Parce que
LeFantome
joined 2 years ago
I think it is a smart play.
They are getting lots of press in the places where the people that care go.
On their website, they focus on the benefits of their platform for customers. From that perspective, the new COSMIC is just a refinement on what they were shipping before.
And this is just the beginning. COSMIC itself is still fairly basic. And the “new” Pop!OS is based on an LTS base that is already 2 years old. None of that is a problem but it is not a hand they want to overplay.
They may actually make a bigger deal about the benefits when 26.04 ships. Things will be a bit more “industry leading” by then.
Ya, it seems odd to be releasing a 24.04 in 25.12 for sure. That said, 24.04 is still the current LTS and so it is the version we would be on now if they had released earlier (even a year ago).
They plan on releasing a 26.04 LTS as well. So, Pop!OS is not lagging. It just feels strange now.
As I said in another comment, critical parts of Pop!OS 24.04 are also quite up to date including the kernel, Mesa, NVIDIA drivers, and of course COSMIC itself.
Moving forward, I expect the versions of COSMIC in 24.04 and 26.04 to be the same. If I was them, I would even consider syncing Mesa between the two. It will make support and testing so much easier and they are already shipping a newer Mesa in 24.04 anyway.
Fair enough.
But before you scare anybody off, it is worth pointing out that Pop!OS 24.04 is quite up-to-date in those areas.
- kernel 6.17
- Mesa 25.1
- NVIDIA 580 drivers
- current Wayland (COSMIC)
Those are what is going to drive your GPU and Wayland experience and they are about the same as you get in Kubuntu 25.10
A lot of the 24.04 packages will be older for sure but it is not fair to compare COSMIC in Pop!OS to the old KDE version you would have been using on Ubuntu 24.04 (Kubuntu).
And I expect Pop!OS 24.04 LTS to see steady COSMIC updates on the road to 26.04. It would kind of shock me if they do not harmonize the desktops between those two releases.
I find it pretty solid if a bit bare bones. With the basics in place, it should improve fairly quickly.
They are planning a 26.04 LTS release. We will see what it looks like by then.
When a Canadian goes to the US, the collective IQ of both countries goes up.
The next chance to do something will be in November.
If anything less comes out of that than a Dem House and Senate that immediately moves to impeach both the President and at least one SCOTUS judge, the entire country is a permanent write-off for me.
2 out of 3 Americans did not vote against Donald Trump. That is a high enough percentage that I wasting time trying to figure out what third to feel sorry for.
PP ought to be a shoe-in at the Leadership convention.
It is not about how well it works. They can certainly disagree about that.
The problem was that they clearly have no idea how it works and “warned” people about a bunch of things that will not happen.
If posted a comment about Linux and told you not to use it because it would cause your hardware to disintegrate, the issue is not my level of Linux advocacy. The problem would be that I am peddling actual falsehoods.
This is a dramatically misinformed comment. You have clearly never used Distrobox.
-
The first bullet is the worst. This is like saying that Flatpak will break your host. It will not break the host OS. Learn what a container is.
-
It is not tools on top of tools. It is tools in a container that you can optionally create entries for in your app menu. Apps in the container execute directly on the host kernel in the container. If you are in a shell, you are either in the container or out and the full environment will reflect that.
-
It is not for “new users”. That said it dramatically simplifies many intermediate ones and make some advanced things possible. Once again though, there are no “stacks on stacks”.
-
The exact opposite. Containers do not change the host. If something goes wrong in the container, you can purge it completely without impacting the host.
In fact, one of my core uses for Distrobox is to keep my host system clean. I use Distrobox to setup dev environments, to try out new software, to work around Distro constraints, and a host of other tasks. It is great.
There seems to be quite a fundamental misunderstanding of what a container is. Everything runs directly on the host kernel. Apps outside the container run on the host OS and do not interact with the container. Apps in the container run in the environment inside the container. The only interaction apps in the container have with the host is the kernel, the filesystem, and via servers like Wayland, Pipewire, and XDG portals.
The entire point of a container is to create an application environment that looks exactly like an application expects regardless of system it is running on. It is clean and consistent between deployment. This is the exact opposite of what the above comment implies.
In practice, it “feels” like apps in Distrobox are running native on the host OS but in fact everything above the kernel is running inside the container and each container is distinct from the others.
The “example” is totally wrong. GCC is either on the host or it is not (from the host). GCC is either in the container or it is not. They do not interfere with each other. They do not even share C libraries. However, the can see the same source files.
Distrobox is absolutely nothing like Homebrew. Homebrew is a package manager. If you are using a package manager in Distrobox, it will be the one intended to work with whatever distro you are running in Distrobox. This is a fantastic illustration of the confusion of ideas at work here.
If you have to compare Dostrobox to something, compare it to Flatpak.
Fair comment.
That said, it just installs in vanilla Arch at this point, which means it would work on EndeavourOS and probably CachyOS as well.
You can of course switch to another DE at any time.
So, you are really not exposing yourself to any risk. If it gets broken or abandoned, just stop using it.
Not that this means you should bother with it. But it is clearly a low risk option to try.