cshalle wrote: --- Improving touch screen support is a goal for both GNOME and Fedora, simply because laptops are heading that way and if we offer something that doesn't even try to use the touchscreen where it makes sense, users will go elsewhere. ---
Linux based solutions oriented towards tablets, hybrids and touch-screen-laptops already exist: Chrome OS and Android. Realistically nobody in the Linux community has the resources that Google does or the mainstream market-share. Everyone knows the obvious, Google will dominate that space. In the technology press everyone is saying the "Year of the Linux Desktop" will be because of Google and Chrome OS as well as Android. Please, don't waste your time and resources, leave this use-case alone.
Red Hat and Fedora as well as similar platforms thrive in the more traditional desktop space and specifically workstations. If you walk into Pixar, NASA or CERN you'll find Fedora/RHEL on workstations and servers. Look at Ubuntu at Google or Facebook. If I call up any major OEM I can get workstations with two different operating systems: Windows 7 and RHEL. Not surprisingly the product is called Fedora Workstation.
There's a theme at play here that if you focus 100% on traditional workstations you can having a winning product that will even compete on equal footing against Mac OS and Windows.
Linux distributions and desktop projects have been chasing the hybrid and tablet and it's had the effect of throwing their workstation users off of the bus. When Ubuntu and Fedora abandoned Gnome 2 and focused on Unity and Gnome 3 this caused major disruption and chaos. Around the same time many developers and engineers ended up going with Mac OS because Apple provided a conservative traditional desktop experience that was highly polished and professional.
Last night I was looking at Gnome Shell extensions and I realized that most of them had the effect of turning Gnome 3 back into Gnome 2. The fact is that the community and specifically the workstation use-case is desperately seeking a Gnome 2 replacement that's why we have XFCE, LXDE and Cinnamon. This is why MATE exists and is rising in popularity. The Linux community still loves Gnome 2 and wants it back.
This is why I propose a compromise that will work for both Gnome and Fedora Workstation:
Have both Gnome 2 (MATE) and Gnome 3 (Gnome Shell) be parallel but related branches of the same Gnome product.
1. Have the Gnome Foundation adopt MATE as a Gnome 2 legacy project. Provide development and support resources to accelerate MATE's efforts to transition to GTK3, systemd and Wayland. Make sure that both Gnome 2 and Gnome 3 are based on the same modern infrastructure.
2. Modify Mutter so that it can become the official compositor of MATE and replace the practice of bundling Gnome 2 with Compiz which is now an Ubuntu product. This would ensure that Gnome 2 and Gnome 3 have similar look, behavior and feel. Another option is to use Compton but that could be seen as a short-term fix until Mutter was fully integrated into Gnome 2.
3. Keep Gnome 3 as is in the present and don't interfere with that project or dictate design to them. Gnome 3 will exist as a development project focused on innovation, experimentation and creativity. Their focus would continue to be on pushing desktop boundaries and exploring alternative paradigms. If appropriate, innovations developed in the Gnome Shell would be occasionally fed back into Gnome 2. This will create a healthy Gnome innovation cycle.
4. Make Gnome 2 the default desktop for Fedora Workstation with Fedora branding and themes as well as the current Gnome default applications. Have Gnome 3 be an optional extra at installation. Also offer KDE as well for diversity.
5. Promote Gnome 3 to Gnome 2 users. When the user runs Gnome 2 for the first time have a prompt that says "Would you like to see the future? Try out Gnome 3". And it would be installed side-by-side with Gnome 2.
That way the traditional desktop can be addressed by Gnome 2 (MATE) and Fedora Workstation doesn't have to interfere and disrupt the activities happening over at Gnome with Gnome 3 (Gnome Shell). Both projects can happily cooperate and coexist. This sort of collaboration is what Linux is famous for and Fedora Workstation should epitomize.
----- Original Message -----
From: "Alex GS" alxgrtnstrngl@gmail.com To: desktop@lists.fedoraproject.org Sent: Monday, February 3, 2014 3:09:43 PM Subject: Re: Underlying DE for the Workstation product
cshalle wrote:
Improving touch screen support is a goal for both GNOME and Fedora, simply because laptops are heading that way and if we offer something that doesn't even try to use the touchscreen where it makes sense, users will go elsewhere.
Linux based solutions oriented towards tablets, hybrids and touch-screen-laptops already exist: Chrome OS and Android. Realistically nobody in the Linux community has the resources that Google does or the mainstream market-share. Everyone knows the obvious, Google will dominate that space. In the technology press everyone is saying the "Year of the Linux Desktop" will be because of Google and Chrome OS as well as Android. Please, don't waste your time and resources, leave this use-case alone.
Nobody in the Linux community got the resources that Microsoft or Apple does either, yet here we are. And the reason we are looking at it is that even the technical workstation users are going to be at least partly running laptops and will be looking for some basic support. We are not looking at turning the system into a pure touch screen experience, rather add some touch functionality where it makes sense.
