Re: [Elementary-dev-community] Mir Discussion with Jono

[Sergey "Shnatsel" Davidoff]

Some perspective on Xmir setup seen in Ubuntu 13.10 and 14.04:
>From Canonical: http://blog.cooperteam.net/2013/07/xmir-performance.html
>From X/Wayland upstreams: http://mjg59.dreamwidth.org/26254.html

-- 
Sergey "Shnatsel" Davidoff
OS architect @ elementary
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Sergey "Shnatsel" Davidoff]

2013/7/13 Sergio Costas 

> And what about creating a WaylandMir? This is: as there is XMir that
> allows to run X apps over Mir, there should be possible to create a piece
> of code that translates Wayland calls to Mir calls. It should be even
> lighter than XMir because there are a lot of things in X that you must
> implement, even if you don't use them, but Wayland and Mir, conceptually
> are so similar that this should be much easier, and need much less
> resources...
>

In theory it's possible, but I can't see how it would be useful. Also, it's
probably easier to fork the relevant Wayland rendering code and make it
render to Mir than to translate calls dynamically.

Also, it seems there's significant confusion surrounding X compability
layers. What XWayland does is running tiny rootless per-app X servers and
rendering that to Wayland compositor which may or may not forward the
frames to a system compositor. What Xmir currently does is running
full-blown rooted X servers that render frames to Mir system compositor.
Which compability layer scenario are you suggesting?

-- 
Sergey "Shnatsel" Davidoff
OS architect @ elementary
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Sergio Costas]

Hi:

And what about creating a WaylandMir? This is: as there is XMir that 
allows to run X apps over Mir, there should be possible to create a 
piece of code that translates Wayland calls to Mir calls. It should be 
even lighter than XMir because there are a lot of things in X that you 
must implement, even if you don't use them, but Wayland and Mir, 
conceptually are so similar that this should be much easier, and need 
much less resources...


El 11/07/13 00:57, Daniel Foré escribió:

Obviously that's ridiculous to think elementary would write all those apps. Our 
mission is to build a platform, not every app you would expect to ever find. 
That's what 3rd party developers are for.

Anyways, as Jono clarified, Gtk+ apps will run in rootless X so no need to 
freak out.

Best Regards,
Daniel Foré

El jul 10, 2013, a las 3:48 p.m., Jo-Erlend Schinstad 
 escribió:


On 11 July 2013 00:28, Daniel Foré  wrote:

See Jono's clarification :)

There is absolutely no reason to believe that not supporting Gtk+ would doom 
Ubuntu to fail. Android and iOS launched just fine without supporting Windows 
or OS X apps.

However, neither Android or Ios were designed to be Desktop operating systems. 
Ubuntu is. When you connect a big screen, keyboard and mouse to your phone, 
you're supposed to get exactly that. In order for _that_ to succeed, you really 
do need all the apps you can get. Removing all GTK applications from the 
desktop would seriously limit the chance of success.
  

And in fact, this is not so different from our expectations either. A serious 
OS needs apps that are built specifically around it's platform. Cross platform 
apps suck.

Then Elementary OS is planning on writing everything from office suits to movie 
editors because none of the current ones are specifically designed for 
Elementary? Good luck with that. Guess it'll take a while.


I would fully expect it to seem odd that someone would try to run GIMP on 
Ubuntu 16.04 just like it would be odd for someone to want to run Krita on 
Ubuntu now.


I don't think that's particularly odd at all, since it's possible now. Removing 
the _possibility_ of running Krita on Ubuntu, however, would be a seriously odd 
thing to do.

Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and Robert 
Carr replied: «We've always said that we were creating a GTK  backend. but it's 
behind anything for the phone or the system compositor of course :)»

_That_ makes sense.

Robert Carr is one of the big names in Mir development.







--
Nos leemos
 RASTER(Linux user #228804)
ras...@rastersoft.com  http://www.rastersoft.com

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 4:38 PM, Daniel Foré wrote:

> Hey Robert,
>
> Thanks for the clarification! From our discussion with Jono we left with
> the impression that Canonical wasn't intending to get Gtk+ working natively
> on Mir
>
>
Oh, interesting. I spoke to another member of the Mir team about this
earlier, but maybe Robert has a clearer determination of this.

   Jono


-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

Hey Robert,

Thanks for the clarification! From our discussion with Jono we left with the 
impression that Canonical wasn't intending to get Gtk+ working natively on Mir

Best Regards,
Daniel Foré

El jul 10, 2013, a las 4:18 p.m., Robert Ancell  
escribió:

> 
> 
> 
> On Thu, Jul 11, 2013 at 10:48 AM, Jo-Erlend Schinstad 
>  wrote:
>> 
>> Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and Robert 
>> Carr replied: «We've always said that we were creating a GTK  backend. but 
>> it's behind anything for the phone or the system compositor of course :)»
> 
> 
> To clarify - it's been in the blueprints for some time, see "GTK+ support":
> https://blueprints.launchpad.net/ubuntu/+spec/client-1303-mir-converged
> 
> The reason we're not working on it right now is it's not required for the 
> phone, or for an XMir based desktop which are the two goals we are working 
> towards. There's been a number of people who have expressed interest in 
> developing a GTK+ backend for Mir, the most recent being 
> https://lists.ubuntu.com/archives/mir-devel/2013-July/000275.html.
> 
> We don't expect it to be an enormous amount of work due to the refactoring 
> that has occurred in GDK to support Wayland and other next generation display 
> servers and we've been hoping that someone with the time will implement it 
> sooner rather than later :)
> 
> --Robert
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Robert Ancell]

On Thu, Jul 11, 2013 at 10:48 AM, Jo-Erlend Schinstad <
joerlend.schins...@ubuntu.com> wrote:

>
> Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and
> Robert Carr replied: «We've always said that we were creating a GTK
> backend. but it's behind anything for the phone or the system compositor of
> course :)»
>


To clarify - it's been in the blueprints for some time, see "GTK+ support":
https://blueprints.launchpad.net/ubuntu/+spec/client-1303-mir-converged

The reason we're not working on it right now is it's not required for the
phone, or for an XMir based desktop which are the two goals we are working
towards. There's been a number of people who have expressed interest in
developing a GTK+ backend for Mir, the most recent being
https://lists.ubuntu.com/archives/mir-devel/2013-July/000275.html.

We don't expect it to be an enormous amount of work due to the refactoring
that has occurred in GDK to support Wayland and other next generation
display servers and we've been hoping that someone with the time will
implement it sooner rather than later :)

--Robert
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

Obviously that's ridiculous to think elementary would write all those apps. Our 
mission is to build a platform, not every app you would expect to ever find. 
That's what 3rd party developers are for.

Anyways, as Jono clarified, Gtk+ apps will run in rootless X so no need to 
freak out. 

