I’m not sure how well it would work in a self-hosted or server-type context, but GPT4all has built in RAG functionality. There’s also a flatpak in addition to the Windows, Mac and .deb installs.

I’ve had no problems with the Discord flatpak and updates. I use a KDE variant of UBlue and the package manager prompts me to update all flatpaks when I start the computer (whenever there are updates available), which I do with one click.

For a new user, i would think that individually managing package updates is probably not a desired workflow.

Oh, that’s very interesting. I knew about flak in the military context, but never realized it was the same word used in the idiom. The idiom actually makes a lot more sense now.

You’re getting a lot of flack in these comments, but you are absolutely right. All the concerns people have raised about “AI” and the recent wave of machine learning tech are (mostly) valid, but that doesn’t mean AI isn’t incredibly effective in certain use cases. Rather than hating on the technology or ignoring it, the FOSS community should try to find ways of implementing AI that mitigate the problems, while continuing to educate users about the limitations of LLMs, etc.

Weren’t the devs also supposed to be adding the ability to rebind the overview to just the super key (like GNOME)? Is that still in the works?

Yes, I don’t know all of the details, but most of the system config files like fstab and such are modifiable. I automount my NAS by putting a command in fstab.

What instructions or guides do you recommend?

Maybe I’m misunderstanding this, but their FAQ specifically says:

By default, your documents are stored inside the docker volume paperless_media. Docker manages this volume automatically for you.

It also says that documents are removed from the consumption directory, renamed, and put into a folder that you shouldn’t modify.

And that’s my problem with the project. I want to be able to keep my file name and organizational structure.

I experimented with self hosting RSSHub a while back and I do believe you could get past rate limiting on Instagram by configuring the app with your personal Instagram account credentials. I think those without an Instagram account were advised to follow pages via the Pikuki route instead.

You need to use amdgpu-pro, according to the page here. I tried installing the pro drivers on Linux Mint a few weeks ago and still couldn’t get hardware acceleration in Handbrake to work, but it might be worth a shot.

The project linked is a fork of the version on F-Driod. You can download the APK directly from the Github and use it just fine.

Does WhatsApp not work at all on Graphene OS or do you just need to enable Google Play services for it to work? (I do understand why you personally may not want to enable the Play services, but I’m just curious about the potential capabilities).

Thanks for the additional reading and information. Maybe it’s just me, but I feel like I hear about a security vulnerability in “processor microcode” or packages or other software basically every day. As a relatively non-technical user, it’s always very difficult to tell how much these things actually matter for normal users. Flatpaks are incredibly convenient because they “just work” and are easily compatible with immutable distributions. For better or worse, I suspect many people are not going to be dissuaded from using them by hypothetical/abstract security risks.

Can you say more or provide a source on why you shouldn’t use a browser as a Flatpak? Is it just because the sandboxing is potentially weaker?

They need to work on their branding. “Cloud Native” triggers images of subscription services and data mining. But the idea here is that the whole OS and its components are all sort of containerized, so you can just pull pre-configured “cloud” images that are guaranteed to work out of the box to your machine.