Hey everyone,
About 10 months ago, this commit introduced the “Edge Barrier” feature in KWin’s Wayland session:
invent.kde(dot)org/plasma/kwin/-/commit/ad13765348403122747b02e7c26539b55a624dd5 (I’m not allowed to post links)
Currently, the default is set at 100px. In practice, this creates a “dead zone” when moving your mouse between screens. Instead of freely moving your cursor across displays, you have to apply extra pressure to break through the barrier. While I understand the idea behind preventing accidental transitions, it ends up feeling more like a bug than a helpful feature—especially if you move your mouse slowly and keep getting stuck.
Other platforms don’t do this everywhere. For example, Windows might offer slight resistance near the top to help with window controls, but it doesn’t enforce it along every edge. Having this friction on all sides just makes the multi-monitor experience feel unnatural.
I propose setting the default “Edge Barrier” value to 0px. This way, the default experience is smooth and intuitive, and if someone actually wants a barrier, they can still turn it back on manually.
What do you think? Is this the most appropriate place to talk about this, or should I open a bug report?
Hi! Forewarning, this is mostly personal opinion about to come here
I think the catch is, the appropriateness of the feature likely depends on how folks use individual screens:
Mostly with loosely floating windows filling part of the screen
Mostly with maximized or tiled windows filling the whole screen
And how they use multiple monitors:
As one combined workspace - e.g. if they had an ultra-wide, they’d use it
As a physical version of multiple workspaces/virtual desktops - a place to contain different tasks
My intuitive assumption*, from seeing folks I’ve worked with who use multiple monitors, is that more folks tend toward the second option in each of those - most windows are tiled or maximized, and the contents of the screens are intended to be discrete. In that way of working, the most common slow mouse movements near a screen edge would be trying to reach a scrollbar, sidebar, titlebar, panel, or toolbar at the edge of your current screen.
To get even more anecdotal… on my wife’s Windows work laptop with an external monitor, she typically moves her glance from one to the other as a sudden jump - as if she’s pressing Meta+Ctrl+Right in her brain. I also see her, decently frequently, need to make multiple mouse motions to get to the scrollbar on an Excel spreadsheet, because she lost track of the cursor as it jumped a few pixels too far to the right on the continuous extended screen (which was about 12 in / 30 cm in actual physical space).
I don’t have a horse in this race, so to speak, as I don’t use multiple monitors myself - but I thought it might be helpful to talk through the case for “on by default”. There’s also a general KDE design principle that if a new feature was useful enough to spend time building, it should usually be on by default: Get Involved/Design/Frequently Discussed Topics - KDE Community Wiki
For folks who dislike the “Edge barrier” - is your use of multiple monitors more as one combined workspace, or is there a different way of looking at multi-screen setups that would be helpful to think through here?
*Telemetry would be needed for less anecdotal perspectives
@johnandmegh in theory that might sound nice, but in practice it feel really bad for multiple-monitor setup. When I google “edge barrier kde reddit”, these are the first results that pop in:
yes, my time is spent with floating windows on one giant workspace (the desktop metaphor is alive and well at my house).
i have my vertical panel with the taskbar on the right edge of the left monitor so that basically splits my workspace in two visually.
which, from your argument, would tend to benefit from the barrier as i try to pick items from the task manager for the system tray, but i’ve never had any issue with that and having it catch there before moving over to the right monitor just threw a wrench in my workflow whenever i crossed that barrier.
That would be my case as well. Generally when I move my mouse to those corners it’s because I want to click the X button, click a hamburger menu, use the scrollbar, or open the clock so I can see the timezone my friends are in, and my windows are always maximized. If I want to switch screens I do a large cursor movement.
I actually loved when this feature came out because until then I was doing a workaround where I’d have my screens in an arrangement like this with specific scaling:
Just so I’d move my mouse easily to the top right or bottom right corners without having the cursor move to the next screen. Especially useful for my trackball as well, since often I just roll it imprecisely to the vicinity of where I want to click and then I make more accurate movements.
Basically to eliminate the annoying thing where you move quickly to the X button but you go slightly to the side and miss it and the cursor ends in the other screen.
