Who wants to test MIR?

[serdotlinecho]

Working really hard since 2008

[Gyokuro]

In my opionion they feel now more pressure and wayland/weston shows mature engineering (if you check git commits alot of various people sending patches and MIR shows most commits are from @canonical.com) work which I think lets to follwing problem: Ubuntu will go with MIR (if they succed) and the rest of linux projects/distributions with wayland/weston. I would really like to see that MIR get cancelled and following wayland/weston.

[Mr.JJ]

I added emphasis to what I think is the most important point in the quote, as developing something new is the way Ubuntu works.

True. But something new do not always equates to something better. And in canonical's decisions, none of them are any better (technologically or performance wise). As others pointed out, compiz, upstart, unity, etc... Sure they are different, but not any better.

.....if Wayland development was where Ubuntu expected it to be by this time, we'd all be testing it now, but as it is, it is just a curiosity that some of us play with and discard, as it really isn't ready for everyday usage.

How long you think linux kernel took to reach version 1.0? Three long yrs. Even then it had only 1% total no of lines of code (compared to now). Then it took another decade to reach 50% size. If you think there are shortcuts/quick-fixes to wonderful and original pieces of software, then welcome to software engineering.

Adding on to that quick-fix part, recently I see another possibility coming out in the comments. Canonical might be thinking about using the android (patched) kernel and android display servers. So, they do not need any long development and they can use the readily avialable drivers etc. Considering their work in ubuntu for android (It was also around the same time they stopped working on wayland) and their history, I see that as a very possible solution.

[Mr.JJ]

They are still working on getting the windows to minimize and maximize. So even with all that backing I can only ask why all of this is happening so late in the game.

From that comment alone, I know that you do not have any experience at all in software development.

It is not difficult (and do not take much time) to add a feature like that. What takes time and is difficult is to decide on what to implement and how to implement that; creating the standards. I can write thousands of lines of code in a single day, working code. But that will not be good code- extendable, scalable, adaptable code.

In one person projects they are creating applications in a day or two but only that developer can work on it and it cannot use, say a different toolkit etc. That is fine at the app level. But at the low level, and when creating standards it is important to consider various use cases and get everyone agree on the solutions. This is no flat development structure. As I said above, there are no quick-fixes for original quality software.

Say, in car manufacturing what takes time is to decide/finalise the design. The actual development in itself is not difficult or time consuming. The performance etc is directly related to the effort that is went into the design. I can use old engines and spare parts from different manufactures and make a car; for personal use. But if I expect it to be the largest selling/best performing car in the world, I am being stupid.

[Stinger]

@ cariboo907
I like the article you linked to, I read it a couple of times now.
Ubuntu Mir: Is This the Future of Linux Everywhere?
It raises a couple of questions though in the sense of why is Canonical trying to 'reinvent the wheel' with projects like MIR (and Unity) ?
The answer that springs into mind is control. Canonical want's to control it's projects.

That's necessary and maybe OK from a business point of view but it leaves us users without any influence because, to quote from the article:

Rather than working out in the open and accepting feedback on message boards — or code contributions through a system like Github — Canonical chose to work in secret, allowing only its inner circle to contribute. Some call this approach “throwing code over the wall.” It’s another sign that Canonical wants greater control over its open source projects.

Take MIR as an example, why didn't Canonical try to influence and turn wayland in the direction they wanted it to be ? IMO because Canonical can't control Wayland

Unity as another, why didn't Canonical try to 'bend' Gnome Shell the way they wanted ? IMO because they can't control the Gnome project (still presenting a challenge, remember Nautilus 3.6 )
Gnome 3.8 components doesn't fit Unity and UGR will have to release yet another 'stable' Gnome 3.6 version because, like me, the UGR developers have no control.

As Ubuntu user, I care.
I care about the Ubuntu community, that's why I haven't left yet and still try to influence the Ubuntu development.

'We will just have to wait and see what MIR will bring' EgoGratis said, and I'm afraid that's all we can do, because as community user I feel I have lost my influence. Here is another quote from the article:

Canonical’s decision to keep development of Ubuntu 13.04 closed was also controversial. Although by releasing its code under a GPL license, Ubuntu remains open source by the Open Source Initiative’s definition, it violates the spirit of openness and transparency.

I couldn't have said it better myself, Canonical 'violates the spirit of openness and transparency' and I might add 'makes me feel I have lost my influence as user'.

[ventrical]

From that comment alone, I know that you do not have any experience at all in software development.

That's a really cheap presumption on your part and Iv'e considered the source.

It is not difficult (and do not take much time) to add a feature like that. What takes time and is difficult is to decide on what to implement and how to implement that; creating the standards. I can write thousands of lines of code in a single day, working code. But that will not be good code- extendable, scalable, adaptable code.

In one person projects they are creating applications in a day or two but only that developer can work on it and it cannot use, say a different toolkit etc. That is fine at the app level. But at the low level, and when creating standards it is important to consider various use cases and get everyone agree on the solutions. This is no flat development structure. As I said above, there are no quick-fixes for original quality software.

Say, in car manufacturing what takes time is to decide/finalise the design. The actual development in itself is not difficult or time consuming. The performance etc is directly related to the effort that is went into the design. I can use old engines and spare parts from different manufactures and make a car; for personal use. But if I expect it to be the largest selling/best performing car in the world, I am being stupid.

