Solved Gnome/GTK-free Desktop

[sidetone]

Why don't you guys dive into LxQt and/or Lumina and help there?

Lumina is heavy. LXDE, associated with LxQT, is also heavy, despite it being called light.

JWM and other window managers are light, and fully functional. These do very well.

 

[Mjölnir]

LXDE: new name reflecting switch Gtk -> Qt: LXQt (or whatever upper/lowercase they choose...)

 

[Mjölnir]

For me personally, I don't dislike MOC because it is not pleasant to use. I dislike MOC because in the past I have seen it reduces the lifespan of the software and makes it a considerable maintenance nightmare.
Unfortunately using another language doesn't really solve that because the bindings underneath many layers still rely on MOC. Something like this: https://github.com/woboq/verdigris would be more ideal (MOC-less C++) but it is currently a little too niche.

and

It's not attractive to me because of the licensing and being based on C++. As you might have noticed i am (for similar reasons) not sold on FLTK either but it actually comes out ahead from my perspective by having the "right" license.

So then you think the world is waiting for a BSD licensed [light] BSDesktop toolkit? [L]BSDTk.org ready to catch up with 25+ years of GUI development experience just by levering modern approaches in software engineering, development techniques & languages & KISS? Well, why not?

 

[sidetone]

So then you think the world is waiting for a BSD licensed [light] BSDesktop toolkit? [L]BSDTk.org ready to catch up with 25+ years of GUI development experience just by levering modern approaches in software engineering, development techniques & languages & KISS? Well, why not?

gtk is +25 years of more bloat than progress.

An Athena variant can do it, and there's no reason it can't be improved to be better. The problem is, when there's something good, GPL slaps their license on it, and BSD loses a lot of control over its direction.

As for a desktop that have BSD like licenses, JWM and i3. i3 for those who like that kind of window manager.

 

[kpedersen]

and

It was a response to your suggestion of using a language binding to avoid MOC and why this is not a viable solution in the long run. It doesn't eliminate the MOC precompiler.

For reference.

If you don't like the moc precompiler, you can use many other modern languages than C++.

 

[ekvz]

So then you think the world is waiting for a BSD licensed [light] BSDesktop toolkit?

I don't really know what the world is waiting for but to me that sure sounds great

[L]BSDTk.org ready to catch up with 25+ years of GUI development experience just by levering modern approaches in software engineering, development techniques & languages & KISS? Well, why not?

Exactly. In any case i see some reasonable potential. While the devil is of course always in the details i still feel like there really is an actual chance for this to yield a decent result. I am almost certain i'll try it at some point. It's just way to tempting (+ i already have some code i could at least partly recycle).

 

[kpedersen]

I think it could be really interesting. In many ways barely any progress has been made in terms of GUI development since the 80's and many of them were written by tiny teams back then too.

If you do decide to take this forward, I look forward to seeing some examples. It may even inspire me to knock out some simple things like a notepad and calculator with it to start with. The world needs more of them!!!

 

[sidetone]

It may even inspire me to knock out some simple things like a notepad and calculator with it to start with.

x11/xcalc is the only desktop utility I use that is from Twm. I use the pad on my keyboard for it, because it's easier to use more accurately than clicking with a mouse.

It's good as it is. A replacement or upgrade would be good though.

 

[Beastie7]

I’m surprised no one here mentioned Hikari. I’d rather FreeBSD work on having a modern display server in base, then work up the stack from there. X.org is just gross to me.

 

[ekvz]

I think it could be really interesting. In many ways barely any progress has been made in terms of GUI development since the 80's and many of them were written by tiny teams back then too.

If you do decide to take this forward, I look forward to seeing some examples. It may even inspire me to knock out some simple things like a notepad and calculator with it to start with. The world needs more of them!!!

To be honest i've already done what i should not do and looked over the existing sources. I can actually see this happening. The code posted above is already part of a library anyways. Just one with a really broad scope (it was basically meant to not stop at brewing coffee but also drink it) and tons of unfinished/experimental stuff in it. It would need to be heavily trimmed down and somewhat reorganized. Would surely take a bit of time but seems doable. I just need to resist the urge to expose to much or redo anything fundamental. Likely not going to happen during the next weeks though as i should basically be working 24/7. Well, maybe i could do a tiny, tiny bit here or there

x11/xcalc is the only desktop utility I use that is from Twm.

Xcalc is great.

Btw while looking for a screenshot of xmmix i stumbled up this page: https://cyber.dabamos.de/unix/x11. It's very much on topic here i guess.

