Mirout: Sort Of Like XRandR Reporting For Mir

[phoronix]

Phoronix: Mirout: Sort Of Like XRandR Reporting For Mir

A Canonical developer has introduced a new utility this week called "mirout" for handling display information on Mir...

http://www.phoronix.com/vr.php?view=MTU2NTE

[V10lator]

Does that mean every program that needs "information like the display's connector, power mode, graphics cards, display sizes, and display resolutions." (I think about games, multimedia apps or generally apps that want to avoid drawing a window bigger than the screen) has to support Mirout? That would invalidate the statement that only toolkits have to care about Mir vs Wayland.

[newwen]

Does that mean every program that needs "information like the display's connector, power mode, graphics cards, display sizes, and display resolutions." (I think about games, multimedia apps or generally apps that want to avoid drawing a window bigger than the screen) has to support Mirout? That would invalidate the statement that only toolkits have to care about Mir vs Wayland.

I'm afraid so. The Freedesktop group should push for standardization for such utility interfaces, like XRandR, for new generation display protocols. Otherwise, the Linux desktop stack faces a huge fragmentation problem like never before.

[tuubi]

Does that mean every program that needs "information like the display's connector, power mode, graphics cards, display sizes, and display resolutions." (I think about games, multimedia apps or generally apps that want to avoid drawing a window bigger than the screen) has to support Mirout? That would invalidate the statement that only toolkits have to care about Mir vs Wayland.

I'd assume games would be provided for by SDL and its ilk (so toolkits in this case), but I do share some of your concern. On the other hand, Mir will most likely stay a Canonical-only project, which kinda means it'll simply have to provide the same or very similar interfaces as Wayland for high-level stuff like resolution queries if it wants to be compatible with software designed for the other desktops. I'd bet that won't be hard though. Wayland will most likely provide a rather stable and versatile design for basic stuff like this. It's not a one company / one desktop project after all.

[Pajn]

Does that mean every program that needs "information like the display's connector, power mode, graphics cards, display sizes, and display resolutions." (I think about games, multimedia apps or generally apps that want to avoid drawing a window bigger than the screen) has to support Mirout? That would invalidate the statement that only toolkits have to care about Mir vs Wayland.

Nah. They get this from SDL. And if they don't want SDL they already have problems today, like different DE's and different soundsystems.

[TheBlackCat]

Does that mean every program that needs "information like the display's connector, power mode, graphics cards, display sizes, and display resolutions." (I think about games, multimedia apps or generally apps that want to avoid drawing a window bigger than the screen) has to support Mirout? That would invalidate the statement that only toolkits have to care about Mir vs Wayland.

I would think that any toolkit that deserves the name would handle this sort of stuff itself.

[Delgarde]

I'm afraid so. The Freedesktop group should push for standardization for such utility interfaces, like XRandR, for new generation display protocols. Otherwise, the Linux desktop stack faces a huge fragmentation problem like never before.

Except that the FD group is almost certainly backing Wayland, as the successor to X. The standard will be whatever Wayland does, and if Canonical don't support the same thing in Mir, that's their problem for not going along with what the wider community is doing.

[mrugiero]

Except that the FD group is almost certainly backing Wayland, as the successor to X. The standard will be whatever Wayland does, and if Canonical don't support the same thing in Mir, that's their problem for not going along with what the wider community is doing.

And in any case, the FD group doing any kind of standard doesn't guarantee Canonical will not do their own and ignore the standard anyway.

[curaga]

Completely unrelated to this, but having something be a FDO standard does not mean it's good.