It's Becoming Easier To Develop New Wayland Extensions For Mir

[phoronix]

Phoronix: It's Becoming Easier To Develop New Wayland Extensions For Mir

The Mir 1.2 release was aiming to make it easier to develop Mir servers with custom Wayland extensions easier, but in dog feeding the work, Canonical's long-time Mir developer Alan Griffiths realized some shortcomings in the experience...

http://www.phoronix.com/scan.php?page=news_item&px=Mir-Wayland-Extensions-Smoother

[plonoma]

Alan quickly realized the issues

What were those issues then?
Tell us about them.

[pgoetz]

Can someone please explain what the point of Mir is at this point. Is it a compositor?

[pracedru]

Can someone please explain what the point of Mir is at this point. Is it a compositor?

Yeah. I don't see the point either.

[kbumsik]

Mir for now is a very different thing from it was announced by Canonical at the beginning. It is now a Wayland server/compositor like Sway while it used be a totally different display server from Xorg/Wayland. If I were Mir maintainer I would rename the project considering such a bad reputation...

[pgoetz]

Mir for now is a very different thing from it was announced by Canonical at the beginning. It is now a Wayland server/compositor like Sway

Yeah, I was thinking it had to be something like this. So basically they ditched Google protocol buffers and first layer glue and substituted the Wayland protocol instead. This makes it of interest, as all the current Wayland compositors appear to be shit.

[tildearrow]

Can someone please explain what the point of Mir is at this point. Is it a compositor?

It is both a compositor and a building block for other compositors (like Weston and libweston, or Sway and wlroots).