My complaints are generic to human-computer interfaces. Providing small icons in close proximity is poor design per [1]. Providing icons without text is also poor design, as summarised at [2]. We can argue more about how detailed labels should be--as I would equally complain about Gnome's context labels too. For example, "Open with..." is more useful as a menu with maximum 5-10 other applications to choose from. For example, will "Compress..." replace/remove my original file if I click on this action, or just create a compressed archive alongside the original?

[1] https://en.wikipedia.org/wiki/Fitts%27s_law

[2] https://www.nngroup.com/articles/icon-usability/

Why would I as a Linux user want windows users to switch? I don't. If many of them do, Linux will become more like windows to appease them, and also because there's more money in the masses.

>your complaints boil down to "I don't like it because I'm not familiar with it" which is not valid critique

dhx cited Fitts's law, and you answered with "you're just not used to it" and a pompous, hand-waving non sequitur about traffic signs you pulled out of your reactive ass.

Fitts's law doesn't give a shit what you're used to. The time to hit a target goes up with its distance and down with its size. A row of tiny unlabeled icons crammed together is slow and easy to miss on the thousandth click, not just the first. Getting used to them teaches you what they mean. It doesn't make them any bigger.

Paul Fitts published it in 1954. Card, English and Burr used it at Xerox PARC in 1978 to show why the mouse beat the joystick and the cursor keys. It's the first thing anyone who designs pointing interfaces should know. If you've never heard of it, you have no business telling anyone what counts as valid critique of a menu. Neither do the people at Microsoft who shipped that row of icons, whether they never heard of it or heard of it and ignored it.

As for traffic signs: they're standardized by international treaty, you have to pass a test on them before you're allowed to drive, and the European stop sign still says STOP. Show me the treaty and the driving test for the Windows 11 context menu icons.

Familiarity is something a design has to earn, not something users owe it. Pie menus are self-revealing: they lead, follow, and get out of the way:

1) A novice clicks, sees the labels, and the menu leads them to the item.

2) An intermediate starts moving before reading, and the menu follows, highlighting the slice to confirm the direction.

3) An expert flicks in the direction without waiting, and the menu gets out of the way and never even draws. It's the same motion every time, so using it is learning it through rehearsal.

Every item is the same short distance from the click, and every target is a whole wedge out to the edge of the screen.

Jack Callahan, Mark Weiser, Ben Shneiderman and I measured it at Maryland in 1988: pie menus were significantly faster than linear menus with fewer errors, even for novices.

An Empirical Comparison of Pie vs. Linear Menus, CHI'88:

https://donhopkins.medium.com/an-empirical-comparison-of-pie...

I've been building them ever since, in shipping open source and commercial products and reusable components: an X10 window manager (the one we ran that experiment on), NeWS, SimCity in both NeWS/HyperLook and X11/Tcl/Tk, ScriptX, The Sims, ActiveX, a personalizable Palm TV guide and universal remote you could stroke with one finger instead of using the pen, and numerous implementations in JavaScript for web browsers.

The earliest known pie menus were in PIXIE, at Cambridge in 1969, by Neil Wiseman, Heinz Lemke and John Hiles: radial buttons that traveled with the light pen's tracking cross on a PDP-7. Its 1972 program runs again in the browser, so you can try them yourself.

Pie Menus: A 30 Year Retrospective, covering all of these:

https://donhopkins.medium.com/pie-menus-936fed383ff1

X10 "uwm" window manager with pie menus, scriptable in Forth (1987):

https://donhopkins.com/home/archive/piemenu/uwm/

NeWS pie menus in PostScript:

http://www.donhopkins.com/home/code/piemenu.ps.txt

Tk pie menus from multiplayer SimCity for X11 (1992):

https://github.com/SimHacker/micropolis/blob/master/micropol...

ScriptX pie menus (Kaleida Labs):

https://donhopkins.com/home/archive/scriptx/piemenu.sx

The Sims pie menus:

https://donhopkins.medium.com/the-sims-pie-menus-49ca02a74da

ActiveX pie menus:

https://www.donhopkins.com/home/catalog/piemenus/ActiveXPieM...

ConnectedTV, the Palm TV guide and remote with finger pies:

http://www.pencomputing.com/palm/Pen44/connectedTV.html

jQuery pie menus:

https://github.com/SimHacker/jquery-pie

PIXIE, running on an emulated PDP-7 and Type 340 display (the mouse is the light pen):

https://hyperties.org/cabinet/applications/pixie-symelec/

PIXIE and FORTH on PDP-7 Cabinet Emulator with Type 340 Vector Graphics Display:

https://www.youtube.com/watch?v=lo8kdY-5i6c

Why PIXIE matters: the people, the 1969 film, and how the 1972 listing was brought back to life:

https://hyperties.org/databases/pixie/

Flight of the PIXIE, the 1969 film set to Yuja Wang playing Flight of the Bumblebee:

https://www.youtube.com/watch?v=jDrqR9XssJI

And by your own admission, half the Windows 11 menu is an ad for cloud storage. Get as familiar with that as you like. It'll still be there on every click.

Instead of pulling reactionary bullshit out of your ass and calling it user interface design, you should base your beliefs on empirical science and actual measurements in the real world, not irrelevant metaphors. Like the way the midterm elections just resoundingly proved beyond a doubt, once and for all, that Trump really is a stupid idiot.