GNOME 3.32 Planning To Retire Application Menus

[polarathene]

disbelief that a difference between vbox and metal could possibly be of any material difference to anyone.

There are a few things that can give a different experience, performance can be an obvious one especially with graphics. When I last tried KDE mid 2016 in vbox I had a subpar experience because the compositor support was nerfed among other issues.

With QEMU and KVM trying to boot Manjaro would fail when using modern UEFI.. their distro works fine with legacy BIOS or UEFI on metal, but for some reason their auto driver detection at boot selects a driver that tries to init with int10 failing to do a proper boot to desktop, it can be avoided by disabling the functionality with kernel boot parameter and relying on X to figure it out correctly. Mentioned it to the devs long ago and it got dismissed :\

KaOS, similarly has virtualization support for vbox and vmware but refused to support equivalent of vbox and vmware tools that improve the guest experience, based on it being too much choice or something...

[polarathene]

Is this the same as the global menu that macOS has and KDE brought out of experimental with 5.14? Gnome taketh, KDE giveth, aha..

But on Windows 10 it is quite nice to right-click on the app icon in the task bar and you get a context menu where you can open recent files or such.

Example in Visual Studio you can get a list of recent solutions.
In Firefox you can open private window.

You can do the same in macOS, it's just a common sense design that users expect to work.

You can also do that on KDE and I assume other DEs?

There is nobody in the Linux space who cares about usability. The UI / DE areas are a mess of random, ill-thought out experiments and the users are on the losing end.

I take it you haven't been following/aware of the usability and productivity initiative going on in KDE community for a while now? https://pointieststick.wordpress.com...-productivity/

KDE has been getting better and better at this with each release :|

[dkasak]

Seems sensible. Gives trolls some content to show their virtual, incel faces over. Everybody wins

[polarathene]

I'm just an Average Joe desktop users, frustrated and stuck with Gnome3.. because I like it!
But frustration and misunderstanding stays.

Doesn't quite sound like you're stuck with it because you "like" it with all that frustation and misunderstanding. Perhaps give another DE a try? A variety of distros will provide the same DE with different default experiences, or you can venture into some customization yourself if you already haven't, might make Gnome still relevant for you?

If you'd like to have more control, this is something KDE promotes as one of it's goals as a DE, it's kind of the opposite with Gnome(which suits people that just want to do work and aren't fussed/affected by Gnome decisions that remove/break stuff.... at least afaik Gnome is the common DE chosen for enterprise distros?).

You can also choose to sacrifice any future improvements I guess and stay on an older version of Gnome that suits you. If you do feel like trying KDE out, I quite enjoy Manjaro, others seem to like Neon(stable LTS Ubuntu w ith latest KDE) or openSUSE(Leap for stable releases, TumbleWeed for rolling). Fedora looks like it's been getting quite good recently too but may be less enjoyable for you.

This is open source, so I guess I'm welcome to contribute to it and improve it.. but seriously can't: no programming capabilities.
So what options Gnome gives me now? Switching to another DE?
Does my use of their project help them somehow?

Can't this kind of project benefit from a large userbase and if not should they improve a way for them to automatically report bugs or unexpected behaviours?
Shouldn't they make use of the large user base to collect metrics to understand their userbase (if they care) and improve their product?

Should they collect input and feature request from the "community" in a public space (or too much exposure to trolls and saboteurs?

You don't need programming abilities to contribute to a project community You can help out other users in that community via online communities like this forum and what I'm doing for you right now, or places like reddit for example, others help with more trouble shooting oriented places like unix stack exchange or superuser.

If you look at the blog posts by Nate Graham for KDE Usability and Productivity( https://pointieststick.wordpress.com...-productivity/ ), he often cites how there are parts of KDE that can be improved on or helped by the community without a need for programming experience, and he's very happy to help onboard you so you're comfortable and not lost figuring out how to do it alone, we have a very friendly community over at KDE reddit.com/r/kde is quite nice and welcoming, feel free to visit! As for the types of things you could do, it can be as simple as fixing/improving some text for a translation or adding some for a missing tooltip(it's not hard to do these small things and you can get guidance) or changing/adding an appropriate icon where it needs to be.

Your use of a project/DE can help them yes. You can be a part of the community and help others, make others aware of it, you may indirectly introduce others that are able to help in other ways. In addition you can share your voice in the community on topics/discussions like this one or elsewhere which may/may not make a difference(it seems to within KDE at least).

You can learn to file bug reports or feature requests. This isn't difficult to do and most projects should support it, usually through a bug tracker like bugzilla(KDE, I think they also use Phabricator for some of this, but that is more oriented to developers?) or Gitlab/Github issues(Gnome has moved from bugzilla to Gitlab afaik), it's true that this could perhaps be improved a bit for the regular user with some local app to streamline the process I guess. KDE is quite receptive to feature requests, especially if they're small improvements to existing software that make sense.

