You miiight be able to recover the array if you’re lucky

I don’t see how this would apply. Having the disks connected externally is the same as having them connected internally, maybe over a different bus/protocol, but the principle is the same. No RAID solution I know of would lose the array on a power outage (AFAIK).

the problem is the target not being able to manage its own interrupt

Honestly I don’t see how interrupt handling would be any different between internally or externally connected devives, except for different buses/protocols handling it differently intrinsicly. Are you absolutely sure this is a thing or are you just speculating?

you have two different states

Maybe I’m too spolied by using ZFS, but again I don’t think this would actually be a problem. But AFAICT you don’t even need a CoW filesystem for that to be not a problem. Every journaling filesystem (e.g. ext4) would solve this by dismissing the newest non-consistent data and restore a working state.

I mean, there are 60-bay 19" expansion units for enterprise storage systems. I doubt these would be a thing if having the drives connected externally was a problem.

Nope, just JBOD enclosures.

I like this one: https://www.qnap.com/en/product/tl-d800c

Or here, the new version: https://www.qnap.com/en/product/tl-d810tc4

Those are just nclosurs though, no built-in RAID functionality.

Please elaborate on point 3.

UPDATE: Even more exciting… Per this comment from a prominent AMD Linux developer, it looks like a full HDMI 2.1 implementation for AMDGPU could be coming!

I wish Zulip was fully E2EE

This only holds true if you’re talking about Debian Stable, there are definitely Sid or Testing based distros for which that doesn’t hold true.
See PikaOS for example.

For rocm, old is bad.

But It’s Months Out-Of-Date

So, par for the course for Ubuntu, no?

Sure, but that doesn’t make any difference in the maintenance effort required.

If I write my own Dockerfiles to tag my own images I also have to run my own update management.
I like to outsource this, so all I have to do is a simple docker compose pull (or in reality, let Dockhand or watchtower handle it).

Never heard of them, but they can fuck right off.

Today, AI can be pointed at an open source codebase and systematically scan it for vulnerabilities.

Well, then do that.

It’s not a perfect solution, but we have to do everything we can to protect our users.

All you do is shipping unaudited software, you cunts.

Unfortunately it only has tag suggestions when there’s letters in the tag field already. I’m probably going to use a lot of tags/categories and I might not be able to remember all of them. In this case I’d need to cycle through all first letters to make sure to not miss one.
I’m not sure that’s what I want to do.

Nothing really meant to use as a daily driver. There’s some sort of WebView implementation in Tauri and apparently a not-really-maintained browser project: https://gitlab.com/verso-browser/verso

https://gosuki.net/
This one?

Looks like it doesn’t come with docker and seems like you don’t get access to the full source on the free version?

https://mozilla-services.github.io/syncstorage-rs/

This still relies on Mozilla’s login servers, so not fully selfhosted unfortunately.

It seems like you can selfhost the whole FxA stack in theory, but I haven’t found much documentation for it.

Cool tool, but not really what has been asked, is it?

Little Snitch has nothing to do with ad blocking. I don’t know what you’re talking about.

Alternatives:

• https://github.com/safing/portmaster
• https://github.com/evilsocket/opensnitch

Ugh