Your hardware is actually more than capable. The stuttering in browsers is because Qubes no longer allows hardware acceleration. See https://doc.qubes-os.org/en/latest/user/troubleshooting/media-troubleshooting.html

The other big OOTB problem is the default private storage size (i.e. /home/user) for VMs is 2GB, which is insanely small for any sort of interactive use, and needs to be bumped up just to enable a web browser’s cache to reach a reasonable size. Let alone being able to store a useful amount of data to be stored in the VM.

I think when you get past these few things, it’s more smooth sailing.

Pointless additional rambling that can be ignored:

I also at first really liked what QubesOS offers in terms of VM templating, and being able to have multiple AppVM share the same base OS template as more or less immutable (i.e. on next AppVM boot, any changes you made to the system partition in the AppVM are completely reverted to the template’s copy). So one can run multiple AppVMs without wasteful duplication of a unique system partition for each VM. This does mean one must boot up the template VM every time new system packages are needed for the AppVMs based on that template, installing the software, shutting down the template VM, then restarting the AppVM.

Ultimately I ended up just creating a bunch of Standalone VMs, and installed different software into each, rather than trying to come up with an efficient, combined template that had all the software I needed for all my app VMs. Less secure, but just more practical.

Getting used to the clipboard manager is another thing that just takes time, but eventually you get it down. All these things are there for security and I understand why they are there, but it definitely has a huge learning curve.

You can definitely generate real downvotes! This looks like it fills up the entire Fediverse downvote Bingo card.

Big Netbird fan, and they support a fully self hosted version.

Since you know Debian, Devuan is there to try it. Given how almost unavoidable systemd is nowadays I basically try all of Devuan, Artix, Void and the first one that works for the task at hand is the one I stick to :)

The real benefit of PBS is it does not store the backup as just a 1:1 dump of the VM disk, it’s incremental - if the VM you’re backing up for say the 3rd or 4th or 20th time doesn’t have much new or changed data, it’s not a linearly increasing amount of storage on your backup server, the new backup is actually trivially small. This really helps you have longer backup retention if files aren’t actually changing all that often.

I got tired of stressing about whether my backup host might be down when my primary Proxmox host dies, even if they’re in different data centers - well, as far as I know anyway! Now I just use BackBlaze B2 storage for the PBS off-site backup (compatible with Amazon S3), albeit it’s a little finnicky to set up and working 100%. Yes it’s still ‘just someone else’s computer’, but for the ~12 bucks it’s costing me for almost 2TB of backup data I am not stressing about it.

Well aside from the obvious giveaways in the presentation format of the README.md, and .gitignore, the very first question of its FAQ is:

“You use LLMs? 🤯”

Yes. Heavily. Claude and GLM-5.2 wrote a lot of this repo.

“This report draws on a survey conducted online by Wakefield Research on behalf of GitHub from February 26, 2024 through March 18, 2024 among 2,000 non-student, enterprise respondents in the U.S., Brazil, India, and Germany who are not managers and work at companies with 1,000-plus employees.”

I generally want to use smaller FOSS projects, and honestly dont care what tools enterprise companies are making their employees use. But I have college age children and know its being mandated there as well. Though it doesnt mean they have to use it in their own projects.

As was discussed in precious slop threads, these projects are fairly easy to identify when they get posted here, today. Sure in a year that wont be as easy, but its clear from the downvotes thrown at projects since this rule was proposed that many people here just dont want AI-written running in their self hosted systems.

Yes this is needed. Thank you for the proposal here.

I would suggest that this probably needs to be really explicit about any AI involvement, i.e. a minimum if AI is used in any capacity in the coding process, it should require the tag. And ideally an explanation if it was used in other parts of the process.

That last post that came up said they used AI ‘for code review only’. In my mind even that deserves the tag, because these terms are so easy to work around. Someone can ‘code up’ the following:

#include <studio.h> int main(void) { printf(“Program that does X thing”); }

(yes, I know the main arguments are not written correctly. You get the point)

and then have the AI reviewer ‘fix’ their code by doing all the actual work. A strict requirement for this tag, for any AI involvement in the creation of the code seems like the key. The code part is going to be where the security issues crop up, and where it’s really important to know who or what is producing the code you’re about to run on your home server.

I think we’re fairly used to a world where people use templates for their websites, documentation, etc. AI use there bothers me less, but an honest disclaimer saying what the AI did would sure go a long way to reducing the hate comments. I think people will still drive-by downvote, but that can’t (and IMO really shouldn’t be) prevented. But without a rule, people aren’t going to be honest.

The scary part is just how emboldened people feel nowadays to just entirely use an AI for all the coding, documentation, website, and then not even put their name on the project. These to me feel like borderline state actor trojan horses disguised as open-source projects.

Legitimate open source developers can spend years writing code to do something very sinplle but useful, and for them to be drowned out by a bunch of completely AI driven, slop posts really bothers me.

Replaced my third system with artix and could not be happier.

I’m not sure I really understand your use case, but OwnCloud and OpenCloud offer fewer features than Nextcloud and are mostly about file sharing.

Other things you could consider: pydio cells, or Copy party (very simple and more just cloud storage, no syncing).

You are dead on. I hate this, if I want this sort of trite shit I can just ask ChatGPT. Instead people now take AI slop and make articles about it so more AIs can ingest it and spit it back again. One step closer to Dead Internet.

Yes, unfortunately. You might be able to use it as a normal user if you are a member of the ‘disk’ group, but its not consistent and it may still actually need root to do all the loopback setup and so forth that you need to make the Windows install ISO sane.

Windows install ISOs won’t work like that, you must use WoeUSB to write it or Ventoy to boot it

Plus ventoy has the advantage that you can just throw a bunch of ISOs on the drive and it’ll boot them all.

Or Netbird, both work very well and Netbird can be completely self hosted; Tailscale requires Headscale AFAIK.

It depends on how the document is written, but \> stops matching on a period, comma, apostrophe, space, newline, what have you. Word boundary matching is just very handy.

As to why its that set of characters… Honestly I have no idea :) Regexes are just what they are and I assume the special escape made sense to the inventor at least.

/-o\>

Seems to work, albeit you’ll hit later mentions doing it more than once, but yeah word boundary searches are awesome.

Agree with your overall suggestion, just a tip for when the man page doesnt cooperate.

Once I had to start running daemon-reload on changing freaking /etc/fstab I was done with systemd. Give me any other init system at this point.

You may not hear this from others, but I would NOT switch your phone OS and desktop OS at the same time. Assuming you live a digital life there is just TOO MUCH to have to deal with while dealing with tbe hassle of things that don’t make sense, or require you to stop what you’re doing and google the answer.

I love /e/OS and would absolutely recommend it, but there are just shortcomings that slow me down all the time (are you OK with the play store compatibility suddenly stopping and you cant download apps - do you wait and see if it starts working after a few minutes? reboot your phone? Log out and back into your google account? Switch to/from anonymous? This happens more than I think folks would readily admit and the solution is just different each time)

Once you are up and running fot a few weeks with whatever change you want to make first then sure go for the other one.

Best of luck!

From what I’m reading, autoclean would remove any local packages which couldn’t be download (i.e. they are out of date). This would indeed fix the issue, but your packages could still be missing critical udpates.

When possible you should definitely run ‘sudo apt update && sudo apt dist-upgrade’ to get the latest stuff installed. If it says you’re up to date then that’s awesome and no further action is needed.

This is almost always the issue. Apt is not like yum and pacman, you must first run apt update or it will inevitably fall out of sync with the current packages when you try to run install or upgrade, etc.