Middle Click Paste
This blog post is just a personal commentary. I am not involved with the GNOME project, and all I do is use it as my desktop environment.
On January 3rd, a GNOME developer posted a short message:
Starting 2026 strong, by disabling the paste on middle-click abomination.
GNOME merged the linked patch on 9th, while Firefox (imo correctly) closed the bug report, stating that it respects Wayland compositor advertisement.
I have been using the middle click (PRIMARY) clipboard since I have discovered it over a decade ago. I will live with this patch just fine — at first I might enable it on new systems, but eventually I'll get used to its absence and move on.
What I don't like about it is something we (non-GNOME developers) have been seeing for years now. Some of you might know where I'm going already.
I'll quote Jordan first, because he started this wave:
This is an X11ism, originally an xsetting which frequently results is in unexpected behavior when people pressing the middle mouse button.
This is a little known feature and behavior that leads into user confusion when they click the middle mouse button without knowing about its functionality. Most of the time, its also clicked by accident, and its very weird o have the clipboard dumped on such occasions.
Let's decompose the text.
Frequently. Most of the time. — What does that mean, how often is that? Who's to make the call?
Little known feature. — Just as Ctrl+Alt+Fn to switch TTY is.
Unexpected behavior. Without knowing. — Who defines "unexpected"? I'd say "Ctrl-C" terminating running program is also very unexpected, and yet everyone just goes with it. Hiding files starting with a dot is also very confusing and unexpected, coming from Windows world.
There is universal design consensus that this is a bad idea to have it enabled by default. — How can it be universal if it caused a relatively big uproar?
I was ready to let the situation be, but then I stumbled over Victoria's post. She tries to explain the history of the PRIMARY selection, though she's still missing out the technical reasons some don't like it.
The technical details
There's a chance for user confusion caused by the active v. passive method: the unexpected possibility of the clipboard being erased if the source is gone, or the fact that some applications don't stop highlighting the text when another text is selected in different application.
Victoria writes
There is no evidence that any contributor for the GNOME project is interested in outright removing the middle-click paste feature.
I wouldn't be surprised a single bit if some were. But I'm giving them the benefit of the doubt.
What follows is a list of GNOME 3 achievments. You might, my smart reader, alrady have a feeling I might not be fond of all of them.
No desktop icons. — I have got rid of those even before switching to Linux. I don't miss them, but they made the desktop useful; now every user is required to open a menu before starting any work.
No minimize button. — "Use virtual desktops". I guess that might be fair, but I haven't seen this workflow explained anywhere ever. There are some obvious problems with this: one fullscreen application which you minimize every so often to see two half-screen apps side-to-side (in-browser documentation with editor and terminal, anyone?). I guess I'm supposed to switch between workspaces (shortcut for this isn't displayed anywhere, isn't it confusing users?) or alt-tab all the time, fighting window z-index (or whatever the equivalent is for GNOME/Wayland).
Overview instead of the dash. — This one I never understood, because it makes me nauseous, even with animations disabled. I remember which part of the monitor the windows were, and now they are all scattered! It also removes zero screen estate, unlike Victoria claims, since the currently open applications can be displayed on the same bar as the time and applications menu is (thanks God to Dash to Panel/Dash to Dock).
No notification tray. — "No nagware". As if "Discord's red unread badge" wasn't solvable by a introducing a checkbox for "Display application icon instead of app-provided one".
At first, it wasn't clear that Jordan wanted to limit his change to different default; he phrased his message very unanimously and uniequivocally. Whether it was intentional or not I don't know, but we all saw what followed.
It’s apparently okay to stalk, harass, and misrepresent volunteers contributing to free software, if it happens to be the accepted scapegoat.
Of course it isn't, Victoria. I get it is a personal blog, but this sentence is playing it as fair as the texts it complains about.
[I]f the feature is so valuble [...], then it must be worth somebody’s time to keep it alive. If GNOME’s middle-click paste languishes and flounders without love, you’ll only have yourselves to blame.
No. Not everyone is able to learn how a 26 years old desktop environment works, dig through its C, XML, or Javascript stack to identify and fix a problem.
It is simply not true, and everyone knows it. If you want to say what you mean to say, just say it: "if you care, step up and learn all of that".
Side note: In the whole debate, I have seen the Wikipedia article about the X selection linked exactly zero times, just as often as any other documentation explaining the problem space. Not having a common understanding of a problem is certainly a great way of solving it.
It's not surprising to me the GitLab MR got locked immediately after being open; I've been seeing a lot of walled-garden behavior from the GNOME developers. I'm not saying The Register post was phrased the best, or even that its arguments were formed well, but that's what you get if you behave like an authoritan: an opposition equipped the same as you are.
Once you achieve something in software development, you find out that tech is the easy part of it. Talking to people, listening to their opposing opinions, weighing options and coming up with something that is a sane compromise is hard.
If there is something I wish, it is for GNOME developers to realize they aren't just coding their weekend project; they are maintaining a digital equivalent of critical infrastructure, with all positive and negative that comes with it.
With great power comes great responsibility.