Recent Features
| |||||||||||||
Canonical reveals plans to launch Mir display server (The H)Canonical reveals plans to launch Mir display server (The H)Posted Mar 5, 2013 5:06 UTC (Tue) by rsidd (subscriber, #2582)Parent article: Canonical reveals plans to launch Mir display server (The H)
The complaints above are along the lines of
None of which is relevant, I feel. The point is Ubuntu is Shuttleworth's pet project. He became incredibly rich by being at the right place in the right time with the right service (Thawte), and has indulged himself since with a space trip and Ubuntu. And Shuttleworth's not stupid. Very likely the reasons for not using Wayland were bogus and the real reasons are something like "We want to put out a project by 2014 and are not confident Wayland will be ready". They are willing to work to port the most common toolkits, as well as a rootless X server, to Mir. And Ubuntu has been good at implementing their ideas, be it Upstart or Unity or Ubuntu One. Unity is not universally beloved but it has fared better than GNOME 3, or, for that matter, KDE 4. Shuttleworth has a product and a timeframe in mind. He has decided that this is what it will take to get there. Good luck to him. I use Ubuntu but not Unity (it's been i3wm for a couple of months now). Maybe Mir+rootless-X will work for me, maybe it won't, but as long as they keep xorg in the repositories, I don't particularly care what else they do :)
(Log in to post comments)
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 6:28 UTC (Tue) by misc (subscriber, #73730) [Link]
> "We want to put out a project by 2014 and are not confident Wayland will
> be ready".
That would be disturbing to think that, as this mean they would think that collaboration with others slow down projects.
In fact, I could see the point if Wayland didn't exist ( or surface flinger, for that matter ). But since a good part of the work have already been done, starting from scratch to go faster would be rather surprising.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 7:37 UTC (Tue) by jrn (subscriber, #64214) [Link]
> as this mean they would think that collaboration with others slow down projects.
Which is often true as far as I can tell, unless someone involved is very good at release management. Though it can be worth the wait.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 8:57 UTC (Tue) by rahulsundaram (subscriber, #21946) [Link]
Collaboration with existing projects to modify them to suit you needs or atleast having a conversation with them is not really more of a slow down than rewriting the entire thing from scratch especially for a key component that they don't have a lot of in-house expertise to speak of. They didn't even try to talk to Wayland developers before launching this project.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 10:03 UTC (Tue) by anselm (subscriber, #2796) [Link] But since a good part of the work have already been done, starting from scratch to go faster would be rather surprising. I'd guess that the people at Canonical are afraid that the Wayland people will get sidetracked into stuff that is not important to what Canonical needs, and that will hold up the project. For example, suppose the Wayland people made supporting multi-screen setups a priority. This isn't all that relevant to phones or tablets, which are the reason why Canonical is in this game in the first place. The people at Canonical – who are under pressure because they essentially need to catch up with Android etc. in the mobile space – may think that it is faster to hack up something that supports exactly their use case without having to depend on an external project whose focus is different than Canonical's. It's not as if we hadn't seen that sort of thing before, either.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 11:14 UTC (Tue) by airlied (subscriber, #9104) [Link]
The thing is Canonical has to assign engineers to the task, if the upstream engineers aren't focused on their requirements, then their engineers can work in the project to focus it on their requirements.
It's almost like Mark doesn't understand open source development models, and the inability of Canonical to participate in any projects upstream is a great display of this.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 12:36 UTC (Tue) by pboddie (subscriber, #50784) [Link]
From a pure productivity perspective I can understand that the benefits of working with existing projects and code can be outweighed by the benefits of forking projects or having full control of your own projects, but in these latter cases you have to be well-resourced for such efforts to not be ultimately futile.
It's like watching a rider in the Tour de France break away from the pelaton half way through every stage and get caught every time long before the finish. It may make for good entertainment, but it won't win that rider anything.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 16:03 UTC (Tue) by NAR (subscriber, #1313) [Link]
Actually that gains points for the most combative rider classification, points at intermediate sprints, king of the mountains points, TV exposure and the occasional stage win. Actually this strategy makes sense if one does not have the talent to mix it with specialists - "let's try and hope it's my lucky day". If Canonical really can deliver something that works in a couple of months while Wayland gets delayed, Firefox OS crashes and Android stumbles, they might get lucky. However, I wouldn't bet a fortune on it.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 7, 2013 21:49 UTC (Thu) by JanC_ (guest, #34940) [Link] It's like watching a rider in the Tour de France break away from the pel[o]ton half way through every stage and get caught every time long before the finish. It may make for good entertainment, but it won't win that rider anything. You obviously don't understand the cycling sport business. ;) It will probably win that rider a lucrative contract for the years to come, because it gives the team's sponsors a lot of media exposure without having to pay even bigger amounts of money for a rider that wins many stages or ends up on the final podium. (And if you think it's easy to do those breakaways, check out how many riders succeed in catching the "right" one that can remain in front for a long time—occasionally even until the finish.)
Canonical reveals plans to launch Mir display server (The H) Posted Mar 5, 2013 6:37 UTC (Tue) by HelloWorld (guest, #56129) [Link]
> Very likely the reasons for not using Wayland were bogus and the real reasons are something like "We want to put out a project by 2014 and are not confident Wayland will be ready".
The Weston project is very open to contributions, so Canonical could just pay developers to add whatever functionality is needed to Weston. And if that doesn't work out for whatever reason, they can write their own Wayland compositor and add whatever protocol extensions they deem necessary. There's no reason to believe that developing their own display server will save them any time, to the contrary.
The real problem with Canonical is that they ultimately have the mindset of a proprietary software company. They require copyright assignments for no good reason, they develop proprietary software (the Ubuntu One Server), they develop things behind closed walls, they suffer from serious NIH and they generally seem to focus on competition instead of cooperation. I personally just encourage people to switch away from Ubuntu due to this attitude. The Linux ecosystem would be better off without Canonical.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 6, 2013 5:25 UTC (Wed) by Serge (guest, #84957) [Link]
> so Canonical could just pay developers to add whatever functionality is needed to Weston. And if that doesn't work out for whatever reason, they can write their own Wayland compositor and add whatever protocol extensions they deem necessary.
Or they could use DirectFB and patch ilixi compositor to their needs. On the other hand Wayland people could also write X11 protocol extension instead of reinventing X11 into core Wayland + bunch of compositor extensions.
It's kind of fun to read how people were excited about new Wayland announcements and how those people now worry about "one more windowing system" and "what should Steam@Valve be now developed for".
If people really cared about Linux being successful and attractive for developers they would stick to common stable solutions, i.e. Xorg, and try making it better.
Canonical reveals plans to launch Mir display server (The H) Posted Mar 6, 2013 9:25 UTC (Wed) by tnoo (subscriber, #20427) [Link]
The Wayland develpers are mostly the Xorg developers. So they are not re-inventing the wheel, but are working hard to develop a round, smoothly running wheel instead of the established roundish thing with bumps, spikes, and struts and in all directions that is Xorg.
|
|||||||||||||