KDE, GNOME, Unity, Razor-Qt Developers Met Up

[uid313]

Anyway fdo has a very broad scope. Covering Linux desktops instead of free desktops would be much better for Linux.

What do you mean?

Any standard that requires D-bus is a failed standard.

What is wrong with D-Bus?

[Guest]

Any standard that requires D-bus is a failed standard.

No, it's actually great idea to use D-bus, because it's not available in BSD. It's Linux community that's driving FLOSS and there's no contribution coming from BSD tiny word. Just ask KDE, Gnome or X developers about this or check which bug reports are meaningful. For example: PC-BSD has debugging disabled while even Ubuntu has this turned on (which makes Ubuntu vs BSD benchmarks quite unfair).

[Banu22elos]

lets hope they come to agree on something.

[curaga]

What is wrong with D-Bus?

It's a bloated, slow, and unnecessary daemon. My systems are usually dbus-less.

Re Pawlerson, I don't know about it on BSD. If it's not available there, then that's another good reason against it.

[Teho]

It's a bloated, slow, and unnecessary daemon. My systems are usually dbus-less.

We should have kdbus ("kernel dbus") before the end of the year that makes it fast, lean and daemon-less.

Re Pawlerson, I don't know about it on BSD. If it's not available there, then that's another good reason against it.

D-Bus is available for BSDs. According to Greg K-H some BSD developers are interested in kdbus too.

[Guest]

If it's not available there, then that's another good reason against it.

This one is not the reason against it.

[curaga]

We should have kdbus ("kernel dbus") before the end of the year that makes it fast, lean and daemon-less.

Perhaps. Basing current work on something that may or may not happen in the future is not smart.

[Akka]

Perhaps. Basing current work on something that may or may not happen in the future is not smart.

We can use dbus today. If kernel bus implementation is successful it shouldn't be any problem with a port to it when the dbus inteface is done.

[c117152]

We can use dbus today. If kernel bus implementation is successful it shouldn't be any problem with a port to it when the dbus inteface is done.

The only good thing about dbus is the precedent in introducing other generic client ipc (other than pipes and sockets). Maybe now plumber will get ported and we can finally have small and sane programs again.

[bkor]

It's a bloated, slow, and unnecessary daemon. My systems are usually dbus-less.

That you do not need it for your use case does not make it unneccessary. Most desktops require dbus, systemd uses dbus. Suggesting unneccessary is funny, but does not make it true.