It’s 100% fair that your experience, and that of the upvoters on those Reddit links, is that the feature is unwanted for them.
What we don’t know, from either those negative anecdotes or my positive anecdotes, is how many people who don’t post about Linux to social media liked it, hated it, loved it, left Plasma because of it, etc. Ironically, that’s the group that matters most on topics of new “on by default” settings, as they’re the ones least likely to be plugged into release announcements, and least likely to recreationally open the System Settings app and discover the ability to tweak things like this.
I do think this is an absolutely valid discussion topic for this place, and I continue to be curious to hear about how you and others are using their desktops, as I’m sure the actual developers are as well - it would just seem like a few dozen folks on Reddit being on the side of “off by default” shouldn’t be sufficient to render a verdict on that decision.
Perhaps one actionable item would be finding some way to more explicitly or pointedly get folks to review the release announcements (the one that focuses on features that users will interact with, like Plasma 6.1 - KDE Community - not suggesting every user go through the detailed release notes) in the Welcome Center app on new “major” X.X version releases?
I understand that anecdotal feedback isn’t conclusive. However, the fact that we’re seeing multiple discussions—in this thread, on various forums, and on Reddit—where people consistently look for ways to disable or reduce the edge barrier suggests the concern might be more widespread than just a handful of vocal users.
Try a neutral search for “edge barrier kde” (without additional biasing terms like “disable”). On my end, most of the initial results involve users asking how to turn it off or calling it out as an annoyance. While not a formal metric, this does indicate that dissatisfaction emerges naturally in general searches. Additionally, other popular operating systems don’t enforce a universal barrier across all edges by default. I have used many O.S in my life, including my current dual boot with Windows 11, I’ve used Macs, Manjaro with X11 and other flavors of Linux, and I had never seen this behavior before. I thought it was a bug caused by different DPI between my monitors.
I’m not sure what more would be considered “proof.” Polls or formal usability tests could help, but given the consistency of what we’re already seeing, it’s fair to wonder how many more signals we need before acknowledging that the current default isn’t working for a lot of users.
i’ve always found those “new feature added” highlights until you interact with it to be a good way to bring attention to new features or options in a given piece of software.
the UX is intuitive and straightforward with little orange dots as bread crumbs to lead you to the new setting.
when i upgraded from kubuntu 24.04 to 24.10 and this new feature added to my screen edges settings, i would have found it and realized it was on by default, then i could have chosen to just turn it off rather than abandon wayland entirely.
Similar to Herzenschein’s approach, I’ve had my monitors in often awkward ‘housebrick-style’ staggered arrangements to allow me to throw the mouse and trap it in the corners. For decades, Fitts’s Law and multiple displays have been at loggerheads. No more. This edge barrier feature is an absolute gem, an accessibility and usability boon, and hiding it would be a shame. Shoutout to the devs, this is killer.
How would that person know that such a thing is possible now? If it’s set to 0px there’s no sign it exists.
I get it though, it might be mistaken for a bug, so:
Perhaps something like the ‘Highlight Screen Edges and Hot Corners’ desktop effect, could be used to show a ‘glow’ where the cursor crosses and this effect engages? Visual feedback could maybe give a hint that ‘hey, this is on purpose’, so the user thinks, ‘maybe there’s a setting for this feature’, rather than ‘why is my mouse stuck, must be a bug’.
But to be honest, I think that any visual interference could just generate a bunch of posts on reddit asking what is that annoying new effect and how to hide that. I’d want to toggle it off, for sure. I rest my mouse on that barrier a hundred times a day.
I feel like time and people using a search engine (rather than post another dupe on reddit) will best take care of this.
This feature is divisive. It seems these edge barriers on by default are unique to Wayland KDE across all Operating Systems, which can be frustrating and confusing for anyone coming from another OS or even a different compositor like X11.
If it really has to be on by default, then a highlight could be a good UX solution to give feedback to the user that there’s a “barrier” to overcome when moving the mouse across monitors.
This edge barrier highlight could be turned off as well, for those that are used to this feature.
Enable Edge Barrier Highlight (?)
“When enabled, a subtle highlight will show at screen edges to indicate the barrier. Push through the highlight to move the cursor to the adjacent screen.”