Best Regards,
Daniel Foré

El jul 10, 2013, a las 3:48 p.m., Jo-Erlend Schinstad 
 escribió:

> On 11 July 2013 00:28, Daniel Foré  wrote:
>> See Jono's clarification :)
>> 
>> There is absolutely no reason to believe that not supporting Gtk+ would doom 
>> Ubuntu to fail. Android and iOS launched just fine without supporting 
>> Windows or OS X apps.
> 
> However, neither Android or Ios were designed to be Desktop operating 
> systems. Ubuntu is. When you connect a big screen, keyboard and mouse to your 
> phone, you're supposed to get exactly that. In order for _that_ to succeed, 
> you really do need all the apps you can get. Removing all GTK applications 
> from the desktop would seriously limit the chance of success. 
>  
>> 
>> And in fact, this is not so different from our expectations either. A 
>> serious OS needs apps that are built specifically around it's platform. 
>> Cross platform apps suck.
> 
> Then Elementary OS is planning on writing everything from office suits to 
> movie editors because none of the current ones are specifically designed for 
> Elementary? Good luck with that. Guess it'll take a while. 
> 
>> 
>> I would fully expect it to seem odd that someone would try to run GIMP on 
>> Ubuntu 16.04 just like it would be odd for someone to want to run Krita on 
>> Ubuntu now.
> 
> 
> I don't think that's particularly odd at all, since it's possible now. 
> Removing the _possibility_ of running Krita on Ubuntu, however, would be a 
> seriously odd thing to do. 
> 
> Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and Robert 
> Carr replied: «We've always said that we were creating a GTK  backend. but 
> it's behind anything for the phone or the system compositor of course :)»
> 
> _That_ makes sense.
> 
> Robert Carr is one of the big names in Mir development. 
> 
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 3:48 PM, Jo-Erlend Schinstad <
joerlend.schins...@ubuntu.com> wrote:

> On 11 July 2013 00:28, Daniel Foré  wrote:
>
>> See Jono's clarification :)
>>
>> There is absolutely no reason to believe that not supporting Gtk+ would
>> doom Ubuntu to fail. Android and iOS launched just fine without supporting
>> Windows or OS X apps.
>>
>
> However, neither Android or Ios were designed to be Desktop operating
> systems. Ubuntu is. When you connect a big screen, keyboard and mouse to
> your phone, you're supposed to get exactly that. In order for _that_ to
> succeed, you really do need all the apps you can get. Removing all GTK
> applications from the desktop would seriously limit the chance of success.
>
>

Again, we are *not* removing GTK apps. We will simply be running them as
rootless X apps; the user will notice no difference. :-)

We are however encouraging app devs to use our SDK to build apps, which
includes features for presenting different interfaces dependent on which
device you are on.


>
>> And in fact, this is not so different from our expectations either. A
>> serious OS needs apps that are built specifically around it's platform.
>> Cross platform apps suck.
>>
>
> Then Elementary OS is planning on writing everything from office suits to
> movie editors because none of the current ones are specifically designed
> for Elementary? Good luck with that. Guess it'll take a while.
>
>
>> I would fully expect it to seem odd that someone would try to run GIMP on
>> Ubuntu 16.04 just like it would be odd for someone to want to run Krita on
>> Ubuntu now.
>>
>
>
> I don't think that's particularly odd at all, since it's possible now.
> Removing the _possibility_ of running Krita on Ubuntu, however, would be a
> seriously odd thing to do.
>
> Agreed.

   Jono

-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Allen Lowe]

However, as Jono said, GTK+ apps will continue to function as rootless X
apps anyway, so no one is removing any possibilites.


On Wed, Jul 10, 2013 at 4:48 PM, Jo-Erlend Schinstad <
joerlend.schins...@ubuntu.com> wrote:

> On 11 July 2013 00:28, Daniel Foré  wrote:
>
>> See Jono's clarification :)
>>
>> There is absolutely no reason to believe that not supporting Gtk+ would
>> doom Ubuntu to fail. Android and iOS launched just fine without supporting
>> Windows or OS X apps.
>>
>
> However, neither Android or Ios were designed to be Desktop operating
> systems. Ubuntu is. When you connect a big screen, keyboard and mouse to
> your phone, you're supposed to get exactly that. In order for _that_ to
> succeed, you really do need all the apps you can get. Removing all GTK
> applications from the desktop would seriously limit the chance of success.
>
>
>>
>> And in fact, this is not so different from our expectations either. A
>> serious OS needs apps that are built specifically around it's platform.
>> Cross platform apps suck.
>>
>
> Then Elementary OS is planning on writing everything from office suits to
> movie editors because none of the current ones are specifically designed
> for Elementary? Good luck with that. Guess it'll take a while.
>
>
>> I would fully expect it to seem odd that someone would try to run GIMP on
>> Ubuntu 16.04 just like it would be odd for someone to want to run Krita on
>> Ubuntu now.
>>
>
>
> I don't think that's particularly odd at all, since it's possible now.
> Removing the _possibility_ of running Krita on Ubuntu, however, would be a
> seriously odd thing to do.
>
> Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and
> Robert Carr replied: «We've always said that we were creating a GTK
> backend. but it's behind anything for the phone or the system compositor of
> course :)»
>
> _That_ makes sense.
>
> Robert Carr is one of the big names in Mir development.
>
>
> --
> Mailing list: https://launchpad.net/~elementary-dev-community
> Post to : elementary-dev-community@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~elementary-dev-community
> More help   : https://help.launchpad.net/ListHelp
>
>
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jo-Erlend Schinstad]

On 11 July 2013 00:28, Daniel Foré  wrote:

> See Jono's clarification :)
>
> There is absolutely no reason to believe that not supporting Gtk+ would
> doom Ubuntu to fail. Android and iOS launched just fine without supporting
> Windows or OS X apps.
>

However, neither Android or Ios were designed to be Desktop operating
systems. Ubuntu is. When you connect a big screen, keyboard and mouse to
your phone, you're supposed to get exactly that. In order for _that_ to
succeed, you really do need all the apps you can get. Removing all GTK
applications from the desktop would seriously limit the chance of success.


>
> And in fact, this is not so different from our expectations either. A
> serious OS needs apps that are built specifically around it's platform.
> Cross platform apps suck.
>

Then Elementary OS is planning on writing everything from office suits to
movie editors because none of the current ones are specifically designed
for Elementary? Good luck with that. Guess it'll take a while.


> I would fully expect it to seem odd that someone would try to run GIMP on
> Ubuntu 16.04 just like it would be odd for someone to want to run Krita on
> Ubuntu now.
>


I don't think that's particularly odd at all, since it's possible now.
Removing the _possibility_ of running Krita on Ubuntu, however, would be a
seriously odd thing to do.

Had to take a trip to #Ubuntu-mir on Freenode. I asked about this and
Robert Carr replied: «We've always said that we were creating a GTK
backend. but it's behind anything for the phone or the system compositor of
course :)»

_That_ makes sense.

Robert Carr is one of the big names in Mir development.
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

See Jono's clarification :)

There is absolutely no reason to believe that not supporting Gtk+ would doom 
Ubuntu to fail. Android and iOS launched just fine without supporting Windows 
or OS X apps.

And in fact, this is not so different from our expectations either. A serious 
OS needs apps that are built specifically around it's platform. Cross platform 
apps suck.

I would fully expect it to seem odd that someone would try to run GIMP on 
Ubuntu 16.04 just like it would be odd for someone to want to run Krita on 
Ubuntu now. 

Best Regards,
Daniel Foré

El jul 10, 2013, a las 2:53 p.m., Jo-Erlend Schinstad 
 escribió:

