| |||||||||||||
Moving the goalpostsMoving the goalpostsPosted Mar 4, 2013 23:57 UTC (Mon) by JoeBuck (subscriber, #2330)In reply to: Moving the goalposts by man_ls Parent article: Canonical reveals plans to launch Mir display server (The H) Without Google's commitment, Samsung would take over Android development, with plenty of money to fund it (though they might no longer share with other players). Android has hundreds of millions of paying users. Neither Wayland nor MIR have end users yet. Perhaps MIR has good ideas, but the fragmentation of the quite small free desktop space makes me nervous.
(Log in to post comments)
Moving the goalposts Posted Mar 5, 2013 0:06 UTC (Tue) by man_ls (subscriber, #15091) [Link] Me too, but instead of shouting that the sky is falling, maybe we should be asking more interesting questions: will there be a standard ABI for drivers? Will the protocols be interoperable? Will there be a common API for applications and desktop environments? If we accept that both sides have good ideas and can learn from the other side, then we can wait patiently until Lennart Poettering reinvents the graphics stack with his own solution (preferently integrated into systemd, because where else? :)
Moving the goalposts Posted Mar 5, 2013 2:39 UTC (Tue) by rahvin (subscriber, #16953) [Link]
My questions are more simple. Why couldn't they devote their limited resources to expanding/improving an existing solution? Wayland needs the help and for shuttleworth to run off and spend his limited resources on yet another fork is insanity. If he devoted those same resources to wayland he might be able to get what he wants while improving wayland for everyone.
Moving the goalposts Posted Mar 5, 2013 3:46 UTC (Tue) by drag (subscriber, #31333) [Link]
Probably the same reason they are using upstart. They like what they are doing more then what others are doing.
Moving the goalposts Posted Mar 5, 2013 5:03 UTC (Tue) by raven667 (subscriber, #5198) [Link]
This seems different than with bazaar or upstart, in those cases they saw the need and were early to the party but failure to build a consensus around their design caused them to be left by the wayside as the industry moved on to technically better solutions (git and systemd). In this case they are late to the part, the early one is google with surface finger and the design that will probably win out on quality is Wayland so that doesnt leave much room for Mir. Maybe it will be like Phonon and just wrap the real API, as long as it doesn't get in the way
Moving the goalposts Posted Mar 6, 2013 5:18 UTC (Wed) by Serge (guest, #84957) [Link]
> In this case they are late to the part, the early one is google with surface finger and the design that will probably win out on quality is Wayland so that doesnt leave much room for Mir.
Not late. Currently there's nothing but Xorg in Linux. Yes, there're Wayland builds in some distributions, there's even RebeccaBlackOS livecd with Wayland on it. But most people have never seen it, they don't know how Wayland works or even looks. Even those who've seen it usually say that it's in early development stage, and lacks many features.
So actually there's Xorg and tales about legendary Wayland. :) Now we've just got one more tale about legendary Mir.
Moving the goalposts Posted Mar 5, 2013 5:53 UTC (Tue) by HelloWorld (guest, #56129) [Link]
The stated reason for sticking with upstart is that it took them five releases to make it work somehow and they expect the same kind of breakage from systemd.
> http://ubuntu.5.n6.nabble.com/upstart-beyond-Ubuntu-12-04...
It obviously never occurred to them that making upstart work was hard because it, you know, is upstart.
Moving the goalposts Posted Mar 7, 2013 20:27 UTC (Thu) by JanC_ (guest, #34940) [Link]
From my point of view, it took so long because of lack of manpower. That problem didn't get solved until recently (the last 6 months or so?).
Moving the goalposts Posted Mar 5, 2013 9:08 UTC (Tue) by cladisch (✭ supporter ✭, #50193) [Link]
> then we can wait patiently until Lennart Poettering reinvents the graphics stack with his own solution
There's no need to wait for Lennart; Kristian Høgsberg is quite capable of doing this by himself with Wayland (and already did this with the Linux FireWire stack, which he replaced with Juju).
|
|||||||||||||