In addition to X.Org and Wayland developers criticizing Canonical
on Google+ about the Mir display server
, there was a colorful discussion about this new open-source project on the Wayland IRC channel.
Shared via this forum post
is a copy of the #Wayland
IRC channel that took place with exchanges was Chris Halse Rogers "RAOF", he's the only Canonical employee that participated in the discussion, who works on X for Wayland and is one of the Mir Canonical developers
. Participating in the IRC exchange on the Wayland side were Kristian Høgsberg "krh", David Airlie "airlied", Daniel Stone "daniels", and other Wayland stakeholders.
Chris Halse Rogers, the Canonical employee, was quick to joke around that "It's [Canonical's] turn to pull a systemd!" He admitted that he knew internally about Mir and that's his reason for his "lack of work on the wayland system compositor branches."
Kristian was quick to let the Canonical developer know he doesn't agree with Mir and that "the technical reasons on the Mir page just don't add up." In particular, "everything about input" that the Ubuntu developers say as shortcomings with Wayland. Kristian added, "don't go out and tell the whole world how wayland is broken and has all X's input problems...that's what pisses me off...you can do whatever you want and you don't need my permission but don't piss on wayland in the process."
The vibrant IRC discussion then came about Chris Halse Rogers saying Canonical was unsure if their contributions would be accepted into upstream Wayland, followed by the Canonical developer admitting, "I'm unfamiliar with wayland's input handling." Kristian ended up responding to Rogers saying that he doesn't think Canonical meant to "piss" on Wayland by saying, "so don't go and tell the world it's broken if you don't know what it is." Kristian additionally said, "I'll have fun explaining how that's not the case to everybody for the next few months."
Open-source graphics driver developer Jerome Glisse got in on the conversation by saying, "if you look at mir example it's scary...doesn't seems to have the notion of atomic commit...or frame synchronisation...there is a bunch of usleep in them with bogus value to supposedly do 60fps."
The discussion did turn to technical merits and just not flaming between the polarized parties. However, in the end it seems that all of the "advantages" of Mir seem to be features already implemented in Wayland or could be achieved without touching the core Wayland protocol.
Rogers also said when talking about porting tool-kits is that they will just be looking at porting the "master" state of the tool-kits, i.e. Qt5 and GTK3 but not the older versions. He also said "we seem to like the android input stack" but "I'm not totally familiar with [its input]."
Those are the highlights from the lengthy IRC discussion that happened on Monday about Ubuntu's Mir. At this point I remain very unconvinced that Mir will be in a state for widespread Ubuntu desktop deployments by the time of Ubuntu 14.04 LTS in one year. If there is one thing though that Canonical is becoming wildly successful at is alienating and fragmenting the core upstream Linux communities with their "Not Invented Here" syndrome.
If you didn't yet try building Mir, see Building & Running The Ubuntu Mir Display Server
where I also included a video of Canonical's display demo.