Sometimes you'll see a community reached out to for their feedback and metrics like RedHat has been doing lately with their work towards improving power management, often though the Linux community seems a bit averse to data collection / telemetry and within good reason for privacy. If a project is interested in this type of thing, they usually offer a way to enable it.

Or shouldn't they find a way so the Average-Fan-Boy Joe could eventually sponsor a feature request, like a bounty to collect some revenue/resources? (so many users are happy to pay for open source apps/services!)

BountySource and other sites can cater to this for you. You can link to an existing bug report or feature request and contribute some money towards seeing it resolved properly. Others can join in with their own contribution to make it more attractive to someone with the time to spend on it. It gets resolved and then if those who offered money are happy they release it to the developer that resolved it. You just need to spread the word and get support from others in your community to support that bounty.

Others will accept things like Patreon where you can contribute an amount of money monthly to fund a certain community member to do their thing, rather than a specific task like bounties cater to. Examples for this would be Nate Graham's efforts to make KDE a great experience for users as he communicates with other KDE developers and tries to push/steer efforts towards what users voice as important to see as features or bug fixes, he also provides his weekly posts informing the userbase of what hard working community developers have been doing. Another is Geoffrey McRae and his LookingGlass project, which is for users interested in a particular virtualization experience, he also has contributed bug fixes to that virtualization community that had been causing grief for some time.

Shouldn't they consider this very particular moment with the exposure of being the default desktop on Ubuntu to gain more momentum and push forward for a little revolution of the project (and inside the project)?

They seem more focused on other demographics I think? Once nvidia susses out their shit with Wayland support, such that it becomes viable to use on other mainstream distros like KDE, and if Gnome continues it's current trend/reputation that I've seen(not too positive), you may find Ubuntu makes a switch again at a later point. What sort of things are you thinking Gnome could do though to improve here?

They mostly seem to be making perf improvements, and focusing on being more minimal with less flexibility(which reduces bugs and lets development time go elsewhere I guess), along with their different take on UI from other DEs, GTK4 also seems like it'll be interesting once that arrives, so you could say they're more in a transitional period and forward thinking about where Gnome is to be positioned in the future already, it's just not neccessarily what you'd want to deal with(presently at least).

[NateHubbard]

an awesome feature like the global menu

I've always thought that the global menu on Apple devices was terrible. They tried to get rid of it themselves when OS X was in Beta, but the users wanted things the old way.
It feels really old too. Like 1980s old.

[brrrrttttt]

But on Windows 10 it is quite nice to right-click on the app icon in the task bar and you get a context menu where you can open recent files or such.

Example in Visual Studio you can get a list of recent solutions.
In Firefox you can open private window.

Am I missing something? Gnome 3.30.1

[brrrrttttt]

We gave up on gnome when we qualified it in a vbox session with users, then when we rolled it out on a couple real boxes, we got an entirely different experience. When we went to the support forums to find out what was up, we were met with disbelief that a difference between vbox and metal could possibly be of any material difference to anyone. Apparently the unnamed distro was deciding on old and new gnome depending on virtualization or not.

Having said that, I really liked gnome-terminal's default colors and fonts in the recent Ubuntu that I tried (vs my regular Tumbleweed+KDE).

That sounds like a distro issue, not a Gnome issue?

[brrrrttttt]

I liked the idea when I came to Linux back on 2012. However, it never worked properly. They should've moved everything to this menu, or made a global menu macOS-like. If none of these, it's a valid idea to get rid of.

GNOME improving more and more, and leaving its concurrency far behind.

This. It works really well on macOS, always being in exactly the same place regardless of what you do with the application's windows.

[hrkristian]

sorry but gnome fails in just about every possible way.
the usability is getting worse in every new version in a way that you can't even imagine anymore

This just in.

Lonely man seeks recognition from assumed peers.
(Fails, because all he knows is sour whining.)

[Delgarde]

We gave up on gnome when we qualified it in a vbox session with users, then when we rolled it out on a couple real boxes, we got an entirely different experience. When we went to the support forums to find out what was up, we were met with disbelief that a difference between vbox and metal could possibly be of any material difference to anyone. Apparently the unnamed distro was deciding on old and new gnome depending on virtualization or not.

How long ago was that? Very early in the Gnome 3 cycle, I think that was default behaviour - if the hardware wasn't up to scratch for running a composited desktop, they'd fall back to something closer to the old Gnome 2... and I can easily see a VM triggering that, given the usual issues around hardware acceleration...