hoop at the very least, but my concern is more to do with the growing trend to shit on Flatpaks as an ecosystem, not just this particular instance of Fedora head-assery.
I think it's decent software and has really solid use-cases, far from unreliable shit at least in my own anecdotal experience. But my experience is limited, which was why I asked the OP to elaborate on actual flaws they see with the Flatpak ecosystem.