> 
> 
> On 10 July 2013 22:45, Daniel Foré  wrote:
>> I can confirm that from speaking to Canonical employees and attending UDS 
>> that the tone has been for a long time that they will eventually stop 
>> supporting Gtk+ in favor of Qt
> 
> For Ubuntu SDK apps, that has been known for quite a while and there are good 
> reasons for it. But that's something _very_ different from not supporting 
> GTK+ on their display server, though it's obviously vice versa, purely 
> technically speaking. That would effectively kill Mir right from the 
> beginning and this seems a highly unlikely goal for any software company – to 
> design your products to fail. 
>  
>> 
>> At the current rate, 14.04 may be the last version of Ubuntu under which you 
>> can run Gtk+ apps unless the community wants to build Mir support for Gtk+.
>> 
>> Canonical is building a suite of default apps in Qt and all their 
>> third-party dev documentation is now focused on Qt. This is happening. 
>> Ubuntu is for Qt.
> 
> You seem to still be talking about Ubuntu SDK apps? Yes, on Touch, GTK+ is 
> not supported and according to Michael Hall, this is primarily because of 
> scaling GUIs to make applications look the same even if the resolution 
> changes. You should of course still be _able_ to run GTK+ apps, but they'd 
> look horrible because of lacking support from the SDK. But on the _desktop_? 
> It makes absolutely no sense in removing all those applications because they 
> themselves want to write in Qt. 
> 
>> 
>> But like ConciousUser has stated, there is absolutely nothing wrong with 
>> that. That is 100% their choice to make and in my opinion, having a 
>> dedication to a single toolkit is a great choice. That's why elementary also 
>> has a dedication to a single toolkit and if we built a new display server I 
>> can tell you right now we'd have no intention of making Qt run on it.
> 
> There are lots of things wrong with that, which is why I don't believe it. 
> But let me get this straight; if X has to be replaced and nobody else writes 
> a new display server for you, then Elementary OS is willing to remove 
> software like GIMP, Evolution, Inkscape all all other great GTK+ 
> applications? You wouldn't want to support Evolution for your 
> business/corporate users – even if those represent your primary income? That, 
> to me, seems like an amazingly bad idea. Perhaps you'd decide to port those 
> applications to Qt instead? After all, you want your OS to be a success, 
> right? Then you need to have good apps. 
> 
> I can't believe Canonical intends to do any of that, and I won't believe it 
> until I see an official statement. The only other option I can see is to keep 
> supporting XMir forever, which in time will mean supporting X.org alone. That 
> would most likely be a lot harder than supporting a GTK+ backend. Right? 
> Likewise, porting everything from GIMP to Evolution would also mean an insane 
> amount of work. 
> 
> 
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 2:10 PM, cameron  wrote:

> "I can tell you right now we'd have no intention of making Qt run on it."
>
> You would probably have no expectation of making KDE run on it, though. It
> is an odd stance Canonical is taking. They want elementary OS (and other
> spins) to use Mir, but they put no effort into supporting the toolkits (and
> other needs) of these derivatives.
>

To be clear: we want to provide as much help and guidance as possible to
help all distros and upstreams explore Mir as a display server that may
work well for them. Mir is going to ship in 13.10 as a well supported
stable display server, and in 14.04 we will have it fully supported for
five years. From the perspective of a derivative I think this is an
attractive option for this foundational piece of the stack - it means that
you folks don't have to worry about it.

Canonical has limited engineering resources like any organization, and we
don't have the resources to build Mir support for every toolkit out there.
We would rather focus our efforts on making Mir rock solid and production
ready for 13.10 and then also provide support and guidance for those who do
want to build or integrate Mir support for their distro/upstream. This was
why I wanted to reach out to the team, to see how we can help.

   Jono

-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

Sorry, I might have made it more clear that was speculation.

Best Regards,
Daniel Foré

El jul 10, 2013, a las 3:01 p.m., Manish Sinha  
escribió:

> On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré  wrote:
>> At the current rate, 14.04 may be the last version of Ubuntu under which you 
>> can run Gtk+ apps unless the community wants to build Mir support for Gtk+.
> 
> I don't think I understand this properly. You mean to say that after
> 14.04 Ubuntu cannot run GTK+ apps as they will go pure Mir and Mir
> won't have GTK+ support?
> Is this true? What about all the apps written in GTK+? What will they
> be replaced with?
> 
> -
> Manish

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré wrote:

> I can confirm that from speaking to Canonical employees and attending UDS
> that the tone has been for a long time that they will eventually stop
> supporting Gtk+ in favor of Qt
>
>
There has been quite some discussion in the past about focusing more on a
single toolkit, and as Daniel says, overall we have focused on Qt and QML.
It is the basis of our SDK, we are re-writing Unity in it (the convergence
Unity), and it meets our needs well for the different devices we are
building for.


> At the current rate, 14.04 may be the last version of Ubuntu under which
> you can run Gtk+ apps unless the community wants to build Mir support for
> Gtk+.
>
>
See my previous message about the GTK support - GTK support will continue
to work in Ubuntu, but will be rootless X support unless there is a GTK Mir
backend. Some may wonder why Canonical is not investing in this
backend...well, we are focusing our new development efforts on Qt/QML and
we can still deliver our GTK apps via rootless X sessions.


> Canonical is building a suite of default apps in Qt and all their
> third-party dev documentation is now focused on Qt. This is happening.
> Ubuntu is for Qt.
>
> But like ConciousUser has stated, there is absolutely nothing wrong with
> that. That is 100% their choice to make and in my opinion, having a
> dedication to a single toolkit is a great choice. That's why elementary
> also has a dedication to a single toolkit and if we built a new display
> server I can tell you right now we'd have no intention of making Qt run on
> it.
>
>
Agreed.

   Jono

-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 3:12 PM, Jono Bacon  wrote:

>
>
>
> On Wed, Jul 10, 2013 at 3:01 PM, Manish Sinha wrote:
>
>> On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré 
>> wrote:
>> > At the current rate, 14.04 may be the last version of Ubuntu under
>> which you can run Gtk+ apps unless the community wants to build Mir support
>> for Gtk+.
>>
>> I don't think I understand this properly. You mean to say that after
>> 14.04 Ubuntu cannot run GTK+ apps as they will go pure Mir and Mir
>> won't have GTK+ support?
>> Is this true? What about all the apps written in GTK+? What will they
>> be replaced with?
>>
>
> I checked into this with the Mir team and the 14.04 plan is to run non-Qt
> apps that don't have a Mir backend as rootless X apps, so all GTK apps will
> continue to work. Currently Canonical is investing in building a Qt backend
> and providing help and guidance for those who want to build Mir backends
> for other toolkits. In fact, there is an active discussion underway on
> mir-devel as we speak about a GTK Mir backend (
> https://lists.ubuntu.com/archives/mir-devel/2013-July/000275.html).
>
> So in a nutshell, there will be no regressions at all with regards to GTK
> apps in 14.04.
>
> Oops, sorry it turns out I had this wrong - for 14.04 we will continue to
ship Unity 7 on XMir on Mir, which will obviously continue to run GTK apps.
In 14.10 the GTK apps will be rootless, although if the community GTK Mir
backend work continues to make improvements, we may well have native Mir
GTK support by then.

   Jono

-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jono Bacon]

On Wed, Jul 10, 2013 at 3:01 PM, Manish Sinha wrote:

> On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré 
> wrote:
> > At the current rate, 14.04 may be the last version of Ubuntu under which
> you can run Gtk+ apps unless the community wants to build Mir support for
> Gtk+.
>
> I don't think I understand this properly. You mean to say that after
> 14.04 Ubuntu cannot run GTK+ apps as they will go pure Mir and Mir
> won't have GTK+ support?
> Is this true? What about all the apps written in GTK+? What will they
> be replaced with?
>

I checked into this with the Mir team and the 14.04 plan is to run non-Qt
apps that don't have a Mir backend as rootless X apps, so all GTK apps will
continue to work. Currently Canonical is investing in building a Qt backend
and providing help and guidance for those who want to build Mir backends
for other toolkits. In fact, there is an active discussion underway on
mir-devel as we speak about a GTK Mir backend (
https://lists.ubuntu.com/archives/mir-devel/2013-July/000275.html).

So in a nutshell, there will be no regressions at all with regards to GTK
apps in 14.04.

-- 
Jono Bacon
Ubuntu Community Manager
www.ubuntu.com / www.jonobacon.org
www.identi.ca/jonobacon www.twitter.com/jonobacon
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Manish Sinha]

On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré  wrote:
> At the current rate, 14.04 may be the last version of Ubuntu under which you 
> can run Gtk+ apps unless the community wants to build Mir support for Gtk+.

