Mir Developer Hopes Community Will Use It & Add Wayland Compatibility

[pal666]

We get it. You hate.

no, you don't. why should i hate? most engineers are of low level

[pal666]

Maybe some of the developers did voice concerns at the start of the project

all of developers just failed to evaluate wayland. wikipedia quote: When originally announcing Mir, Canonical made various claims about Wayland's input system, which the Wayland developers quickly rebutted. ^{36 37}
both links to phoronix, lol

[pal666]

A reads of the mir stuff says to me "oh look. we don't understand wayland at all, neither do we grok X11 that much either".

i can't resist https://www.phoronix.com/scan.php?pa...tem&px=MTMxNzY

[pal666]

Airlie followed up with, "they barely have anyone competent enough to write a display server, the fact that they are actually quite ignorant of how wayland works makes it even more apparent."

[genstorm]

When originally announcing Mir, Canonical made various claims about Wayland's input system, which the Wayland developers quickly rebutted.

They even ended up adopting Wayland's input protocol, iirc.

[pal666]

They even ended up adopting Wayland's input protocol, iirc.

wayland's libinput

[TheBlackCat]

A different design philosophy. Mir was supposed to be a display server with an API (one that was testable) where as wayland uses a protocol to pass messages.

I don't know how Mir supports got this idea, it is completely wrong. Both Wayland and Mir use a protocol for communication between clients and the server, and both Wayland and Mir provide an API to access that protocol. The difference is that under Wayland the protocol is considered a backwards-compatible, public interface that other projects can use if they want to (although most just use the API), while under Mir the protocol is considered private and not backwards-compatible so other projects are not supposed to use it.

[TheBlackCat]

It's easy enough to judge from a position of ignorance, especially with the quantity of disinformation available on this subject through social media where your ignorance is as valuable as my facts.

That is pretty hypocritical when you have been spreading flat-out wrong information about Wayland, information that ten seconds of googling would have shown you was wrong.

[boxie]

I don't know how Mir supports got this idea, it is completely wrong. Both Wayland and Mir use a protocol for communication between clients and the server, and both Wayland and Mir provide an API to access that protocol. The difference is that under Wayland the protocol is considered a backwards-compatible, public interface that other projects can use if they want to (although most just use the API), while under Mir the protocol is considered private and not backwards-compatible so other projects are not supposed to use it.

Because the inner core of Mir is this, the outer shell might be different though

[littleowl]

if you want to read one person opinion, read the comments in response - here's one i picked
"The TL;DR is that he wants to write a toolkit (or a compositor?) and is pissed that it's not simple. There should be a library that does exactly what he wants but there isn't and every question he has should be answered just the way he thinks about the problem but it isn't. And that's of course Wayland's fault."
If you read down the comments, he reduces his argument to "bad documentation".

Exacly. I use wayland on embedded targets and have to say that the protocol is well designed.