Red Hat and Fedora as well as similar platforms thrive in the more traditional desktop space and specifically workstations. If you walk into Pixar, NASA or CERN you'll find Fedora/RHEL on workstations and servers. Look at Ubuntu at Google or Facebook. If I call up any major OEM I can get workstations with two different operating systems: Windows 7 and RHEL. Not surprisingly the product is called Fedora Workstation.
I am well aware of this, I meet or call with a lot of these customers on a regular basis. And I also spend time working with them to prepare their transition to GNOME 3 as part of their RHEL 7 transition.
There's a theme at play here that if you focus 100% on traditional workstations you can having a winning product that will even compete on equal footing against Mac OS and Windows.
Linux distributions and desktop projects have been chasing the hybrid and tablet and it's had the effect of throwing their workstation users off of the bus. When Ubuntu and Fedora abandoned Gnome 2 and focused on Unity and Gnome 3 this caused major disruption and chaos. Around the same time many developers and engineers ended up going with Mac OS because Apple provided a conservative traditional desktop experience that was highly polished and professional.
Last night I was looking at Gnome Shell extensions and I realized that most of them had the effect of turning Gnome 3 back into Gnome 2. The fact is that the community and specifically the workstation use-case is desperately seeking a Gnome 2 replacement that's why we have XFCE, LXDE and Cinnamon. This is why MATE exists and is rising in popularity. The Linux community still loves Gnome 2 and wants it back.
This is why I propose a compromise that will work for both Gnome and Fedora Workstation:
Have both Gnome 2 (MATE) and Gnome 3 (Gnome Shell) be parallel but related branches of the same Gnome product.
<SNIP>
I am generally negative to any such solutions as they tend to suck resources away from advancing something over to trying to keep multiple options sorta working together. Any solution one choose at any given point in time will have holes or missing functionality. I don't think the best solution is ever to start filling those gaps with complete alternate solutions, as they will just be bringing in their own set of holes and bugs and in the end you are not moving forward anymore, you are just jumping around trying to avoid regressing and trying to plug security issues.
Christian
On 02/03/2014 03:23 PM, Christian Schaller wrote:
I am generally negative to any such solutions as they tend to suck resources away from advancing something over to trying to keep multiple options sorta working together. Any solution one choose at any given point in time will have holes or missing functionality. I don't think the best solution is ever to start filling those gaps with complete alternate solutions, as they will just be bringing in their own set of holes and bugs and in the end you are not moving forward anymore, you are just jumping around trying to avoid regressing and trying to plug security issues.
Which would not be a problem if the workstation group defines an solid criteria that is required to be meet before becoming a workstation product.
You must realize by Alex statements ( as well as the fact the several alternatives of desktop environments exist in the first place ) that end user will chose what *he* thinks what works best for him and his workflow not what *we* think is best for him or what *we* think is workflow is which in turns shows in the end of the day it's better for us to provide a larger inventory of products since it will increase the odds that the end user will find something *he* likes and *can use* for *himself* with us.
JBG
On Mon, 2014-02-03 at 15:35 +0000, "Jóhann B. Guðmundsson" wrote:
You must realize by Alex statements ( as well as the fact the several alternatives of desktop environments exist in the first place ) that end user will chose what *he* thinks what works best for him and his workflow not what *we* think is best for him or what *we* think is workflow is which in turns shows in the end of the day it's better for us to provide a larger inventory of products since it will increase the odds that the end user will find something *he* likes and *can use* for *himself* with us.
That's only true for a certain subset of users (users like you, pretty tech savy and with strong opinions about the features and setting of your environment).
However most people don't choose much. The OEM manufacturers chose for them, and eventually the only choice you can make is between a Mac and a Windows PC. They just want _a computer that works_, they don't care about what's in there as long as they can install the applications they care about and they can perform the tasks they need the computer for.
The bottom line is this, when you don't know the difference between two options, you can't possibly make an informed decision, and you rely on other people to make that choice for you (should I buy a Mac or a PC? Android or iPhone?).
From that perspective, letting the user choose seems pretty much like
asking them to stick to what they know (pressummably a propietary OS).
On the other hand, we can't possibly offer a well defined set of integrated features if we don't have a unified, single offering. It is simply unrealistic to support every single DE and let the user choose.
Because at that point, at the face of choosing on which one we implement things first, we've no knowledge on where should we focus. But possibly worse, when people out there want to target their apps for Fedora, they don't know what environment should they be aiming for as they all have different interface guidelines.
This is in essence the difference between a project and a product. Projects tend to be an inclusive set for people to cooperate, however, products are on the other hand opinionated and restricted by nature to be able to serve to its audience.
Linux was never about choice[0] [0]http://www.redhat.com/archives/fedora-devel-list/2008-January/msg00861.html
On 02/03/2014 04:08 PM, Alberto Ruiz wrote:
However most people don't choose much. The OEM manufacturers chose for them, and eventually the only choice you can make is between a Mac and a Windows PC. They just want_a computer that works_, they don't care about what's in there as long as they can install the applications they care about and they can perform the tasks they need the computer for.
