Introduction ------------
Based on discussions at and around Flock, the Fedora Project Board has approved a proposal for a big change in the way we put Fedora together. Rather than presenting one Fedora with multiple slightly-different install options, future Fedora will be designed, developed, and promoted as three separate products built around a common core.
To take that idea from talk to reality, we're making a corresponding change to Fedora leadership. Each product will be guided by a Working Group, which will function as an independent subcommittee of FESCo, (the Fedora Engineering Steering Committee). FESCo will resolve issues which impact multiple working groups, and the Fedora Board will continue to set overall strategic direction, but the working groups will be largely autonomous within their own areas.
The Groups ----------
We are creating a group for each of the three initial products the Board has approved:
* Fedora Workstation * Fedora Server * Fedora Cloud
The Board asks that the Working Groups determine their own target audience definition and product description as a first task; the names aren't set in stone.
We're also creating groups to focus on the common core, and to work on policies and practices for software operating outside of Fedora's traditional packaging model, alongside the existing (and continuing) Fedora Packaging Committee.
* Base Design Working Group * Environments & Software Stacks Working Group
Composition -----------
The working groups' initial membership will be chosen by FESCo from volunteers. This message is the request for those volunteers to self-nominate.
Each group will have at least one FESCo member, who will act as a liaison to FESCo and as a representative of the group at FESCo meetings.
We would like each group to also have representation from other major areas of Fedora - Quality Assurance, Infrastructure, Release Engineering, Documentation, Design, Websites, Ambassadors, Feature Wrangler, Marketing. We don't intend for this to be additional work to current members of those teams; working group members should interact with and augment the existing subprojects.
What's Expected ---------------
These working groups will be more formal than the existing SIGs, with a documented governing structure and process of operation, voting members, up-to-date and maintained project materials, and regular open communication. All working group members will need to actively participate.
The first responsibility will be to establish a governance charter, followed by a product requirements document. We're obviously in the middle of Fedora 20 development, and that's the priority for many of us right now. For that reason, these deliverables won't be required until after the F20 release, but we do want to start organizing as soon as possible.
Interlude: Interested in Something Else? ----------------------------------------
Are the projects listed above not your interest? That's fine; you can keep on working the way you are now. Or, perhaps you're interested in a target product, but something different from the ones described above. In that case, you may start a "secondary product" working group following the same model; if that is successful the Fedora Board may elect to promote that product to a primary target.
How to Self-Nominate --------------------
To volunteer to serve on one of the new Fedora working groups, simply add yourself to the appropriate section in the wiki page below, along with a brief description of your current involvement with Fedora and plans for participation in this group.
https://fedoraproject.org/wiki/Fedora.next/WG_Nominations
Next Steps ----------
The nomination period will be at least one month from this announcement. FESCo will review the applications, appoint the initial members, and assist with the development of each group's independent governance.
I feel this is a an exciting evolution for Fedora. At the same time of course it will be a big change, and there could be some risk of the increased complexity fragmenting Fedora development somewhat, but it should FESCo to scale to support the needs of these separate products better.
I know this is work-in-progress and details are still going to emerge so forgive me if I am jumping the start gun here going into nitty-gritty details, but I am not getting an overall picture of how the current images/trees and spins of Fedora releases will map into the new WGs and products:
Presumably the DVD.iso may be replaced by images for the individual new products.
- Fedora Workstation
Will this subsume Live-Desktop.iso and Live-KDE.iso? What about other current desktop Spins? Presumably some of these might have a secondary WG.
- Fedora Server
I am assume this would include a GUI
- Fedora Cloud
and this no GUI.
- Base Design Working Group
@core @standard @?
What about the main toolchain, devel languages, and X/Wayland, etc? Would they fit in here too, or would they be covered by FESCo?
((Will each product with iso's will have its own netinst.iso, or would it allow choosing the product? TBD presumably...))
- Environments & Software Stacks Working Group
Might this involve separate repos building on top of the main fedora repos?
Will all the products remain on the same schedule? Will spins require secondary WG's? What will be the process for starting a secondary WG?
I guess my general question is also about clarifying the general scope of each of the WG's though I gather this is still to be defined.
I am sure there are a lot of things to think about, but these are some of the initial questions that have been coming up in my mind. I am just curious to hear more of the general overall release vision without pre-empting the coming discussions and proposals of the individual WG's.
Jens
ps I wasn't at Flock alas so I may have missed some of the earlier discussions that might already have covered some of this...
