And what especially galls me is that if you suggest this "skeumorphic" model to OSS devs, they act like you're crazy. Even if it's something simple like "the Indigo Magic Desktop's scalable line art icons that had animated states". Suggesting visual interfaces seems to grossly offend people who develop software.
You say that, but I don't think it's true. (I think you even argue against it later in the comment but I'm not sure??)
I liked some things that looked more physical/tactile.
To be clear, I don't care if my calculator looks like a desk calculator as much as if clickable buttons show they are clickable. Underlining a link might be inelegant, but it is also functional and immediately knowable.
These days I use Linux with both swaywm on a laptop with Arch do-it-yourself Linux, and mainstream Gnome on a Fedora desktop (workstation edition) with almost no fiddling. Both desktop metaphors allow me to organise my digital world very well.
From that point of view the mess of icons on a Windows desktop looks horrible to me. I dread the very thought of using Windows ever again. I dread I’ll be given the locked Windows laptop and forced to use it.
I know it’s not that bad. Especially, if you just need to perform some tasks in some GUI apps, especially if those are Microsoft apps anyway. Still, I have the fear. My Linux familiarity created me a very vulnerable place of comfort and possibly an illusion the things would be that way forever (for me). Which can be true, actually.
I want several, hard, discrete desktops. I want one that I can toss a bunch of graphics files onto that has something kin to "Fences" so I don't have to navigate the file explorer. And with the flick of a button I can move to a different desktop that I've set up for work, one for play and etc... I think Ubuntu used to have something similar, though I wouldn't be surprised if it actually behaves much like Windows and is just a window manager.
I do a lot on the desktop to specifically avoid the file manager because file managers seem to be intrinsically unnavigable. And then there's inconsistencies between applications which is enough to press me ever closer to an aneurysm. Going between the Win10 file explorer and one from Win95 a la ImageJ is awful.
Shit, with a system like that I reckon you could almost get away without a file explorer.
You know how you swipe between pages on a phone desktop? Organize all the logically coherent things together for productivity and badda bing badda boom.
And Fences does that, mostly. It has "pages" and the "Fences" themselves. Implementation is... Passable. And unfortunately Stardock's reputation isn't stellar.
It'd be nice to see a native system for Windows or Linux that replicates it.
I can have a dozen windows spread out all across my 3 monitors on desktop 1 and a different windows placed on Desktop 2. when I switch to desktop 2, it switches the whole view to the other windows.
And now, as I write this and test it out, I realize that you must be talking about the icons on the desktop rather than the open programs? That would make sense, because those icons do stay the same no matter which desktop.
I guess I just never thought about that because I almost never use desktop icons anymore. I just hit the windows key on the keyboard and start typing to find what I want. I actually rarely even see my actual desktop, because that would require me to close everything or minimize everything to see the icons and that is just an extra step.
The ability to switch to a project, and see all my notes, documents, apps for only that project. In a way that doesn't collide/interfere.
maybe have ~/Desktop/project1, and all apps use that to store settings, notes, etc.
could be a group of desktops for a project.
maybe like virtual desktops for a project in columns, but each row is a different a separate project.
The idea would be close to macos with fast user switching, but one user per project.
https://github.com/bumptop/BumpTop/wiki#background
Interesting to see it's up on github
There have been a few cheap looking, janky and slow demos in the ‘oughts and late 90’s, but I’m confident that personal computing is able to do this without significant performance loss, and it’s not such a massive leap from tiled window managers today.
I firmly believe that our current way of managing and interacting with information could benefit from an extra dimension: ragged corners, thickness, persistent placement on a desk, and other attributes that make a paper object much easier to identify, locate and differentiate in real life, compared to a bunch of documents named something something.pdf/docx