I don't think I understand this properly. You mean to say that after
14.04 Ubuntu cannot run GTK+ apps as they will go pure Mir and Mir
won't have GTK+ support?
Is this true? What about all the apps written in GTK+? What will they
be replaced with?

-
Manish

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jo-Erlend Schinstad]

On 10 July 2013 22:45, Daniel Foré  wrote:

> I can confirm that from speaking to Canonical employees and attending UDS
> that the tone has been for a long time that they will eventually stop
> supporting Gtk+ in favor of Qt
>

For Ubuntu SDK apps, that has been known for quite a while and there are
good reasons for it. But that's something _very_ different from not
supporting GTK+ on their display server, though it's obviously vice versa,
purely technically speaking. That would effectively kill Mir right from the
beginning and this seems a highly unlikely goal for any software company –
to design your products to fail.


>
> At the current rate, 14.04 may be the last version of Ubuntu under which
> you can run Gtk+ apps unless the community wants to build Mir support for
> Gtk+.
>
> Canonical is building a suite of default apps in Qt and all their
> third-party dev documentation is now focused on Qt. This is happening.
> Ubuntu is for Qt.
>

You seem to still be talking about Ubuntu SDK apps? Yes, on Touch, GTK+ is
not supported and according to Michael Hall, this is primarily because of
scaling GUIs to make applications look the same even if the resolution
changes. You should of course still be _able_ to run GTK+ apps, but they'd
look horrible because of lacking support from the SDK. But on the
_desktop_? It makes absolutely no sense in removing all those applications
because they themselves want to write in Qt.


> But like ConciousUser has stated, there is absolutely nothing wrong with
> that. That is 100% their choice to make and in my opinion, having a
> dedication to a single toolkit is a great choice. That's why elementary
> also has a dedication to a single toolkit and if we built a new display
> server I can tell you right now we'd have no intention of making Qt run on
> it.
>

There are lots of things wrong with that, which is why I don't believe it.
But let me get this straight; if X has to be replaced and nobody else
writes a new display server for you, then Elementary OS is willing to
remove software like GIMP, Evolution, Inkscape all all other great GTK+
applications? You wouldn't want to support Evolution for your
business/corporate users – even if those represent your primary income?
That, to me, seems like an amazingly bad idea. Perhaps you'd decide to port
those applications to Qt instead? After all, you want your OS to be a
success, right? Then you need to have good apps.

I can't believe Canonical intends to do any of that, and I won't believe it
until I see an official statement. The only other option I can see is to
keep supporting XMir forever, which in time will mean supporting X.org
alone. That would most likely be a lot harder than supporting a GTK+
backend. Right? Likewise, porting everything from GIMP to Evolution would
also mean an insane amount of work.
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[cameron]

"I can tell you right now we'd have no intention of making Qt run on 
it."


You would probably have no expectation of making KDE run on it, though. 
It is an odd stance Canonical is taking. They want elementary OS (and 
other spins) to use Mir, but they put no effort into supporting the 
toolkits (and other needs) of these derivatives.


On Wed, Jul 10, 2013 at 1:45 PM, Daniel Foré  
wrote:
I can confirm that from speaking to Canonical employees and attending 
UDS that the tone has been for a long time that they will eventually 
stop supporting Gtk+ in favor of Qt


At the current rate, 14.04 may be the last version of Ubuntu under 
which you can run Gtk+ apps unless the community wants to build Mir 
support for Gtk+. 

Canonical is building a suite of default apps in Qt and all their 
third-party dev documentation is now focused on Qt. This is 
happening. Ubuntu is for Qt.


But like ConciousUser has stated, there is absolutely nothing wrong 
with that. That is 100% their choice to make and in my opinion, 
having a dedication to a single toolkit is a great choice. That's why 
elementary also has a dedication to a single toolkit and if we built 
a new display server I can tell you right now we'd have no intention 
of making Qt run on it.


Best Regards,
Daniel Foré

El jul 10, 2013, a las 1:03 p.m., Conscious User 
 escribió:


 
 Em Qua, 2013-07-10 às 21:25 +0200, Jo-Erlend Schinstad escreveu:
 
 It also seems kind of weird that a desktop distribution would want 
to
 make such an enormous amount of applications second class citizens 
in

 this way.

 
 They already did. With or without Mir support, The GNOME stack and
 Compiz are clearly second class citizens in Ubuntu now. For 
evidence,

 look no further than the schedule of the last UDSes and pretty much
 *all* communication from Canonical employees ever since Ubuntu Touch
 was announced.
 
 And just to be clear, because people got confused about this the

 last time I brought it up: I DO NOT MEAN THIS AS CRITICISM AGAINST
 CANONICAL, NOR I AM IMPLYING ANY SORT OF MALICE OR HOSTILITY FROM
 THEIR SIDE. *No* distro can devote the same amount of attention to
 all apps. KDE apps have always been second class citizens in Ubuntu,
 and that is perfectly fine.
 
 
 
 -- 
 Mailing list: https://launchpad.net/~elementary-dev-community

 Post to : elementary-dev-community@lists.launchpad.net
 Unsubscribe : https://launchpad.net/~elementary-dev-community
 More help   : https://help.launchpad.net/ListHelp


--
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

I can confirm that from speaking to Canonical employees and attending UDS that 
the tone has been for a long time that they will eventually stop supporting 
Gtk+ in favor of Qt

At the current rate, 14.04 may be the last version of Ubuntu under which you 
can run Gtk+ apps unless the community wants to build Mir support for Gtk+. 

Canonical is building a suite of default apps in Qt and all their third-party 
dev documentation is now focused on Qt. This is happening. Ubuntu is for Qt.

But like ConciousUser has stated, there is absolutely nothing wrong with that. 
That is 100% their choice to make and in my opinion, having a dedication to a 
single toolkit is a great choice. That's why elementary also has a dedication 
to a single toolkit and if we built a new display server I can tell you right 
now we'd have no intention of making Qt run on it.

Best Regards,
Daniel Foré

El jul 10, 2013, a las 1:03 p.m., Conscious User  
escribió:

> 
> Em Qua, 2013-07-10 às 21:25 +0200, Jo-Erlend Schinstad escreveu:
> 
>> It also seems kind of weird that a desktop distribution would want to
>> make such an enormous amount of applications second class citizens in
>> this way.
> 
> They already did. With or without Mir support, The GNOME stack and
> Compiz are clearly second class citizens in Ubuntu now. For evidence,
> look no further than the schedule of the last UDSes and pretty much
> *all* communication from Canonical employees ever since Ubuntu Touch
> was announced.
> 
> And just to be clear, because people got confused about this the
> last time I brought it up: I DO NOT MEAN THIS AS CRITICISM AGAINST
> CANONICAL, NOR I AM IMPLYING ANY SORT OF MALICE OR HOSTILITY FROM
> THEIR SIDE. *No* distro can devote the same amount of attention to
> all apps. KDE apps have always been second class citizens in Ubuntu,
> and that is perfectly fine.
> 
> 
> 
> -- 
> Mailing list: https://launchpad.net/~elementary-dev-community
> Post to : elementary-dev-community@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~elementary-dev-community
> More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Conscious User]

Em Qua, 2013-07-10 às 21:25 +0200, Jo-Erlend Schinstad escreveu:

> It also seems kind of weird that a desktop distribution would want to
> make such an enormous amount of applications second class citizens in
> this way.

They already did. With or without Mir support, The GNOME stack and
Compiz are clearly second class citizens in Ubuntu now. For evidence,
look no further than the schedule of the last UDSes and pretty much
*all* communication from Canonical employees ever since Ubuntu Touch
was announced.

