Yeah Stalwart seems to have a lot of momentum, I’ll probably be setting up a server with my kubernetes+ceph cluster this month.

Check out NixOS. It can build qcow images from scratch for you to import into proxmox

https://github.com/nix-community/nixos-generators

I have 8 bare-metal servers and I do everything automated with NixOS, I rarely ever access the servers directly.

Here are the nixos configs for my DHCP server and kubernetes servers that you can use as a base.

https://codeberg.org/jlh/h5b/src/branch/main/porygonz

https://codeberg.org/jlh/h5b/src/branch/main/nodes

For what it’s worth, Ive been using Ansible off and on at work for 8 years, and I think it’s pretty outdated and clunky these days, there are much smarter ways to manage workloads such as kubernetes, cloud-init, terraform, and NixOS. If you don’t want to get into Kubernetes then definitely learn NixOS.

Quick google search points out this blog post for tips and tricks for prototyping stuff like game features in Rust: https://corrode.dev/blog/prototyping/

Definitely something that I’m going to try when I have to time to get back into Rust. Probably good advice for most people who are unhappy with Rust. Being attracted by Rust’s unique optimization tools too early on seems like a big beginner trap.

Not here to doubt their decision, they had good reasons to switch.

For the sake of discussion though, would it have been easier though if they had focused more on abstractions with their code architecture? I haven’t done any serious projects in Rust, but those issues with low-level coding and API thrash seem like more of a code architecture problem. Like, that example of a function signature seems like they should have bundled up their paperdoll logic more into a single “PaperdollLoadout” struct and moved that into a separate game logic function separate from the view related code. It’s more code to write, but that’s the up-front cost of strict type checking.

Modding and learning definitely seem like a big barrier for Bevy overall though.

One decision i will question is picking Unity over Godot, though maybe they were still reeling from the learning issues on Bevy.

The legacy jitsi-based rtc used turn, and so does matrixrtc I believe. I haven’t tried it myself, though.

This just sounds like FUD.

What should the Matrix developers have prioritized instead? Chat programs are complicated, especially when you’re making them distributed. You’re comparing an organization and a MSP with like $1M/yr revenue with a soon-to-IPO company with $600M/yr in revenue.

What feature is missing from Matrix that is preventing Discord users from joining?

Matrix uses TURN, and the Matrix protocol is not used for carrying voice/video.

Element is an independent client developed by Element, and not part of the Matrix protocol.

not to mention there are 48 and 64gb dimms out now too that work with basically all alder lake atoms

Yeah, what you’re talking about is called GitOps. Using git as the single source of truth for your infrastructure. I have this set up for my home servers.

https://codeberg.org/jlh/h5b

nodes has NixOS configuration for my 5 kubernetes servers and a script that builds a flash drive for each of them to use as a boot drive (same setup for porygonz, but that’s my dedicated DHCP/DNS/NTP mini server)

mikrotik has a dump of my Mikrotik router config and a script that deploys the config from the git repo.

applications has all my kubernetes config: containers, proxies, load balancers, config files, certificate renewal, databases, clustered raid, etc. It’s all super automated. A pretty typical “operator” container to run in Kubernetes is ArgoCD, which watches a git repo and automatically deploys any changes or desyncs back to the Kubernetes API so it’s always in sync with git. I don’t use any GUI or console commands to deploy or update a container, I just edit git and commit.

The kubernetes cluster runs about 400 containers, most of them just automatic replicas of services for high-availability. Of course there’s always some manual setup steps outside of git, like partitioning drives, joining the nodes to the cluster, writing hardware-specific config, and bootstrapping Argocd to watch git. But overall, my house could burn down tomorrow and I would have everything I need to redeploy using this git repo, the secrets git repo, and my backups of my databases and container /data dirs.

I think Portainer supports doing GitOps on Docker compose? Never used it.

https://docs.portainer.io/user/docker/stacks/add

Argocd is really the gold standard for GitOps though. I highly recommend trying out k3s on a server and running ArgoCD on it, it’s super easy to use.

https://argo-cd.readthedocs.io/en/stable/getting_started/

Kubernetes is definitely different than Docker Compose, and tutorials are usually written for Docker compose.yml, not Kubernetes Deployments, but It’s super powerful and automated. Very hard to crash once you have it running. I don’t think it’s as scary as a lot of people think, and you definitely don’t need more than one server to run it.

nah you’re probably not going to get any benefits from it. The best way to make your setup more maintainable is to start putting your compose/kubernetes configuration in git, if you’re not already.

Ah, no, Kopia uses a shared bucket.

Seems like a good way to do it.

Keep in mind Kopia has some weirdness when it comes to transferring repos between filesystem and S3, so you’d probably want to only keep one repo.
https://kopia.discourse.group/t/exported-s3-storage-backup/3560

Backblaze B2 is a cheap S3 provider. Hetzner storage box is even cheaper, but it doesn’t support S3 natively, so you’re likely to run into issues with the kopia repo compatibility I mentioned.

PHP does actually scale better than something like Lemmy which is written in rust

But sure, you can act like you know more than the Nextcloud devs

Isn’t Opencloud just extended Nextcloud? (Still PHP)

https://nextcloud.com/blog/opencloud-the-digitally-sovereign-workspace-from-hkn-univention-and-nextcloud/

Also, nextcloud core components are written in Rust, the PHP just handles incoming requests.

https://nextcloud.com/blog/nextcloud-faster-than-ever-introducing-files-high-performance-back-end/

Running untrusted Javascript code from the internet without security mitigations is a bad idea. It’s maybe excusable for servers but it still increases the risk of container break out if one of the 100 containers you’re running is attacked.

The ideological brain

In political science, a political ideology is a certain ethical set of ideals, principles, doctrines, myths, or symbols of a social movement, institution, class, or large group that explains how society should work, offering some political and cultural blueprint for a certain social order.

Saying political science and sociology aren’t real science is silly.

Pretty sure the field of political science is not moral relativism

Thinking patterns are reinforced physically, makes sense.

Outside of topical debates, the key difference between conservatism and progressivism is risk taking vs risk avoidance.

I was trying to ask about this at kubecon. As far as I know, Neonephos is more about making software for public cloud services with EU and German funding, and tied to SAP. Public code for public money.

I’m not entirely sure what EuroStack is, but there was an open letter from 100+ companies supporting it and a general idea of investing in European IT and open standards, and divesting from American IT. SAP and LF were not on that letter.
https://berthub.eu/articles/EuroStack_Initiative_Letter_14_March.pdf

Buy used Samsung pm983s on ebay. Super cheap, super fast, and they have power-loss protection. Only downside is that they’re M.2 22110, not m.2 2280. There’s also a bunch of cheap Samsung and hgst u.2 drives on eBay, but you’ll need an adapter.