git LFS might be for you. If the data takes so long to reprocess I think it is fine to check it in (possibly using LFS).

I used to do this, but imho the used language is hardly a useful index. When does it happen that you want to see everything written python? For me that’s never.

Also where do you put multi-language projects? Like, go backend with typescript frontend or whatever.

I push every project I work on right away to my gitea instance. If I expect not to work on something for some time I just delete the local copy.

When I change devices or hit file size limits, I’ll compress and send things to my NAS.

Well, that sounds inconvenient.

Damn right he owes us!! /s

How is being python-based a good thing?

Seems like a year ago he was actually using Linux himself. Wondering what happened that made him feel so butthurt.

Love how the top comment doesn’t even answer the question. But TIL, I guess. If I had known this option existed, I probably would have used it. I’m going to play the “not a native english speaker”-card on this one.

This is the correct way. I wish hetzner had a storage box size between the 1TB and 5TB version though.

ht o you men? You cn typ jut fine ith keybor like tht.

It’s also a good way to never actually getting the ball rolling on a new hobby, and instead obsessively research what the “correct” way of doing xyz is and then be too overwhelmed by all the opinions to actually get started yourself.

I think this somewhat depends on how tech savvy the people you want to give editing access are. If they know how to handle git and write markdown, I’d go with a git repository with (for example) mkdocs and setup CI/CD to automatically deploy to Github Pages. This would be free. If they are more like the typical MS Word andy, I’d go with a self hosted instance of bookstack. You could host it for example on fly. Unfortunately bookstack does not (yet) support sqlite so you’ll also need mariadb, which will make hosting it on fly slightly more expensive (but probably still far below $10), because you’ll need 2 machines in total. One of which you can’t scale to zero. There are probably other cloud providers where its going to be cheaper though.

My setup is simple:

• Pictures: I don’t take many and rarely look at them tbh. So they just sit on my laptops NVMe
• Music: I only ever use cmus for listening to music => Therefore music is also only locally on my laptop, managed with beets
• Movies/Tv Shows: I have jellyfin running on a raspberry Pi 4. For single user use this works fine (even transcoding DVD quality works). For multi user or higher resolution transcoding this probably won’t work.
• Backups: One off-site backup at a cloud storage provider using restic and one backup on a USB hdd I simply plug in every other week.

My recommendation is: Keep it as simple as possible. In the past I created the craziest setups, but it turns out that in every day life I have neither the time nor motivation to maintain that shit.

[…] script something that does this.

Theoretically this pandoc one liner already does it, but depending on the website the layouting is going to be trash.

pandoc -i 'https://the-website-your-rss-items.link.to/' -f html -t epub -o out.epub

Might not be for you if you are not a TUI person, but I like newsboat. I also use it to watch youtube and listen to podcasts (with mpv). For pdf/epub export you can probably script something that does this.

Fourth Wing by Rebecca Yaos. I was expecting a nice fantasy story with dragons and shit. But the romance part of it was just so annoying. “Oh look that dude is so hot…” at every. single. occasion. I could’ve known beforehand that this book is more targeted towards female readers, but sometimes I just like to go to the book store and buy a book based on the blurb. Since then I made the new rule to keep my distance to books that mention TikTok or #BookTok on the cover.

Surprised to see jellyfin here tbh. The docker image needed literally zero configuration to work perfectly for me.

I know this is suppossed to be a good vibes post, but “nobody” is probably a slight underestimation.

Even with an ad blocker it’s insufferable though. Every time a page ask me to sign up for their stupid newsletter I want to punch a hole in my screen.

Well, of course you are right. The problem is that for many people (including me) it is hard to use it in a way that actually brings value, because it is just too easy to spend hours on there without getting anything in return.

I also think that it is highly subjective what can be considered “good” or “bad” content. When it comes to educational content, I also would consider it a waste of time. Sure, if I have a real life problem and the solution happens to be described in a youtube video, there is nothing wrong with watching it. But often times I was just like “Oh, this could be useful at some point in the future” and at the end of the video I could hardly remember what it was about. I also don’t think that “mindless” content is inherently bad. If it helps someone to relax, go for it. I always felt worse afterwards.

Saying youtube would be a big waste of time in general is indeed reductive, but I think for many people it actually is, because it is just not designed to be used to bring you value. The only objective is that you spend as many hours on their platform as possible.

This + DeArrow. DeArrow replaces clickbaity titles and thumbnails with better titles submitted by the community. I wouldn’t ever use youtube without it again. With this setup I don’t even want to watch most videos anymore, which is a good thing, because let’s be real, youtube is a big waste of time.