SUSE Dropping Mainline Work On Their In-Kernel Bootsplash System

[Serafean]

Average consumers don't think like that. To these people the first impression from booting up the OS is important because if they see strange looking text on their screen they might think their computer is broken or has a virus infection.

Guess what: fuck the average consumer.
They can, and should learn. Do we have a problem with cars coughing a bit before starting? No. Do we have problems with the car's dashboard lighting up like a Christmas tree the moment power is turned on? No.
Why should a "diagnostic during boot" cause problems on computers.
</rant>

[starshipeleven]

Guess what: fuck the average consumer.

Guess what: fuck people that can't understand users should not be bothered unless it's really necessary.

All embedded systems hide their boot console and show a splash screen already. It's what makes a system look polished, and "whole".

Why should a "diagnostic during boot" cause problems on computers.

Looks bad (text interface is not used for users) and is useless?

[Guest]

Guess what: fuck the average consumer.
They can, and should learn.

... and that attitude is the reason why the year of desktop linux never happened.

[starshipeleven]

... and that attitude is the reason why the year of desktop linux never happened.

Uhhh... no. Linux desktop did not happen because of other reasons. Mostly because of the legacy Windows ecosystem.

[timmerk]

... and that attitude is the reason why the year of desktop linux never happened.

Exactly! I was thinking the same thing.

[oiaohm]

Average consumers don't think like that. To these people the first impression from booting up the OS is important because if they see strange looking text on their screen they might think their computer is broken or has a virus infection.

I am sorry to say this in kernel splash screen stuff does not make very much sense to me.

https://www.denx.de/wiki/DULG/LinuxSplashScreen

So boot loader sets splash screen Linux kernel has CONFIG_FB_PRE_INIT_FB so it does not stomp on bootloader set splash screen then if you need a splash screen after that you set it from userspace.

Of course that keep contents of video card set by bootloader and hide Linux output until userspace is ready would be a useful feature to add to KMS as well.

There is more than one way to implement a splash screen to hide that strange looking text from end users.

So FirstPersonBSOD problem can be addressed in many different ways. Some ways having if splashscreen is on or off controlled by the bootloader makes way more sense as well. Bootloader can have detected a failure prior and boot with the splashscreen of so user gets to see what the problem is if the bootloader is in control.

So I don't think SUSE path made any sense so it disappearing into history would be a good thing.

[oiaohm]

Really the idea of the average consumer does not exist as a fixed constant. Anyone saying for the average consumer needs to get a clue quickly.

Most things there are more than one way that it can be implemented. I remember when people wanted fast web server.

https://en.wikipedia.org/wiki/TUX_web_server

So why don't we have TUX web server today because it turned out some kernel modifications and userspace performance with a normal web server could be brought up equal.

Just because a group of consumers is demanding X and developer is attempt to address a group of consumers X feature does not mean that developer is putting that feature in the right place.

So I totally agree when it comes to if something should be implement in the kernel or not you mostly say stuff the group of consumers wanting the feature because their wishes are not valid grounds for it to be implemented in the kernel.. Like with the spashscreen technical merit suggest boot loader and user-space.

This is some of the big trouble average consumer wishes for something does not tell developers where it should be implemented.

Consumer wished for features implemented in the wrong place.
1) Bloat that is duplication of code that does not need to be duplicated.
2) Security issues like the TUX webserver running with the privilege of the kernel was not 100 percent a good thing.
3) Poorer customer experience. Like splashscreen in kernel that may not simply disabled for diagnostics means users could be holding a device with no clue why its broken and people repairing can have a hard time.

[duby229]

Ah crap. Oh well. This sucks.

[jpg44]

I dont think the splash screen is a big issue for useability on Linux, with getting Linux onto more desktops. The main blockers still are as they have always been the lack of compatability with Windows apps (wine is only 50%, which isnt good enough). And also, the problems with getting certain kinds of hardware to work. The big thing for a Linux desktop is to get PC manufacturers to install it by default, and the above two issues are intetwined with this. The distros need to engage app developers to encourage the use of cross platform APIs in order to help improve things.

[FirstPersonBSOD]

I dont think the splash screen is a big issue for useability on Linux, with getting Linux onto more desktops. The main blockers still are as they have always been the lack of compatability with Windows apps (wine is only 50%, which isnt good enough). And also, the problems with getting certain kinds of hardware to work. The big thing for a Linux desktop is to get PC manufacturers to install it by default, and the above two issues are intetwined with this. The distros need to engage app developers to encourage the use of cross platform APIs in order to help improve things.

It makes your system looks more polished to the average people and first impression is really important if you want people to actually try your OS.