Edit: Haha https://cyber.dabamos.de/unix/x11/#xlennart. If i ever happen to build a desktop environment i will include this as kind of a minesweeper replacement.

 

[sidetone]

Btw while looking for a screenshot of xmmix i stumbled up this page: https://cyber.dabamos.de/unix/x11. It's very much on topic here i guess.

That's a cool website that shows a lot of neat programs.

The (shade) border bar at the top of windows is different, depending on the window manager.

 

[sidetone]

I’m surprised no one here mentioned Hikari. I’d rather FreeBSD work on having a modern display server in base, then work up the stack from there. X.org is just gross to me.

In that case, x11-wm/cage, x11-wm/sway and x11-wm/wayfire are also Wayland compositors (window managers) with BSD and MIT licenses.

• x11-wm/hikari is a stacking type
• Cage is described as kiosk
• Sway is an i3 replacement
• Wayfire is described as 3D, and based on wlroots

x11-toolkits/wlroots is a Wayland toolkit/library with an MIT license.

 

[Mjölnir]

Thx for the link to xlennart, very amusing! I'd like to suggest you invite senenmut, s/he's diving into Motif programming.

 

[kpedersen]

Wayland

I believe that then restricts you to just two fairly large sloppy GUI toolkits. Isn't that what this thread was trying to avoid?

I do have a little bit of interest replacing the ancient parts of Xorg with Wayland but then splicing the useful part (the underlying protocol for Xlib, XCB) on top. I am really just waiting for the Linux community to see the light and do this themselves (Wayland 2.0).
I am convinced they will once they realize going back to a Windows 95 style single-user display system is daft.

And then of course this re-opens you up to thousands of great Window Managers and *portable* toolkits.

Ideal if they manage to come up with something better than the X11 protocol. But I doubt the community / industry even has the skillset or funds for something like that anymore.

 

[ekvz]

I'd like to suggest you invite senenmut, s/he's diving into Motif programming.

Sure once there is at least something that builds i'd love to cooperate. It just needs to be somewhat cleaned up/trimmed down and glued together. The existing code is pretty clean and tidy but also practically 100% undocumented (outside a couple comments here or there urging to FIXME or giving a short explanation for things that just look really off at first glance). I figure as is it would be rather confusing to a beginner.

I believe that then restricts you to just two fairly large sloppy GUI toolkits.

I have little clue about Wayland but from what i read there is a SDL backend for it so even if there is only 2 toolkits right now there might be a 3rd at some point, who knows?

 

[CoTones]

Thx for the link to xlennart, very amusing! I'd like to suggest you invite senenmut, s/he's diving into Motif programming.

OpenBSD also have the opinion on systemd

GitHub - reyk/systemd-openbsd: This is a game: how fast will systemd eat your system?

This is a game: how fast will systemd eat your system? - reyk/systemd-openbsd

github.com

RedRat tried infect tmux too.

 

[a6h]

OpenBSD also have the opinion on systemd

They have to wait till Linus go to a vacation, then immediately patch it to the Linux kernel.

 

[sidetone]

Motif in the ports has GPL or custom licenses. It's better to start from Athena, because of its license, even if Motif is further developed. Make this backward compatible with motif and other Athena applications. There's no reason that an Athena implementation can't be improved to look comparable to gtk2.

Since the toolkit wlroots for Wayland has libxcb compatibility, go from an Athena implementation to xcb, for xorg and to bridge X11 applications to wlroots or any other potential toolkit on Wayland.

Wlroots doesn't have to be used permanently. Whatever it is that potentially replaces wlroots. Let the Athena to x11/xcb implementation be adaptable for potential Wayland toolkits, if desired.

x11-toolkits/libxaw3dxft is Athena with UTF8 support. x11-toolkits/Xaw3d and x11-toolkits/libXaw are also Athena implementations in ports. They should be replaced with the best implementation and be backward compatible with applications that use these plus Motif.

 

[Mjölnir]

I'd be pleased if a software would arise from this forum! Grab a free cloud resource, install some colaboration tool like devel/fossil, www/trac or www/sogo4-activesync and start to throw in your papers/ideas/code fragments/spike solutions...

 

[a6h]

I'd be pleased if a software would arise from this forum!

I'm all in.

 

[Mjölnir]

I stumbled over Codeberg News : "Providers of cloud services for individual use which do not provide direct access to the public, open source software developing platforms, and online market places whose main activity is online retail of physical goods, should not be considered online content sharing service providers within the meaning of this Directive"