And also, you learn to make programs of a given difficulty by making programs of a smaller difficulty first.

Beware rdiff-backup. It certainly does turn rsync (not a backup program) into a backup program.

However, I used rdiff-backup in the past and it can be a bit problematic. If I remember correctly, every "snapshot" you keep in rdiff-backup uses as many inodes as the thing you are backing up. (Because every "file" in the snapshot is either a file or a hard link to an identical version of that file in another snapshot.) So this can be a problem if you store many snapshots of many files.

But it does make rsync a backup solution; a snapshot or a redundant copy is very useful, but it's not a backup.

(OTOH, rsync is still wonderful for large transfers.)

That's what I use too. Coupled with soju it's an easier experience for me. And they are both in Debian 13!

If you are going to run Jellyfin or some other media sharing, the key is if you need to transcode media (recompress because the playback device cannot handle it or not). Likely not, nowadays, but research that. If you need transcoding, research; you might get by with an old CPU, or maybe hardware transcoding support, but it's difficult.

Outside transcoding, for file sharing/streaming, every simultaneous client will require additional horsepower and disk transfer usage. If you are the sole client, then likely you can do with an old CPU. But if you and three people more in your household are going to be using the system at the same time, it might be a bit complex.

One of my home servers is a 4gb of RAM, with a "Intel(R) Celeron(R) CPU G1610T @ 2.30GHz". It's very old and low end, but for file sharing it works quite well, but it rarely has more than a single simultaneous user.

Remember that Google News has RSS feeds! They are very well hidden, but they are there.

However, they are also a bit bad.

I started https://github.com/las-noticias/news-rss to postprocess a bit Google News RSS feeds and also play with categorization. I found spaCy worked well to find "topics", but unfortunately I lost steam.

I haven't tested this, but I would expect there to be ways to do it, esp for VMs if they are not LXC containers.

(I try to automate provisioning as much as possible, so I don't do this kind of stuff often.)

The Incus forum is not huge, but it's friendly, and the authors are quite active.

When I learned Git I think there were not decent tools, so I got used to the command line.

I occasionally use gitk for reviewing my commits- it's nicer to see the files modified and be able to jump back and forth, although I get I could use git log -p instead.

I'm an Emacs user, but I don't use magit (!)

I like some of the graphical tools- some colleagues use Fork and I like it... but as I've already learned the CLI, I don't see the point for me.

I could use learning some jj because it automates some of the most tedious parts of my workflow, but I'm getting too old.

You need two drives for the OS, four for data. Hetzner boxes are cheap with 2 drives, cost multiplies if you add any other.

I was looking at the Proxmox graphs. Now, looking at iostat, r/s measured over 10s hovers between 0 and 0.20, with no visible effect of spamming reload on a Nextcloud URL. If you want me to run any other measurement command, happy to.

Huh, what?

I see in your link that that image has support for KasmVNC, which is great and you could use to make Emacs work...

But the whole point of VS Code is that it can run in a browser and not use a remote desktop solution- which is always going to be a worse experience than a locally-rendered UI.

I kinda expect someone to package Emacs with a JS terminal, or with a browser-friendly frontend, but I'm always very surprised that this does not exist. (It would be pretty cool to have a Git forge that can spawn an Emacs with my configuration on a browser to edit a repository.)