Reasons For Losing Motivation In Wayland
Phoronix: Reasons For Losing Motivation In Wayland
While many are super excited about Wayland and the thought of X11 finally going away in the coming years, some who have been enthusiastic about Wayland/Weston are starting to lose interest. Here's the reasons by one Wayland enthusiast for losing motivation in the project...
This is why you should have fully tested code whit everything developed in branches witch needs approval of both the build bot and a team member to merge back to master.
On this plane Mir is superior over Wayland, to bad Wayland isn't developed this way.
As noted in the article, the problem is with Cairo and Mesa. Mir also uses these, so Mir users would run into the same issues.
Originally Posted by Pajn
This migration X -> Wayland is taking ages and I wonder if it will append before 2016.
One of the downsides I see is that there appears to be as much developer interest in Wayland as there is in X.
Well, there is XWayland too.
Originally Posted by chenxiaolong
But I agree Cairo and Mesa should also be fully unit tested.
Personally I'm losing interest because it only works on the open drivers. This wouldn't be so much of a problem if they had PM functionality since the performance is already good enough in the open drivers, but without PM the open drivers are pretty much useless for everything except starting house fires.
Last edited by chris200x9; 06-12-2013 at 01:48 PM.
well one thing i like to clear here, the fact you send a patch doesnt mean it has to be 100% of the time accepted and there are certain times when a patch could "fix an issue" but the real problem should be adressed because that patch will fix 1 thing but screwed another more relevant.
mesa in this case is very picky since they don't tend to accept patches that don't solve the entire problem or is in wait list cuz someone is working in a better solution that will be integrated later on and will make this problem/patch irrelevant, so maybe this segfault xwayland bug is more transcedental than a minimal patch to hack it fast and require for optimal use that keith merge the dri3000 branch or example.
in the case of cairo i can't talk since i don't report bugs in that project since long ago[have ages without issues with cairo] but i can say for mesa because since i got my SI card i have it in my bookmark bar reporting issues daily.
beside remember wayland xwayland mesa Xorg drivers Xorg+dri3000 are being developed in parallallel, so is not like they are sitting in their asses watching the roof and Canonical's Mir just leech this changes since they don't contribute anything at all giving the sensation that Mir is getting done faster which is false. make OSS stack license incompatible with Mir from X to Kernel modules and lets see how fast can Mir rewrite all that code themselves
High five. If it won't be using binary drivers (or binary drivers won't have necessary bits for wayland) wayland simply doesn't exist for me. I am using Nvidia hardware and I like to play games - closed drivers have all the features and quality I need and novueau is beyond terrible.
Originally Posted by chris200x9
The problem here seems to be more on 3rd party projects than on Wayland itself.
While it is a pitty that open-source software is constatly subject to this kind of problems, it is the reality, and we should try to further improve the interconections between projects that depend upon another, instead of just giving up.