Ubuntu's X.Org Session Support Now Split Into Separate Package

[Serafean]

See what you just did there is you basically, but not in so many words, said, "it works for me, even though I recognize the scenarios it could happen in, so fuck all you"...

I guess projectors don't exist, or docks, or headsets, or second monitors....

EDIT: Your so called 99% is actually more like 30%...

I use multiscreen under Plasma 6.1, and have literally never seen a window come up offscreen.
My screen configuration never changes, so I can't speak for those situations, but that would be a kwin bug to be reported, reproduced, and fixed. I have reported quite a few bugs to KDE, most have already been fixed. So, you know... the process works.

[Siuoq]

SANDBOXED apps don't need to know any of that.

Not all fucking apps are sandboxed.

Ever thought I WANT to be "fingerprinted" by some apps?

Like... my own app (script) which automates my life?

Literally stop fucking talking in app devs' names. You're not their freaking mom.

Then talk to the window manager from your app, problem solved?

[duby229]

I use multiscreen under Plasma 6.1, and have literally never seen a window come up offscreen.
My screen configuration never changes, so I can't speak for those situations, but that would be a kwin bug to be reported, reproduced, and fixed. I have reported quite a few bugs to KDE, most have already been fixed. So, you know... the process works.

I get it and generally I would agree, but even if we filed flawless bug reports for every possible thing that Wayland can't do, do you really think the KDE devs would actually be capable of going through it all and resolve everything?

I don't. I say that's impossible. Maybe in a thousand years...

[varikonniemi]

I get it and generally I would agree, but even if we filed flawless bug reports for every possible thing that Wayland can't do, do you really think the KDE devs would actually be capable of going through it all and resolve everything?

I don't. I say that's impossible. Maybe in a thousand years...

What you think is a bug is a feature more often than not.

A bug report and feature request are different. And it sounds like you would want a feature request that really should be done on wayland level, so you need to win the argument there, or convince KDE that you know better, wayland working group is wrong, and KDE should implement the feature off spec.

I already told you: it's a bug if application can control window placement.

If it is really needed, then it should become a wayland geometry request API that every time asks user "do you want to allow this geometry request to happen?"

*Point is that it should not be standard, and app should work normally even if request api is not available or permission not granted, not hack the whole system by doing things on it's own.

[Serafean]

I get it and generally I would agree, but even if we filed flawless bug reports for every possible thing that Wayland can't do, do you really think the KDE devs would actually be capable of going through it all and resolve everything?

I don't. I say that's impossible. Maybe in a thousand years...

Maybe, but without filing a good report you'll never know.
Quick story: I filed a bug report about screencasting specific windows using firefox, very minor annoyance. KDE devs asked for a mirro bug in mozilla's bugzilla, which I did. I thought nothing of it, until bugzilla updates from mozilla started coming encrypted. Turns out it was a security issue (already fixed, not updated in browsers) in webrtc that I'd stumbled upon.
Do file the good bug reports. At worst you lose 5 minutes of your time, and nothing happens, at best you start a discussion and make the world a better place. Posting about issues only here makes it a guarantee that nothing happens.

[duby229]

What you think is a bug is a feature more often than not.

A bug report and feature request are different. And it sounds like you would want a feature request that really should be done on wayland level, so you need to win the argument there, or convince KDE that you know better, wayland working group is wrong, and KDE should implement the feature off spec.

I already told you: it's a bug if application can control window placement.

If it is really needed, then it should become a wayland geometry request API that every time asks user "do you want to allow this geometry request to happen?"

*Point is that it should not be standard, and app should work normally even if request api is not available or permission not granted, not hack the whole system by doing things on it's own.

The thing is, this particular issue with apps opening off screen, the KDE devs are fully aware of. They implemented a keyboard+mouse click shottcut to drag Windows back to the main screen. Plus there is an unofficial keyboard combo to snap Windows back to the main screen..

It isn't like the KDE devs aren't aware, they are fully aware.

It is NOT a hack to expect applications to be drawn on screen and it is not a hack to expect an application can minimize. It's also not a hack to expect a smooth mouse curser. (Which btw, it's super fucked up that Wayland still can't draw a mouse curser smoothly)

[varikonniemi]

The thing is, this particular issue with apps opening off screen, the KDE devs are fully aware of. They implemented a keyboard+mouse click shottcut to drag Windows back to the main screen. Plus there is an unofficial keyboard combo to snap Windows back to the main screen..

It isn't like the KDE devs aren't aware, they are fully aware.

It is NOT a hack to expect applications to be drawn on screen and it is not a hack to expect an application can minimize. It's also not a hack to expect a smooth mouse curser. (Which btw, it's super fucked up that Wayland still can't draw a mouse curser smoothly)

You are literally talking about a hard to reproduce (probably subtle) bug in window/ output management logic. It might be an issue of the architecture of how Plasma is coded and the idiosyncrasies it causes, but still it's not something anyone has been able to reproduce.

Why focus on some bug that is hard to reproduce, and only affects a vast minority of users?

Personally, if my computer has been on for a week, it sleeps 10 times per day going into all those troublesome power management states. Uses 1-3 displays daily, and then at day 7 it opens a window on a screen that is not active (but is there), it does not hurt me very much.

Just to be clear: in my case the issue is that a connected external monitor is there, is connected, but not powered on. Most of the time Plasma detects this and does not use it for window placement. But in these rare buggy cases, the window might open there.

[duby229]

You are literally talking about a hard to reproduce (probably subtle) bug in window/ output management logic. It might be an issue of the architecture of how Plasma is coded and the idiosyncrasies it causes, but still it's not something anyone has been able to reproduce.

Why focus on some bug that is hard to reproduce, and only affects a vast minority of users?

Personally, if my computer has been on for a week, it sleeps 10 times per day going into all those troublesome power management states. Uses 1-3 displays daily, and then at day 7 it opens a window on a screen that is not there, it does not hurt me very much.

What are talking about? Unplug a display and then start an app.... It is that simple to reproduce.... And it affects anybody that uses a projector or dock or headset or second monitor... I'm guessing at least 70% of all users will come across this at some point, and at least 30% of all users will experience this regularly...

And this is only one example out of at least a hundred others.

[Serafean]

What are talking about? Unplug a display and then start an app.... It is that simple to reproduce.... And it affects anybody that uses a projector or dock or headset or second monitor... I'm guessing at least 70% of all users will come across this at some point, and at least 30% of all users will experience this regularly...

And this is only one example out of at least a hundred others.

You know what, I'm game... I've never seen it, just tried your reproducer recipe, to no avail.
What window placement strategy are you using? (preferences -> window management -> window behaviour -> advanced) I'm using "under mouse"

[varikonniemi]

What are talking about? Unplug a display and then start an app.... It is that simple to reproduce.... And it affects anybody that uses a projector or dock or headset or second monitor... I'm guessing at least 70% of all users will come across this at some point, and at least 30% of all users will experience this regularly...

And this is only one example out of at least a hundred others.

many messages ago you were already told that it's not like that, now 2 users in this thrad alone have told you that, then i told you again it's something really subtle, and still you continue your shit talking. WHY?

IF it was like you try to say, then it would be a reproducer, and it would have been fixed already.

But since no-one to this day has come up with a way to reproduce it, because it is so rare and dependent on interactions no-one has been able to pin down, it's still an open bug.