System76 Reportedly Developing Their Own Rust-Written Desktop, Not Based On GNOME

[Steffo]

System76 has really talented engineers like Jeremy and mmstick which also develop Redox OS. So normally I'm sceptic if I read somebody wants to reinvent the wheels, but in this case, I'm very curious what they will deliver.

[blackiwid]

Even I am no gnome hater as many other here seem to be, I am not mad about that, while I thought of the Ubuntu move very very badly when they ditched gnome because:

1. They were the most influential distribution at the time, so they had a market dominating position.
2. They did make it more or less incompatible, installing both unity and gnome at the same machine did not work very well, because they used different versions of parts of gnome in unity. (at one point you needed like 3 ppas to get the full gnome 3 experience)
3. From having it as main distribution as a distribution that have like 5-10 variants for every small desktop they had for the majority of the time no gnome version as alternative, so they kind of forced it on the users.
4. They used the time when Gnome 3 was new and people were hesitate like with all new stuff, and then made the switch very fast and radically. which did hurt gnome adoption a lot. So they used a weak point to backstab the gnome project they got huge with before.
5. they don't gave a good reasoning, why they need to fork it, maybe they had good reasons and conflicts but they did not communicate them well.
6. Mir was back then a complete fork, first they made advertisement that wayland is the future then a month later they created their tool that to the user seem to do exactly the same so just for NIH Syndrome they did create it.
7. all the shit they did with spyware, making a big fuzz about systemd by not making any attempts to bring their upstart init system forward, so that it could compete with systemd, if you want to make a competitor don't have the most horrible tooling any init system ever had...
8. all their Contracts you had to sign, their Git clone etc.
So the only critique is a bit what is the selling point for this new gui, is it just the theming that would be a bit thin, it be written in rust also very thin, what could sell me would be if they had better tiling support backed in all the plugins that do that in gnome are very shitty.

Also most other competitors to some degree just want to give users a gnome 2 ish experience, you don't need 20 gnome forks to give them gnome 2 experience... at least unity had some cool things like the hud or the menu at the top...

So if they can make something that really differentiates from all the other desktops I am happy to hear, and if it's not a gnome clone it should be easy to have it installed besides gnome without problems and port it to many distries easily. Without their own Xserver replacement like Mir was.

[stormcrow]

I wonder just how much do they intend to re-invent the wheel?

When the wheel works for them and their customers. I hope they come up with something great. There's a hell of a lot of room for improvement in this space.

I'm also utterly tired to death of the "reinventing the wheel" trope. This is free software, trying new things is what this community is all about. If you don't like that, perhaps it's not for you.

[bple2137]

I'm very interested in this project... but I have it wonder if it can hit the critical mass that KDE and GNOME have.

Of course it can! Ale they've got to do is to make some solid desktop experience

Seuously though, KDE and GNOME both have some issues that stay there for years. Sometimes even something that seems easily fixable require some big rework. They need to deal with huge codebases and 90s designs. System76 won't have these problems as they create this thing from scratch and they can suite it well for modern standards from the ground up

[rmfx]

Nice, something not based on Gtk or Qt, the broken elder and the bloated. Maybe we can have finally something very optimized and clean. Written in one language. Let’s see.

Ps: please do not name your apps with 50 heterogeneous dumb names like the mess kde. More something like deepin with consistency and simplicity.

[skeetre]

There already exists one that performs pretty much better than the others and has many nice eye candy features a la compiz mode. It is called Wayfire check it out!

This looks interesting. I'm definitely going to try it out. Thanks!

[mdedetrich]

From what I can tell, the only thing that's changing is that GTK apps will have to opt-in to using a stylesheet that isn't the platform stylesheet instead of opting out by including a copy of the platform sylesheet.

They're just asking that distro providers don't change the system theme because it can break applications and the end user won't necessarily know what's to blame.

Which is a terrible experience for distros like PopOS since they want to provide their own theme/custom user experience out of the box, thats the whole issue.

Oh boy, yet another minimalistic UI toolkit. These toolkits are simple and elegant for one reason: they do a whole lot less than "real" toolkits like Qt or GTK. And I'm not only talking about fancy features, but inconvenient and complex but important things like internationalization and accessibility. None of these simpler UI toolkits support complex text layout, for instance, and that's something needed all over the world if you look beyond western countries.

Thankfully they are using Rust, which unlike C/C++ has some of these issues (such as internationalization) already figured out, either in crates or in the language itself.

This is the result of having a language that has a defacto package manager, makes it very easy for the community to create high quality libraries that are reused in the ecosystem

Desktops are already written in C or C++, so Rust's speed is a complex thing to discuss. In absolute terms, benchmark contributors seem to have trouble getting it on par with C and C++ but, in real-world terms, Rust tends to be faster because people are willing to leave performance on the table in C or C++ because they don't want to be responsible for debugging and maintaining what would result.



It's very good at letting you teach the compiler to enforce your invariants for you so a bad night's sleep results in a compiler error rather than an insidious bug. For example, because the ownership-and-borrowing system lets it invalidate previous variable bindings, you can implement what's known as the typestate pattern and teach the compiler to enforce correct use of any API that can be represented as a finite state machine. (eg. the Hyper HTTP library makes it a compile-time error to try to set an HTTP header after you've already started streaming the request body.)

Ontop of this, Rust has its own implementation of `Future` which is very important for GUI toolkits since it allows handling of separate events/UI threads in a much saner way

[Mario Junior]

I think the popos developers were very upset that their system completely uninstalled gnome when installing steam from the terminal.

Otherwise, good luck reinventing the wheel.

[Britoid]

I think the popos developers were very upset that their system completely uninstalled gnome when installing steam from the terminal.

Otherwise, good luck reinventing the wheel.

lol what?

Good luck system76, they're about to find out how much effort is required to mantain a desktop, I'm sure the GNOME dev team won't miss their constant toxicity.

[Britoid]

GTK needs to split from GNOME, IMHO.

This fragmentation occur (Solus is thinking about using EFL and Solus something else) because GTK is controled by the GNOME Project. I think that except from KDE nobody should use anything else, but the fact GTK is part of the GNOME project brings more fragmentation than it would be needed.

My suggestion? GTK should to be a freedesktop.org project or an independent project. This way, GTK could be more inclusive and projects can have confidence to use and contribute with the toolkit.

Libadwaita is kinda that split..