And just to be clear, because people got confused about this the
last time I brought it up: I DO NOT MEAN THIS AS CRITICISM AGAINST
CANONICAL, NOR I AM IMPLYING ANY SORT OF MALICE OR HOSTILITY FROM
THEIR SIDE. *No* distro can devote the same amount of attention to
all apps. KDE apps have always been second class citizens in Ubuntu,
and that is perfectly fine.



-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Paulo Truta]

I'm a little out of the scene now, but i don't really think we should do 
everything canonical wants. Today they changed to Mir, tomorrow it can be 
anything else.. And elementary has to follow them.

Also, i think that Ubuntu is no longer a good base distro for elementary to be 
based on. There are way more stable distros (Like Debian), that don't change 
that much in time and allow the Elementary developer team to work without the 
fear of having to modify and / or adapt all the hard work they had. Ubuntu is 
taking a major remodeling and i don't really think it will stop on Mir.

I think we should really think this for Luna +1. 

PS: I really like Mir and all the thinks the Ubuntu developers are doing for 
the Linux community, taking it to the next level. But i think Elementary is 
still a relatively new project and has to have a stable base, or else, 
iterations of this distro will take even longer to launch than Luna.

Cheers ;)

A 10/07/2013, às 18:40, Conscious User escreveu:

> 
> Developers from elementaryOS should not be the main force
> behind porting toolkits to display servers. This is either
> the responsibility of toolkit developers or display server
> developers. If anything, for the deeper technical knowledge
> this depends on.
> 
> Reactions from the GNOME community indicate that they have
> no intention of supporting Mir, so this leaves Canonical as
> the only option. Historically, trusting Canonical with
> something that is not their main focus for Ubuntu's future
> results in "adequate but unpolished" at best, and "long
> duration breakage" at worst.
> 
> Personally, I recommend establishing a strong communication
> with upstream GNOME and preparing the field to report bugs
> and needs involving Wayland. I also emphasize the need to
> either move away from Ubuntu or be prepared to become the
> GTK QA team that one can't really expect Ubuntu to be anymore.
> Perhaps maintaining PPAs for the entire stack.
> 
> 
> Em Qua, 2013-07-10 às 09:27 -0700, Daniel Foré escreveu:
>> Hey everyone,
>> 
>> 
>> So here’s the distilled conversation we had with Jono:
>> 
>> 
>> 1. Canonical wants Mir to be used by more than just Ubuntu. More users
>> tends to mean more developers which means a better product. Jono
>> didn't say it directly, but I think it's obvious that their interest
>> in us is mainly about bringing more attention/support to Mir (which is
>> reasonable/logical).
>> 
>> 
>> 2. With Mir, they’re building a focused display server that
>> prioritizes a convergent cross-form-factor design. Jono referred to
>> Mir as being "thin" and states that because Mir needs to run on a
>> phone it has very strict performance requirements.
>> 
>> 
>> 3. Canonical also has tools built around Mir that could enable us (and
>> others) to easily “flash” our builds onto mobile devices using an
>> Android/Mir base. This is probably the most unique/enticing part about
>> using Mir.
>> 
>> 
>> 4. Jono wants to reach out to Ubuntu flavors and derivatives to see
>> how Mir affects them and how Ubuntu’s engineering team can help ease
>> the transition to Mir. For 14.04, we theoretically could run Pantheon
>> on XMir which means no meaningful change for us to be able to run with
>> a Mir system compositor.
>> 
>> 
>> Obviously there are pros and cons. The biggest con is that it is
>> highly unlikely Canonical will put staff hours into make sure stuff
>> like Gtk+ and Clutter work on Mir. This work is essential to
>> elementary as it stands and blocks us from running natively on Mir. In
>> the spirit of the above #3, Jono has assured us that if we're
>> interested in doing the work to make our toolkit work with Mir that
>> Canonical engineers would be available to answer any questions we may
>> have.
>> 
>> 
>> So we seemingly have these options moving forward:
>> 
>> 
>> 1. Port everything to Qt (including Gala the rest of our shell, all
>> our apps, etc) and use Mir.
>> 
>> 
>> 2. Port Clutter/Gtk+/etc to Mir.
>> 
>> 
>> 3. Port everything to Wayland (significantly less work than the former
>> two options imo).
>> 
>> 
>> 4. Do nothing and use either XMir or XWayland until those are
>> deprecated at which point we need to port to one or the other.
>> 
>> 
>> I think at this point, it's starting to look more and more like
>> Wayland is going to be the path of least resistance for us. I can't
>> imagine we have the development power to try and maintain Gtk
>> +/Clutter/Mutter on Mir. But it may turn out that sometime between now
>> and 14.04 some group decides to do this work and then we're back to a
>> more difficult choice.
>> 
>> 
>> 
>> All in all, the conversation leaves us with more talking points and
>> more question than answers :p I think the only thing we know 100%
>> (even if we do have some other opinions) is that X is dying. We need
>> to make an effort to remove any x-specific code from our apps and our
>> shell and to move away from any libraries that we know won’t exist in
>> a post-X world (like BAMF

Re: [Elementary-dev-community] Mir Discussion with Jono

[Cassidy James]

While Jono is not the engineering lead for Mir, he said that he does 
not expect Canonical to invest any time into making GTK work on Mir. He 
did, however, say that he would bring it up with the engineering team 
as a concern of ours.


On Wed, Jul 10, 2013 at 2:25 PM, Jo-Erlend Schinstad 
 wrote:



On 10 July 2013 18:27, Daniel Foré  wrote:

Hey everyone,

So here’s the distilled conversation we had with Jono:



[snip]
 
Obviously there are pros and cons. The biggest con is that it is 
highly unlikely Canonical will put staff hours into make sure stuff 
like Gtk+ and Clutter work on Mir. This work is essential to 
elementary as it stands and blocks us from running natively on Mir.




[snip]

Did Jono explain why this is unlikely, or is that your speculation? 
Because it sounds contrary to the recent Mir interview. It also seems 
kind of weird that a desktop distribution would want to make such an 
enormous amount of applications second class citizens in this way.


It seems unlikely that they'll add GTK+ as a language for Ubuntu SDK 
anytime soon, but that's a completely different thing and wouldn't 
affect Elementary in any case. I would certainly expect Canonical to 
add a GTK+ backend for Mir. Otherwise, they'll have close to no hope 
of making Mir a successful display server. 

More information about this would be very interesting read. 


Jo-Erlend Schinstad

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Jo-Erlend Schinstad]

On 10 July 2013 18:27, Daniel Foré  wrote:

> Hey everyone,
>
>
> So here’s the distilled conversation we had with Jono:
>

[snip]


> Obviously there are pros and cons. The biggest con is that it is highly
> unlikely Canonical will put staff hours into make sure stuff like Gtk+ and
> Clutter work on Mir. This work is essential to elementary as it stands and
> blocks us from running natively on Mir.
>

[snip]

Did Jono explain why this is unlikely, or is that your speculation? Because
it sounds contrary to the recent Mir interview. It also seems kind of weird
that a desktop distribution would want to make such an enormous amount of
applications second class citizens in this way.

It seems unlikely that they'll add GTK+ as a language for Ubuntu SDK
anytime soon, but that's a completely different thing and wouldn't affect
Elementary in any case. I would certainly expect Canonical to add a GTK+
backend for Mir. Otherwise, they'll have close to no hope of making Mir a
successful display server.

More information about this would be very interesting read.

Jo-Erlend Schinstad
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Manish Sinha]

On Wed, Jul 10, 2013 at 10:40 AM, Conscious User
 wrote:
> Developers from elementaryOS should not be the main force
> behind porting toolkits to display servers. This is either
> the responsibility of toolkit developers or display server
> developers. If anything, for the deeper technical knowledge
> this depends on.

Adding to this. Elementary taking up doing so many things might
end up like how Ubuntu at the moment. Too many downstream
changes languishing and being un-maintained. Overlay scroll-bars
have bugs, compiz is borked.