Hi,
On Thu, 2013-09-19 at 00:58 -0400, Jens Petersen wrote:
ps I wasn't at Flock alas so I may have missed some of the earlier discussions that might already have covered some of this...
some of your questions are answered in Matthew Miller's Flock presentation "An Architecture for a More Agile Fedora": - video: https://www.youtube.com/watch?v=PqoE-tqAQro - slides: http://mattdm.org/fedora/next/
Best regards, Tadej
On Thu, Sep 19, 2013 at 12:58:32AM -0400, Jens Petersen wrote:
- Fedora Workstation
Will this subsume Live-Desktop.iso and Live-KDE.iso? What about other current desktop Spins? Presumably some of these might have a secondary WG.
Right -- one of the key things we need to do is work on the infrastructure for building these products in general, and make that infrastructure available to SIGs for possible products outside the initial primary ones.
- Fedora Server
I am assume this would include a GUI
I'm not assuming either way. From my sysadmin background, I would expect that it _wouldn't_, but I also know that having a server GUI is important to some people.
- Fedora Cloud
and this no GUI.
Right, now, that I'm pretty sure of.
- Base Design Working Group
@core @standard @?
Somewhere like that, except for more carefully designed and maintained, and with a charter for working with the needs of the higher levels as a real, meaningful base os.
What about the main toolchain, devel languages, and X/Wayland, etc? Would they fit in here too, or would they be covered by FESCo?
They'd fit somewhere else -- roughly where they always have been. There is an idea for something like "the Fedora Commons", except we can't call it that because that name is taken by the _other_ Fedora (the digital repository software).
((Will each product with iso's will have its own netinst.iso, or would it allow choosing the product? TBD presumably...))
Yes, TBD. I think that possibly only the desktop will produce a traditional install ISO, the server might be a netinstall plus a collection of shared kickstarts in a git repo, and the cloud image will be a qcow2 + available in cloud providers. But it's all to be worked out.
- Environments & Software Stacks Working Group
Might this involve separate repos building on top of the main fedora repos?
Definitely.
Will all the products remain on the same schedule?
Initially, yes. We can look at diverging in the future if that starts seeming like a good idea.
Will spins require secondary WG's? What will be the process for starting a secondary WG?
Start a SIG or work in the existing one. Basically the process will be just like the promotion of secondary architectures -- start meeting the general criteria that the primaries are held to, show ability to produce the product consistently and meet whatever standards, and then FESCo will approve it. We don't have a process for this worked out, and in the interest of not overwhelming ourselves with paperwork I think we'll cross that bridge when we get to it.
Fedora has long had a distinction between a "sub project" and a SIG. The Working Groups are more like the former -- they need to have a formal governance structure, regular communications, and so on. SIGs are just a collection of interested people, which may or may not strive to become more formal.
Matthew Miller wrote:
On Thu, Sep 19, 2013 at 12:58:32AM -0400, Jens Petersen wrote:
- Fedora Workstation
Will this subsume Live-Desktop.iso and Live-KDE.iso? What about other current desktop Spins? Presumably some of these might have a secondary WG.
Right -- one of the key things we need to do is work on the infrastructure for building these products in general, and make that infrastructure available to SIGs for possible products outside the initial primary ones.
But the current spins will become even more second-class citizens than they are right now, whereas 2 spins of dubious value to our real-world users (Server and Cloud) get featured instead. (How many people will really use those?) The "Workstation" (hidden GNOME) monoculture is also a completely unchanged continuation of the "Desktop" (hidden GNOME) monoculture with just a new name (a name which is all the sillier considering that most Fedora users are home users). The addition of 2 non-desktop spins is only a lame attempt at papering over that GNOME monopoly.
The selection of the 3 "Products" makes the whole concept of Products and Working Groups worthless and counterproductive. The selection of Products should have been based on the existing successful spins, and the Working Groups formed from the existing SIGs.
What about the main toolchain, devel languages, and X/Wayland, etc? Would they fit in here too, or would they be covered by FESCo?
They'd fit somewhere else -- roughly where they always have been. There is an idea for something like "the Fedora Commons", except we can't call it that because that name is taken by the _other_ Fedora (the digital repository software).
Fedora Core? ;-)
Kevin Kofler
On Oct 10, 2013 8:20 PM, "Kevin Kofler" kevin.kofler@chello.at wrote:
Matthew Miller wrote:
On Thu, Sep 19, 2013 at 12:58:32AM -0400, Jens Petersen wrote:
- Fedora Workstation
Will this subsume Live-Desktop.iso and Live-KDE.iso? What about other current desktop Spins? Presumably some of these might have a secondary WG.
Right -- one of the key things we need to do is work on the
infrastructure
for building these products in general, and make that infrastructure available to SIGs for possible products outside the initial primary
ones.