Yes and Gnome has *never* been and never will be the desktop environment that can provide that role and neither have are we as a distribution trying to chase down those users.
Gnome has been in beta state since RHL 5 constantly throwing half implement half integrating application over the wall to see what sticks so honestly I dont know what you are trying to get at here.
With the exception of the core/base layer Linux is about choice and needs that choice to advance itself.
JBG
----- Original Message -----
From: "Jóhann B. Guðmundsson" johannbg@gmail.com To: desktop@lists.fedoraproject.org Sent: Monday, February 3, 2014 4:35:41 PM Subject: Re: Underlying DE for the Workstation product
On 02/03/2014 03:23 PM, Christian Schaller wrote:
I am generally negative to any such solutions as they tend to suck resources away from advancing something over to trying to keep multiple options sorta working together. Any solution one choose at any given point in time will have holes or missing functionality. I don't think the best solution is ever to start filling those gaps with complete alternate solutions, as they will just be bringing in their own set of holes and bugs and in the end you are not moving forward anymore, you are just jumping around trying to avoid regressing and trying to plug security issues.
Which would not be a problem if the workstation group defines an solid criteria that is required to be meet before becoming a workstation product.
You must realize by Alex statements ( as well as the fact the several alternatives of desktop environments exist in the first place ) that end user will chose what *he* thinks what works best for him and his workflow not what *we* think is best for him or what *we* think is workflow is which in turns shows in the end of the day it's better for us to provide a larger inventory of products since it will increase the odds that the end user will find something *he* likes and *can use* for *himself* with us.
Well first of all I have not advocated for banning any desktops from Fedora, this discussion is not about deleting packages from the Fedora repository. And I realize that some users will make the choice to use something else than our default setup.
But that doesn't invalidate the value of having a default, focusing development resources on the default and expecting any alternative solutions to be able to productively co-exist with the default. Because a lot of end users couldn't care less about the 10 different desktops available, they just want something that works with their hardware and software. And if we know the top 10 feature requests we have a realistic chance of doing something about it as we don't have to consider trying to do it 10 places or work around 10 sets of different desktop quirks and bugs.
To me this is a cornerstone of what the products are meant to be about, trying to do active development of an identified product as opposed to just passively packaging whatever a series of upstreams happens to provide.
Christian
On 02/03/2014 04:08 PM, Christian Schaller wrote:
Well first of all I have not advocated for banning any desktops from Fedora, this discussion is not about deleting packages from the Fedora repository. And I realize that some users will make the choice to use something else than our default setup.
But that doesn't invalidate the value of having a default, focusing development resources on the default and expecting any alternative solutions to be able to productively co-exist with the default. Because a lot of end users couldn't care less about the 10 different desktops available, they just want something that works with their hardware and software. And if we know the top 10 feature requests we have a realistic chance of doing something about it as we don't have to consider trying to do it 10 places or work around 10 sets of different desktop quirks and bugs.
To me this is a cornerstone of what the products are meant to be about, trying to do active development of an identified product as opposed to just passively packaging whatever a series of upstreams happens to provide.
You do realize we are an distribution that represents "upstream" in the larger GNU/Linux eco system and as such our "target users" are users that participate and contribute *back* to the community not the "consuming end user" we leave those for ubuntu.
You are excluding contributing back users with the action of elevating one application or application stack over another we ship since in the end of the day those users are the ones that collectively come together and make magic and innovation happen within the community which is what spins did and still do for the project thou as has become apparent many do not see them as "products" since they are not the "right products" .
JBG
On Mon, Feb 3, 2014 at 5:30 PM, "Jóhann B. Guðmundsson" johannbg@gmail.com wrote:
On 02/03/2014 04:08 PM, Christian Schaller wrote:
Well first of all I have not advocated for banning any desktops from Fedora, this discussion is not about deleting packages from the Fedora repository. And I realize that some users will make the choice to use something else than our default setup.
But that doesn't invalidate the value of having a default, focusing development resources on the default and expecting any alternative solutions to be able to productively co-exist with the default. Because a lot of end users couldn't care less about the 10 different desktops available, they just want something that works with their hardware and software. And if we know the top 10 feature requests we have a realistic chance of doing something about it as we don't have to consider trying to do it 10 places or work around 10 sets of different desktop quirks and bugs.
To me this is a cornerstone of what the products are meant to be about, trying to do active development of an identified product as opposed to just passively packaging whatever a series of upstreams happens to provide.
You do realize we are an distribution that represents "upstream" in the larger GNU/Linux eco system and as such our "target users" are users that participate and contribute *back* to the community not the "consuming end user" we leave those for ubuntu.
Yes because a distribution that has no users other then its own developers is the best way to attract contributors. Note: It isn't.
desktop@lists.fedoraproject.org
-
Alberto Ruiz -
Alex G.S. -
Christian Schaller -
drago01 -
Jóhann B. Guðmundsson