There might be some vaild reasons Canonical is doing this, but to
me it looks like they have bitten more than they can chew. Hopefully
elementary doesn't do the same and reuse as much upstream
as possible.

I am not saying that Elemenentary should not contribute to this
feature but they should think twice before doing this because
they do not develop GTK or Wayland.

(Not sure if my response came off sounding like a pessimist)

-
Manish

-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Sergey "Shnatsel" Davidoff]

Oh, I forgot: Mir and Mer are totally different things. Not a typo!

http://merproject.org/

FYI, it's with Wayland and Mer-based Sailfish OS is debuting this year's
end. *drools*


2013/7/10 Sergey "Shnatsel" Davidoff 

> 3. Canonical also has tools built around Mir that could enable us (and
>> others) to easily “flash” our builds onto mobile devices using an
>> Android/Mir base. This is probably the most unique/enticing part about
>> using Mir.
>
>
> Mer also provides this, and more. This is yet another thing that they
> "borrowed" from Mer/Sailfish and now are advertising as their own creation
> ("yet another" because the whole "nextgen display server on Android
> drivers" thing was developed by Sailfish developer for Wayland and Mer).
>
>
>> I think the only thing we know 100% (even if we do have some other
>> opinions) is that X is dying. We need to make an effort to remove any
>> x-specific code from our apps and our shell and to move away from any
>> libraries that we know won’t exist in a post-X world (like BAMF and WNCK).
>>
>
> The problem with this is that we cannot just move *away* and forget about
> the functionality these libraries provide, we have move *to* something
> else as well to provide that functionality. And it's not clear where we
> should move until we make the display server choice.
>
> The thing I'm 100% sure of is that we're sticking to X until at least
> Ubuntu 14.04. It doesn't matter for applications if it will be with or
> without XMir as system compositor, they're still interfacing with X.
>
> We have a conservative and iterative cycle ahead of us. Let's wait and see
> which of the display servers prevails, and make an *educated* decision
> for L+2. We know very little about what either of the display severs will
> be in a year from now, so any decision we make now is not a decision but
> prejudice.
>
> --
> Sergey "Shnatsel" Davidoff
> OS architect @ elementary
>



-- 
Sergey "Shnatsel" Davidoff
OS architect @ elementary
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Sergey "Shnatsel" Davidoff]

>
> 3. Canonical also has tools built around Mir that could enable us (and
> others) to easily “flash” our builds onto mobile devices using an
> Android/Mir base. This is probably the most unique/enticing part about
> using Mir.


Mer also provides this, and more. This is yet another thing that they
"borrowed" from Mer/Sailfish and now are advertising as their own creation
("yet another" because the whole "nextgen display server on Android
drivers" thing was developed by Sailfish developer for Wayland and Mer).


> I think the only thing we know 100% (even if we do have some other
> opinions) is that X is dying. We need to make an effort to remove any
> x-specific code from our apps and our shell and to move away from any
> libraries that we know won’t exist in a post-X world (like BAMF and WNCK).
>

The problem with this is that we cannot just move *away* and forget about
the functionality these libraries provide, we have move *to* something else
as well to provide that functionality. And it's not clear where we should
move until we make the display server choice.

The thing I'm 100% sure of is that we're sticking to X until at least
Ubuntu 14.04. It doesn't matter for applications if it will be with or
without XMir as system compositor, they're still interfacing with X.

We have a conservative and iterative cycle ahead of us. Let's wait and see
which of the display servers prevails, and make an *educated* decision for
L+2. We know very little about what either of the display severs will be in
a year from now, so any decision we make now is not a decision but
prejudice.

-- 
Sergey "Shnatsel" Davidoff
OS architect @ elementary
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion with Jono

[Conscious User]

Developers from elementaryOS should not be the main force
behind porting toolkits to display servers. This is either
the responsibility of toolkit developers or display server
developers. If anything, for the deeper technical knowledge
this depends on.

Reactions from the GNOME community indicate that they have
no intention of supporting Mir, so this leaves Canonical as
the only option. Historically, trusting Canonical with
something that is not their main focus for Ubuntu's future
results in "adequate but unpolished" at best, and "long
duration breakage" at worst.

Personally, I recommend establishing a strong communication
with upstream GNOME and preparing the field to report bugs
and needs involving Wayland. I also emphasize the need to
either move away from Ubuntu or be prepared to become the
GTK QA team that one can't really expect Ubuntu to be anymore.
Perhaps maintaining PPAs for the entire stack.


Em Qua, 2013-07-10 às 09:27 -0700, Daniel Foré escreveu:
> Hey everyone,
> 
> 
> So here’s the distilled conversation we had with Jono:
> 
> 
> 1. Canonical wants Mir to be used by more than just Ubuntu. More users
> tends to mean more developers which means a better product. Jono
> didn't say it directly, but I think it's obvious that their interest
> in us is mainly about bringing more attention/support to Mir (which is
> reasonable/logical).
> 
> 
> 2. With Mir, they’re building a focused display server that
> prioritizes a convergent cross-form-factor design. Jono referred to
> Mir as being "thin" and states that because Mir needs to run on a
> phone it has very strict performance requirements.
> 
> 
> 3. Canonical also has tools built around Mir that could enable us (and
> others) to easily “flash” our builds onto mobile devices using an
> Android/Mir base. This is probably the most unique/enticing part about
> using Mir.
> 
> 
> 4. Jono wants to reach out to Ubuntu flavors and derivatives to see
> how Mir affects them and how Ubuntu’s engineering team can help ease
> the transition to Mir. For 14.04, we theoretically could run Pantheon
> on XMir which means no meaningful change for us to be able to run with
> a Mir system compositor.
> 
> 
> Obviously there are pros and cons. The biggest con is that it is
> highly unlikely Canonical will put staff hours into make sure stuff
> like Gtk+ and Clutter work on Mir. This work is essential to
> elementary as it stands and blocks us from running natively on Mir. In
> the spirit of the above #3, Jono has assured us that if we're
> interested in doing the work to make our toolkit work with Mir that
> Canonical engineers would be available to answer any questions we may
> have.
> 
> 
> So we seemingly have these options moving forward:
> 
> 
> 1. Port everything to Qt (including Gala the rest of our shell, all
> our apps, etc) and use Mir.
> 
> 
> 2. Port Clutter/Gtk+/etc to Mir.
> 
> 
> 3. Port everything to Wayland (significantly less work than the former
> two options imo).
> 
> 
> 4. Do nothing and use either XMir or XWayland until those are
> deprecated at which point we need to port to one or the other.
> 
> 
> I think at this point, it's starting to look more and more like
> Wayland is going to be the path of least resistance for us. I can't
> imagine we have the development power to try and maintain Gtk
> +/Clutter/Mutter on Mir. But it may turn out that sometime between now
> and 14.04 some group decides to do this work and then we're back to a
> more difficult choice.
> 
> 
> 
> All in all, the conversation leaves us with more talking points and
> more question than answers :p I think the only thing we know 100%
> (even if we do have some other opinions) is that X is dying. We need
> to make an effort to remove any x-specific code from our apps and our
> shell and to move away from any libraries that we know won’t exist in
> a post-X world (like BAMF and WNCK).
> 
> 
> I’ve created a blueprint where we can track our progress in ditching
> X: https://blueprints.launchpad.net/elementaryos/+spec/ditch-x
> 
> 
> Please make sure to file bugs against projects you know contain bits
> that rely on X and link them to this blueprint. The better we asses
> the situation, the easier it'll be to make the transition.
> 
> 
> 
> Best Regards,
> 
> Daniel Foré
> 
> 
> elementaryos.org


-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

[Elementary-dev-community] Mir Discussion with Jono

[Daniel Foré]

Hey everyone,


So here’s the distilled conversation we had with Jono:


1. Canonical wants Mir to be used by more than just Ubuntu. More users
tends to mean more developers which means a better product. Jono didn't say
it directly, but I think it's obvious that their interest in us is mainly
about bringing more attention/support to Mir (which is reasonable/logical).


