So they’ve upgraded to 7.2, which is EOL as of a week ago. And not even the latest 7.2 version, they ship .11 when .15 is the latest.

And generally should be more efficient too. Bash and this unix-style tool calling was not designed for efficiency. Which is fine for most use cases, but should be avoided for tools you regular use.

To be more detailed, every time you call a program using bash, your bash program has to fork into two identical processes. Then all that work of forking is wasting as one of the forks then execs to replace itself with the program you called. And that happens with each tool called.

Using other programs, like Python, can more efficiently exec other programs without the same overhead of forking. And for longer running jobs, Python is much more efficient as it converts itself to bytecode, whereas bash is purely interpreted.

The main downside of Python is its longer time to start. So for simpler jobs, bash can be better.

My opinion of VanillaOS + a dev response to it: https://discuss.tchncs.de/post/66437447/27748310

In another comment, I posted a link to the CRAM presentation that mentioned a chip.

This article also mentions some hardware:

Because CRAM is stored in RAM and treated as RAM, with full cacheline/byte access, it can be accessed in a read-only fashion with little delay; just the cost of hardware-offloaded compression.

I don’t think it would say that if they meant the CPU was doing the compression.

It’s not using the same hardware, this seems to use a special compression chip.

To me this feels more like the benefits of a CPU or GPU upgrade. Things run faster, but not due to software improvements. When people talk about things getting optimized due to the high RAM prices, I think most people are referring to software improvements.

In the presentation, they talk about a compression chip: https://www.youtube.com/live/OPRciCSsdS4?t=2060

Does it count as optimization? For supported hardware, it’s certainly cool. But it’s not making inefficient software better, it’s making better use of the hardware.

None, your current hardware lacks support for this.

A package, meant to run anywhere and be sandboxed.

However, they are only properly sandboxed on Ubuntu. They have some downstream AppArmor patches that haven’t all been accepted. They’re decently sandboxed on other AppArmor distros. Basically no sandboxing on distros that don’t use AppArmor, like Fedora.

Surprisingly, it’s LGPL v2.

It’s almost like Jon Seager just really likes Rust.

It can get run automatically by things like Gnome Software.

They don’t. DHH is a rich guy with rich friends. This is just friends donating to a friend’s project to show support.

And that also means that the CEO of 1Password really has no problems with DHH as a person, those who talk about Omarchy being revolutionary/special or anything like that at best have rose-tinted glasses.

I agree in some respects. If you have something like Gnome with 6 month development cycles, you are kinda encouraged to push features even if they are not ready, otherwise you have to wait 6 months to try again. And as we’ve seen with stuff like triple buffering, it was painful. It was almost ready, missed the cycle, then other Gnome changes were made, then it had to be rebased, then go back to step 1.

With a shorter cycle, missing a cycle isn’t as big of a deal and in theory you can polish it more without it being the end of the world.

But Firefox already had a monthly release cycle. I don’t see how going from a monthly to fortnightly is really that different.

I like Fedora Atomic with bootc for this reason. While Nix language is quite confusing and gets even more confusing when you find online advice that has a different setup than you that introduces compatibilities, bootc is just an OCI containers that I can run bash commands in to get stuff set up.

It’s still not one-to-one. Generally, I find that using bash to set things up makes more things easier than harder compared to Nix. Though Nix does make some stuff simpler.

That being said, I haven’t tried NixOS in a while. I’m sure that with a decent LLMs, today its easier than ever to make a competent NixOS config. I’ve been meaning to try, but haven’t found the time.

Debian voted to not ban AI. Proposal E won:

https://www.debian.org/vote/2026/vote_002#texte

Personally, I wouldn’t recommend Vanilla OS.

• They released v1 based off a non-LTS Ubuntu released and then quickly stopped supporting it while they worked on v2. No security updates.
• They released v2 with no upgrade path from v1. v2 was based on either Debian Testing or Sid. It did not regular receive updates. Right now the website says that last updates were released in January. Presumably they stopped working on it to work on v3.
• I’m not sure if you can upgrade from v2 to v3. A fresh install is probably necessary.
• Mirko founded Vanilla OS but his main focus now seems to be Sinty OS.
• On a petty note, the website makes you opt out of sharing and selling your personal info to third parties.

That’s a lot of churn and issues for such as young project that advertises itself to new users.

Personally, I wouldn’t recommend Vanilla OS.

• They released v1 based off a non-LTS Ubuntu released and then quickly stopped supporting it while they worked on v2. No security updates.
• They released v2 with no upgrade path from v1. v2 was based on either Debian Testing or Sid. It did not regular receive updates. Right now the website says that last updates were released in January. Presumably they stopped working on it to work on v3.
• I’m not sure if you can upgrade from v2 to v3. A fresh install is probably necessary.
• Mirko founded Vanilla OS but his main focus now seems to be Sinty OS.
• On a petty note, the website makes you opt out of sharing and selling your personal info to third parties.

That’s a lot of churn and issues for such as young project that advertises itself to new users.

One annoyance about Python is that I also have a Mac. By default, it comes with no Python. If you install XCode, you get Python 3.9. And of course you can get the latest version from Homebrew.

Granted, I don’t think Amber would help much in this situation either. macOS comes with bash 3.2 and Amber probably targets a newer version that that. So either way I would need to homebrew install newer versions.

Not a DE thing. In /etc/crypttab, you can point to.a keyfile or password to unlock the disk.

So you unlock the OS disk, and on that now readable OS disk are keyfiles and passwords that then unlock the other disks.

Sorry, I should note that 26.3 is not released yet. This is a “snapshot”, aka a development build. The full version of 26.3 will be released in a couple of months.

I updated the title to reflect that 26.3 is not released yet.