But the current spins will become even more second-class citizens than
they
are right now, whereas 2 spins of dubious value to our real-world users (Server and Cloud) get featured instead. (How many people will really use those?) The "Workstation" (hidden GNOME) monoculture is also a completely unchanged continuation of the "Desktop" (hidden GNOME) monoculture with
just
a new name (a name which is all the sillier considering that most Fedora users are home users). The addition of 2 non-desktop spins is only a lame attempt at papering over that GNOME monopoly.
The selection of the 3 "Products" makes the whole concept of Products and Working Groups worthless and counterproductive. The selection of Products should have been based on the existing successful spins, and the Working Groups formed from the existing SIGs.
What about the main toolchain, devel languages, and X/Wayland, etc? Would they fit in here too, or would they be covered by FESCo?
They'd fit somewhere else -- roughly where they always have been. There
is
an idea for something like "the Fedora Commons", except we can't call it that because that name is taken by the _other_ Fedora (the digital repository software).
Fedora Core? ;-)
Kevin Kofler--
Are you then nominating yourself for a working group to create the fourth product?
--Pete
Hello, On Fri, Oct 11, 2013 at 4:20 AM, Kevin Kofler kevin.kofler@chello.at wrote:
But the current spins will become even more second-class citizens than they are right now, whereas 2 spins of dubious value to our real-world users (Server and Cloud) get featured instead. (How many people will really use those?) The "Workstation" (hidden GNOME) monoculture is also a completely unchanged continuation of the "Desktop" (hidden GNOME) monoculture with just a new name (a name which is all the sillier considering that most Fedora users are home users).
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
The addition of 2 non-desktop spins is only a lame attempt at papering over that GNOME monopoly.
No, we are not adding 2 new products and all that extra work just to make GNOME look better. That would be an extreme amount of work to tackle a PR issue, and it couldn't even work, as this thread shows.
The selection of the 3 "Products" makes the whole concept of Products and Working Groups worthless and counterproductive. The selection of Products should have been based on the existing successful spins, and the Working Groups formed from the existing SIGs.
The move to 3 products was intended to be a real change, not a relabeling of the things we already do. The fact that we don't have a successful server SIG/spin was seen as a problem that needs to be fixed, not as a reason to continue avoiding server uses. Mirek
[1] As opposed to any of 1) non-GNOME, 2) GNOME changed by Fedora, 3) GNOME upstream changing. I don't know enough to say whether any of these variants is generally preferred within FESCo.
On Fri, 2013-10-11 at 15:58 +0200, Miloslav Trmač wrote:
On Fri, Oct 11, 2013 at 4:20 AM, Kevin Kofler kevin.kofler@chello.at wrote:
But the current spins will become even more second-class citizens than they are right now, whereas 2 spins of dubious value to our real-world users (Server and Cloud) get featured instead. (How many people will really use those?) The "Workstation" (hidden GNOME) monoculture is also a completely unchanged continuation of the "Desktop" (hidden GNOME) monoculture with just a new name (a name which is all the sillier considering that most Fedora users are home users).
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
I look forward to their patches.
- ajax
On Fri, Oct 11, 2013 at 03:58:41PM +0200, Miloslav Trmač wrote:
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
? The intent was very much for the working group to set their own priorities. I don't see any role for FESCo in making that decision.
On 10/11/2013 01:58 PM, Miloslav Trmač wrote:
The fact that we don't have a successful server SIG/spin was seen as a problem that needs to be fixed, not as a reason to continue avoiding server uses.
Interesting when and how was that conclusion reached?
Yeah sure the server sub-community got quiet after constantly having to assure it's role and existance in the "desktop" distribution but inactive...
Did someone even bother to reach out to the community and ask them how they would like to move forward? ( Not that I recall any thread doing just that )
JBG
On Fri, Oct 11, 2013 at 03:53:39PM +0000, "Jóhann B. Guðmundsson" wrote:
Did someone even bother to reach out to the community and ask them how they would like to move forward? ( Not that I recall any thread doing just that )
It was discussed at Flock. It was discussed on this mailing list. The community representatives on FESCo and the board discussed it. All of this happened in public. Which community do you feel was given no opportunity to represent their opinions?
On 10/11/2013 03:59 PM, Matthew Garrett wrote:
On Fri, Oct 11, 2013 at 03:53:39PM +0000, "Jóhann B. Guðmundsson" wrote:
Did someone even bother to reach out to the community and ask them how they would like to move forward? ( Not that I recall any thread doing just that )
It was discussed at Flock. It was discussed on this mailing list. The community representatives on FESCo and the board discussed it. All of this happened in public. Which community do you feel was given no opportunity to represent their opinions?