2. With Mir, they’re building a focused display server that prioritizes a
convergent cross-form-factor design. Jono referred to Mir as being "thin"
and states that because Mir needs to run on a phone it has very strict
performance requirements.


3. Canonical also has tools built around Mir that could enable us (and
others) to easily “flash” our builds onto mobile devices using an
Android/Mir base. This is probably the most unique/enticing part about
using Mir.


4. Jono wants to reach out to Ubuntu flavors and derivatives to see how Mir
affects them and how Ubuntu’s engineering team can help ease the transition
to Mir. For 14.04, we theoretically could run Pantheon on XMir which means
no meaningful change for us to be able to run with a Mir system compositor.


Obviously there are pros and cons. The biggest con is that it is highly
unlikely Canonical will put staff hours into make sure stuff like Gtk+ and
Clutter work on Mir. This work is essential to elementary as it stands and
blocks us from running natively on Mir. In the spirit of the above #3, Jono
has assured us that if we're interested in doing the work to make our
toolkit work with Mir that Canonical engineers would be available to answer
any questions we may have.


So we seemingly have these options moving forward:


1. Port everything to Qt (including Gala the rest of our shell, all our
apps, etc) and use Mir.


2. Port Clutter/Gtk+/etc to Mir.


3. Port everything to Wayland (significantly less work than the former two
options imo).


4. Do nothing and use either XMir or XWayland until those are deprecated at
which point we need to port to one or the other.


I think at this point, it's starting to look more and more like Wayland is
going to be the path of least resistance for us. I can't imagine we have
the development power to try and maintain Gtk+/Clutter/Mutter on Mir. But
it may turn out that sometime between now and 14.04 some group decides to
do this work and then we're back to a more difficult choice.


All in all, the conversation leaves us with more talking points and more
question than answers :p I think the only thing we know 100% (even if we do
have some other opinions) is that X is dying. We need to make an effort to
remove any x-specific code from our apps and our shell and to move away
from any libraries that we know won’t exist in a post-X world (like BAMF
and WNCK).


I’ve created a blueprint where we can track our progress in ditching X:
https://blueprints.launchpad.net/elementaryos/+spec/ditch-x


Please make sure to file bugs against projects you know contain bits that
rely on X and link them to this blueprint. The better we asses the
situation, the easier it'll be to make the transition.

Best Regards,

Daniel Foré

elementaryos.org
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp

Re: [Elementary-dev-community] Mir Discussion

[Daniel Foré]

Sergey,

Like Cassidy said, I'm sure this will be a much more political discussion than 
a technical one. Obviously I'm not going into the call with the intention to 
personally make a decision about Mir vs Wayland. 

This is an opportunity to speak to someone at Canonical directly and get some 
insight into not only why we might want to use Mir, but why they wanted to use 
Mir. Don't forget that Canonical was also in the Wayland camp for quite a while.

And while I don't expect Jono to have vast technical knowledge on the subject, 
I've met him in person and he's a competent fellow. It's part of his job 
description to have a high level overview of just about everything in Ubuntu 
(including the underlying tech). So I imagine he should be able to answer any 
technical questions we might have.

That said, I'm not exactly a dunce myself either ;)

Best Regards,
Daniel Foré

El jul 8, 2013, a las 1:54 a.m., "Sergey \"Shnatsel\" Davidoff" 
 escribió:

>> I was recently contacted by Jono Bacon from Canonical. We'll be on a call 
>> Tuesday to talk about the pros and cons of Mir and whether or not it fits in 
>> with elementary.
>> 
>> If you have any questions or concerns that you'd like me to bring up with 
>> Jono, please let me know.
>> 
>> I'll be sure to take good notes on his responses so that we can make an 
>> informed decision about how to move forward.
> 
> I really appreciate the move, but I'd expect discussions about low-level 
> components like display servers which users don't even notice to take place 
> between people actually competent with the tech, not a community manager on 
> one side and a UX designer on the other. Perhaps a conference call or a G+ 
> hangout is more appropriate, with more people on our side at least.
> 
> That said, here's what I gather regarding Mir so far.
> 
> The first thing to understand about Mir in 13.10 and 14.04 is that it 
> functions as a system compositor. So you still have all your traditional X 
> sessions just like in Precise, but on top of them you have yet another 
> compositor (Mir) which provides smooth transitions between sessions (instead 
> of ugly VT switching) and lets us utilize LightDM as lock screen.
> 
> Using Mir as a system compositor and using it as the only display server are 
> completely different things. For 14.04 we're talking about using it as a 
> system compositor only. The primary concerns with using it as a system 
> compositor are performance and stability - this is why Kubuntu rejected Mir, 
> for example.
> 
> We have some preliminary benchmarks of Mir as a system compositor, but 
> they're not really meaningful because vertical synchronization was disabled 
> and they're just measuring how much raw GPU power another compositing layer 
> gobbles up. But raw GPU power is just about the least of our concerns. 
> Compositor synchronization is a much trickier problem - read some articles on 
> the topic at http://blog.fishsoup.net/ to get an idea of how compositors work 
> and just how tricky synchronization is.
> 
> Synchronizing the application with just one compositor is problematic enough, 
> but it's mostly solved thanks to GNOME's heavy and continuous investment it 
> it. But synchronizing the whole thing with yet another compositor on top of 
> it that has completely different protocols and inner workings - that's not 
> something I believe to be possible at all. And due to poor synchronization 
> we'll get a bouquet of issues like jitter (disgusting video playback, yay!) 
> and high latency.
> 
> Don't get me wrong, having a system compositor is actually a good idea, given 
> a suitable implementation. It's very perspective in the Wayland world because 
> the Wayland protocol is designed to allow nesting compositors - you can stack 
> as many Wayland compositors as you want, and as long as a compositor does not 
> perform any operations with the frame it receives from an underlying 
> compositor, it can pass the frame to the next compositor or display it 
> without introducing any additional overhead. So you effectively have just one 
> compositor in the process and the second one kicks in only to draw 
> transitions between sessions.
> 
> This still holds if you throw some X applications on XWayland into the mix, 
> because they're run in tiny rootless per-app X servers without their own 
> compositors. This is not the case for what Canonical is up to, though - right 
> now they still have a compositor in X and a Mir compositor on top of that, so 
> you inevitably pass every frame through two completely different compositors 
> that cannot even talk to each other, so there's like no synchronization 
> between them whatsoever?
> 
> With all due respect to Jono, I'm not sure he's sufficiently qualified to 
> understand the details of inner workings of compositors and explain them to 
> us. (Hell, I'm not even sure if I'm sufficiently qualified to understand the 
> explanations!)
> 
> Regarding ditching X for Mir, I believe t

Re: [Elementary-dev-community] Mir Discussion

[Cassidy James]

Hey Sergey,

You're right in the notion of Ubuntu's community manager and the
founder/leader of elementary not being the best people to discuss Mir on a
strictly technical level. That alone tells me to expect the discussion to
be more high-level and of a political/business nature.

We already expect to ask the, "Why would we use this instead of Wayland?"
question. That's a given. If you have any other specific questions that we
could ask, let us know.