You are obsfucating the issue again. Wayland is a choice as MIR is a choice. Nobody (or certainly suggested in the topic content) is forcing you to test MIR or even consider it. Please do likewise and allow some of us beta testers to see what MIR is all about. We're not cramming MIR down your throat, please don't cram wayland down ours.

[EgoGratis]

They are still working on getting the windows to minimize and maximize. So even with all that backing I can only ask why all of this is happening so late in the game.


http://www.phoronix.com/scan.php?pag...tem&px=MTMyMjQ

I wouldn't underestimate Wayland this way it's still going to happen and probably it will still be huge and a lot of work was already done.

As others pointed out, compiz, upstart, unity, etc... Sure they are different, but not any better.

And not any worse from average user expectations. I bet a lot of desktop users will miss Compiz for example when it will be retired from all main distros. I would miss Unity too to be honest if it would stop being produced.

Take MIR as an example, why didn't Canonical try to influence and turn wayland in the direction they wanted it to be ? IMO because Canonical can't control Wayland

Unity as another, why didn't Canonical try to 'bend' Gnome Shell the way they wanted ? IMO because they can't control the Gnome project (still presenting a challenge, remember Nautilus 3.6 )
Gnome 3.8 components doesn't fit Unity and UGR will have to release yet another 'stable' Gnome 3.6 version because, like me, the UGR developers have no control.

Probably Canonical does gain value if it holds (copyright) vital components of it's products and there is grater influence that Canonical can impose there is no point in arguing that. But to be honest if in few years time Wayland proves to be the way to go i do believe current Canonical would switch back.

GnomeShell probably couldn't be bended in a way it would be current Unity. Changes would probably have be too big and to hard to maintain. ATM i think Canonical is in the phase when it want's to build something fast and to see if there will be results an as company probably there is more (potential) value in your products doing it like this. But i would not exaggerate ATM because GNU/Linux it's still barely represented in consumer market and for example Canonical using Qt is still indication they are after best FOSS available in the market ATM.

Different FOSS upstreams and community built around them currently do need to revalidate their position and future stance where and how they want to take FOSS and how does Ubuntu fit in all of this. There is nothing wrong with that. I hope common grounds could be something like to deliver FOSS (GNU/Linux and it's ecosystem) to 10x users install base we have ATM in consumer market place in let's say less then 5 years!

[ventrical]

@ cariboo907
I like the article you linked to, I read it a couple of times now.
Ubuntu Mir: Is This the Future of Linux Everywhere?
It raises a couple of questions though in the sense of why is Canonical trying to 'reinvent the wheel' with projects like MIR (and Unity) ?
The answer that springs into mind is control. Canonical want's to control it's projects.

That's necessary and maybe OK from a business point of view but it leaves us users without any influence because, to quote from the article:


Take MIR as an example, why didn't Canonical try to influence and turn wayland in the direction they wanted it to be ? IMO because Canonical can't control Wayland

Unity as another, why didn't Canonical try to 'bend' Gnome Shell the way they wanted ? IMO because they can't control the Gnome project (still presenting a challenge, remember Nautilus 3.6 )
Gnome 3.8 components doesn't fit Unity and UGR will have to release yet another 'stable' Gnome 3.6 version because, like me, the UGR developers have no control.

As Ubuntu user, I care.
I care about the Ubuntu community, that's why I haven't left yet and still try to influence the Ubuntu development.

'We will just have to wait and see what MIR will bring' EgoGratis said, and I'm afraid that's all we can do, because as community user I feel I have lost my influence. Here is another quote from the article:

I couldn't have said it better myself, Canonical 'violates the spirit of openness and transparency' and I might add 'makes me feel I have lost my influence as user'.

Obviously there were some serious security issues with wayland. I think it is a wise thing to have more control over the display server because this way there is more control over possible security leaks. Ubuntu is attractive, not only to FOSS, Ubuntu and it's clients, but also to hackers looking to cause some chaos. I think it is a smart move all around ... lest we have to be fitted with some reasonable facsimile of an AVG or Comodo guard.

[ventrical]

I wouldn't underestimate Wayland this way it's still going to happen and probably it will still be huge and a lot of work was already done.

With all due respect , I never, ever underestimated wayland! I was apporoached about it, I studied it, I beta tested it, I ran all the tests others were testing .. and then it did nothing forsome long period of time. While testing Precise and Quantal we had other segments to test other than wayland. Are we at fault *here* because of our non-testingness of wayland!? Now we have MIR to chomp away. You guys .. go ahead and run, test , bellyache and incorporate wayland into your systems(or you could always convert over to Mint). I'm going to have fun busting my systems with MIR !

[EgoGratis]

Obviously there were some serious security issues with wayland.

No.

And i will end the debate in this thread with this thought. Ubuntu was upfront and transparent as one could get for few years now already and upstreams are not the only one to blame on why Ubuntu hasn't gain more install base in consumer market place but i do believe Canonical would give back more control of the direction Ubuntu is going to usptreams in areas where it was decided to do things differently if upstreams would find a viable solution on how to achieve let's say 10x install base of GNU/Linux and it's ecosystem in few years time. If usptreams are not that interested in achieving this goal then it's quite obvious why Ubuntu is going in the direction it's currently going.