[02:02] <duflu> bregma: Are we skipping the hangout?
[02:02] <duflu> camako, bschaefer ^
[02:02] <bschaefer> umm i think so
[02:02] <bschaefer> duflu, i think its just me and you around
[02:03] <duflu> And then more
[16:48] <mariogrip> alan_g: ping
[16:48] <alan_g> mariogrip: ?
[16:55] <mariogrip> alan_g: I saw your post "a-new-hope", and about canonical maybe continuing mir, and at the same time i'm trying to create a wayland platfrom in mir. but i'm then wondering if I should stop that and keep using mir instead of having mir on top of wayland, since what we dont want is to maintain a display server on our own. And do you know if the android
[16:55] <mariogrip> platform would also be maintained?
[16:59] <alan_g> mariogrip: at the moment I don't know.
[17:00] <alan_g> Canonical hasn't had practice at redundancies and information isn't flowing well.
[17:01] <alan_g> In particular I don't know if any of the proposed IOT projects would use the android stack.
[17:02] <ogra_> unlikely (but still possible)
[17:02] <mariogrip> right, i'm also a bit worried that android would not prioritized
[17:03] <ogra_> not only that ... i doubt anyone at canonical will look at hybris in the future
[17:03] <ogra_> so even if the Mir bits keep working, hybris integration will be in possible limbo
[17:04] <alan_g> mariogrip: sorry, I gotta go just now.
[17:04] <mariogrip> alan_g: no problem
[17:04] <mariogrip> ogra_: well, libhybris we can manage since we are not alone on that, but maintaining our own display server is a bit too much
[17:05] <ogra_> well, the thing is that you need to maintain  the hybris integration of Mir ... neither mir alone or hybris alone
[17:05] <ogra_> mir itself ... at least in form of the mir-kiosk snap will surely persist ... but i dont know if there will even be any development
[17:06] <ogra_> or if it will just go into "maintenance mode" directly
[17:06] <mariogrip> right, that's what im worried about that libhybris will change and mir breaks
[17:06] <ogra_> it does what it is supposed to do ... (display a single standalone kiosk app) ... so not sure there will be any additional investment
[17:07] <mariogrip> that's why i'm trying to create this wayland platfrom in mir, that way we dont have to maintain the part that talks with libhybris
[20:10] <tvoss> mariogrip: libhybris is surprisingly stable. I think you are creating more work for yourself by using wayland as an abstraction layer here
[20:10] <tvoss> mariogrip: because ultimately, you would have to maintain hybris anyway
[20:41] <ogra_> tvoss, well, if the hal stack changes and hybris gets adjustments for that you usually need to adjust Mir as well ...
[20:42] <tvoss> ogra_: sure, but you would need to patch thewayland adaptation layer, too
[20:42] <ogra_> given the amount of maintainers that care for Mir on hybris ...
[20:42] <ogra_> tvoss, upstream will fix all wayland issues
[20:42] <ogra_> since they use wayland
[20:43] <tvoss> ogra_: upstream of what?
[20:43] <ogra_> but nobody will fix the Mir side
[20:43] <ogra_> of hybris
[20:43] <tvoss> ogra_: sure, but the assumption is that the wayland bits would stay untouched and that the abstraction layer in Mir would need changes either
[20:44] <tvoss> ogra_: just saying that introducing an abstraction layer at that level will likely just move the problem
[20:44] <ogra_> indeed
[20:44] <ogra_> but you add another abstraction that probably makes it easier
[20:45] <ogra_> i.e. if the upper side of wayland doesnt change you dont need to adjust Mir ... and the lower side of wayland will be managed by upstream anyway
[20:46] <tvoss> sure, that's a risky bet, though
[20:46] <ogra_> (i'd actually be more concerned about the performance impact ... maintenance wise it could be less work after all )
[20:48] <ogra_> ...but yeah, based on assumptions :)