Regards,
Cassidy James
On Jul 8, 2013 3:54 AM, "Sergey "Shnatsel" Davidoff" <
ser...@elementaryos.org> wrote:

> I was recently contacted by Jono Bacon from Canonical. We'll be on a call
>> Tuesday to talk about the pros and cons of Mir and whether or not it fits
>> in with elementary.
>>
>> If you have any questions or concerns that you'd like me to bring up with
>> Jono, please let me know.
>>
>> I'll be sure to take good notes on his responses so that we can make an
>> informed decision about how to move forward.
>
>
> I really appreciate the move, but I'd expect discussions about low-level
> components like display servers which users don't even notice to take place
> between people actually competent with the tech, not a community manager on
> one side and a UX designer on the other. Perhaps a conference call or a G+
> hangout is more appropriate, with more people on our side at least.
>
> That said, here's what I gather regarding Mir so far.
>
> The first thing to understand about Mir in 13.10 and 14.04 is that it
> functions as a *system* compositor. So you still have all your
> traditional X sessions just like in Precise, but on top of them you have *yet
> another* compositor (Mir) which provides smooth transitions between
> sessions (instead of ugly VT switching) and lets us utilize LightDM as lock
> screen.
>
> Using Mir as a system compositor and using it as the only display server
> are completely different things. For 14.04 we're talking about using it as
> a system compositor only. The primary concerns with using it as a system
> compositor are performance and stability - this is why Kubuntu rejected
> Mir,
> for example.
>
> We have some 
> preliminary
> benchmarksof
>  Mir as a system compositor, but they're not really meaningful because
> vertical synchronization was disabled and they're just measuring how much
> raw GPU power another compositing layer gobbles up. But raw GPU power is
> just about the least of our concerns. Compositor synchronization is a much
> trickier problem - read some articles on the topic at
> http://blog.fishsoup.net/ to get an idea of how compositors work and just
> how tricky synchronization is.
>
> Synchronizing the application with just one compositor is problematic
> enough, but it's mostly solved thanks to GNOME's heavy and continuous
> investment it it. But synchronizing the whole thing with yet another
> compositor on top of it that has completely different protocols and inner
> workings - that's not something I believe to be possible at all. And due to
> poor synchronization we'll get a bouquet of issues like jitter (disgusting
> video playback, yay!) and high latency.
>
> Don't get me wrong, having a system compositor is actually a good idea,
> given a suitable implementation. It's very perspective in the Wayland world
> because the Wayland protocol is designed to allow nesting compositors - you
> can stack as many Wayland compositors as you want, and as long as a
> compositor does not perform any operations with the frame it receives from
> an underlying compositor, it can pass the frame to the next compositor or
> display it without introducing any additional overhead. So you effectively
> have just one compositor in the process and the second one kicks in only to
> draw transitions between sessions.
>
> This still holds if you throw some X applications on XWayland into the
> mix, because they're run in tiny rootless per-app X servers without their
> own compositors. This is not the case for what Canonical is up to, though -
> right now they still have a compositor in X and a Mir compositor on top of
> that, so you inevitably pass every frame through two completely different
> compositors that cannot even talk to each other, so there's like no
> synchronization between them whatsoever?
>
> With all due respect to Jono, I'm not sure he's sufficiently qualified to
> understand the details of inner workings of compositors and explain them to
> us. (Hell, I'm not even sure if I'm sufficiently qualified to understand
> the explanations!)
>
> Regarding ditching X for Mir, I believe this is out of question. We
> already have Wayland support for Gala for free and porting it to Mir is not
> simply a waste of effort - it's a continuous waste of effort because Mir
> protocol is not stable or froz

Re: [Elementary-dev-community] Mir Discussion

[Sergey "Shnatsel" Davidoff]

>
> I was recently contacted by Jono Bacon from Canonical. We'll be on a call
> Tuesday to talk about the pros and cons of Mir and whether or not it fits
> in with elementary.
>
> If you have any questions or concerns that you'd like me to bring up with
> Jono, please let me know.
>
> I'll be sure to take good notes on his responses so that we can make an
> informed decision about how to move forward.


I really appreciate the move, but I'd expect discussions about low-level
components like display servers which users don't even notice to take place
between people actually competent with the tech, not a community manager on
one side and a UX designer on the other. Perhaps a conference call or a G+
hangout is more appropriate, with more people on our side at least.

That said, here's what I gather regarding Mir so far.

The first thing to understand about Mir in 13.10 and 14.04 is that it
functions as a *system* compositor. So you still have all your traditional
X sessions just like in Precise, but on top of them you have *yet
another*compositor (Mir) which provides smooth transitions between
sessions
(instead of ugly VT switching) and lets us utilize LightDM as lock screen.

Using Mir as a system compositor and using it as the only display server
are completely different things. For 14.04 we're talking about using it as
a system compositor only. The primary concerns with using it as a system
compositor are performance and stability - this is why Kubuntu
rejected 
Mir,
for example.

We have some 
preliminary
benchmarksof
Mir as a system compositor, but they're not really meaningful because
vertical synchronization was disabled and they're just measuring how much
raw GPU power another compositing layer gobbles up. But raw GPU power is
just about the least of our concerns. Compositor synchronization is a much
trickier problem - read some articles on the topic at
http://blog.fishsoup.net/ to get an idea of how compositors work and just
how tricky synchronization is.

Synchronizing the application with just one compositor is problematic
enough, but it's mostly solved thanks to GNOME's heavy and continuous
investment it it. But synchronizing the whole thing with yet another
compositor on top of it that has completely different protocols and inner
workings - that's not something I believe to be possible at all. And due to
poor synchronization we'll get a bouquet of issues like jitter (disgusting
video playback, yay!) and high latency.

Don't get me wrong, having a system compositor is actually a good idea,
given a suitable implementation. It's very perspective in the Wayland world
because the Wayland protocol is designed to allow nesting compositors - you
can stack as many Wayland compositors as you want, and as long as a
compositor does not perform any operations with the frame it receives from
an underlying compositor, it can pass the frame to the next compositor or
display it without introducing any additional overhead. So you effectively
have just one compositor in the process and the second one kicks in only to
draw transitions between sessions.

This still holds if you throw some X applications on XWayland into the mix,
because they're run in tiny rootless per-app X servers without their own
compositors. This is not the case for what Canonical is up to, though -
right now they still have a compositor in X and a Mir compositor on top of
that, so you inevitably pass every frame through two completely different
compositors that cannot even talk to each other, so there's like no
synchronization between them whatsoever?

With all due respect to Jono, I'm not sure he's sufficiently qualified to
understand the details of inner workings of compositors and explain them to
us. (Hell, I'm not even sure if I'm sufficiently qualified to understand
the explanations!)

Regarding ditching X for Mir, I believe this is out of question. We already
have Wayland support for Gala for free and porting it to Mir is not simply
a waste of effort - it's a continuous waste of effort because Mir protocol
is not stable or frozen, unlike Wayland protocol. And I doubt we have the
resources to do so in the first place anyway.

Also, I've never seen an explanation of why Mir is superior to Wayland.
There's a wiki 
page"explaining"
that but the only real argument in there was erroneous and got
removed over time. I'm really, really curious what Jono has to say on the
topic.

-- 
Sergey "Shnatsel" Davidoff
OS architect @ elementary
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/Li

[Elementary-dev-community] Mir Discussion

[Daniel Foré]

Hey everyone,

I was recently contacted by Jono Bacon from Canonical. We'll be on a call 
Tuesday to talk about the pros and cons of Mir and whether or not it fits in 
with elementary.

If you have any questions or concerns that you'd like me to bring up with Jono, 
please let me know.

I'll be sure to take good notes on his responses so that we can make an 
informed decision about how to move forward. 

Best Regards,
Daniel Foré
-- 
Mailing list: https://launchpad.net/~elementary-dev-community
Post to : elementary-dev-community@lists.launchpad.net
Unsubscribe : https://launchpad.net/~elementary-dev-community
More help   : https://help.launchpad.net/ListHelp