For the first not everybody is subscribed to the list, secondly not everyone can and never will be able to attend flock or other events.
Was there any attempt to reach out to the relevant sub-community was there a mail or discussion held on the server list even if only to see who where active on it?
JBG
On Fri, Oct 11, 2013 at 04:19:00PM +0000, "Jóhann B. Guðmundsson" wrote:
On 10/11/2013 03:59 PM, Matthew Garrett wrote:
community representatives on FESCo and the board discussed it. All of this happened in public. Which community do you feel was given no opportunity to represent their opinions?
For the first not everybody is subscribed to the list, secondly not everyone can and never will be able to attend flock or other events.
Fedora development takes place on this list.
Was there any attempt to reach out to the relevant sub-community was there a mail or discussion held on the server list even if only to see who where active on it?
Given that the last mail to the server list was over 18 months ago, the answer is that there's nobody active on it.
On 10/11/2013 04:27 PM, Matthew Garrett wrote:
Was there any attempt to reach out to the relevant sub-community was there a mail or discussion held on the server list even if only to see who where active on it?
Given that the last mail to the server list was over 18 months ago, the answer is that there's nobody active on it.
In other words it was not done.
JBG
On Fri, Oct 11, 2013 at 04:33:24PM +0000, "Jóhann B. Guðmundsson" wrote:
On 10/11/2013 04:27 PM, Matthew Garrett wrote:
Was there any attempt to reach out to the relevant sub-community was there a mail or discussion held on the server list even if only to see who where active on it?
Given that the last mail to the server list was over 18 months ago, the answer is that there's nobody active on it.
In other words it was not done.
Because there's no active server sub-community. The people interested in server work are working within the general Fedora development community, which means devel@ is the appropriate list to reach them.
On 10/11/2013 04:41 PM, Matthew Garrett wrote:
On Fri, Oct 11, 2013 at 04:33:24PM +0000, "Jóhann B. Guðmundsson" wrote:
On 10/11/2013 04:27 PM, Matthew Garrett wrote:
>>>Was there any attempt to reach out to the relevant sub-community was >>>there a mail or discussion held on the server list even if only to >>>see who where active on it?
Given that the last mail to the server list was over 18 months ago, the answer is that there's nobody active on it.
In other words it was not done.
Because there's no active server sub-community. The people interested in server work are working within the general Fedora development community, which means devel@ is the appropriate list to reach them.
That's quite the assumption and based on that I assume the next step planned is to kill the server list and just mobiles the people interested here right. <shrug>
JBG
On Fri, Oct 11, 2013 at 04:47:34PM +0000, "Jóhann B. Guðmundsson" wrote:
On 10/11/2013 04:41 PM, Matthew Garrett wrote:
Because there's no active server sub-community. The people interested in server work are working within the general Fedora development community, which means devel@ is the appropriate list to reach them.
That's quite the assumption and based on that I assume the next step planned is to kill the server list and just mobiles the people interested here right. <shrug>
If people doing server development work feel that they'd be better doing so on a separate mailing list, that option's there. But given the number of server-related discussions we've had on this list, they don't seem to currently feel that.
On Fri, Oct 11, 2013 at 05:41:28PM +0100, Matthew Garrett wrote:
Because there's no active server sub-community. The people interested in server work are working within the general Fedora development community, which means devel@ is the appropriate list to reach them.
I also posted to devel-announce, which should go to even people who aren't trying to keep up with this list and all its thrills. Individual posts to the various existing mailing lists probably wouldn't hurt.
I think it's useful to distinguish between the existence of a community (or, even, a group of people with common interests whether or not they are acting like a community) and a mailing list, which just a tool and which people may or may not find useful in practice. The current effort is certainly meant to address that same community and if they (we, actually -- I'm definitely one of 'em) find that mailing list more useful in the future, awesome.
Miloslav Trmač wrote:
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
[snip]
[1] As opposed to any of 1) non-GNOME, 2) GNOME changed by Fedora, 3) GNOME upstream changing. I don't know enough to say whether any of these variants is generally preferred within FESCo.
2 features which would have changed that have been proposed over time: https://fedoraproject.org/wiki/Features/KDE_Plasma_Desktop_by_default https://fedoraproject.org/wiki/Features/Cinnamon_as_Default_Desktop Both have been rejected by FESCo.
Kevin Kofler
On Sat, Oct 19, 2013 at 9:27 PM, Kevin Kofler kevin.kofler@chello.at wrote:
Miloslav Trmač wrote:
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
[snip]
[1] As opposed to any of 1) non-GNOME, 2) GNOME changed by Fedora, 3) GNOME upstream changing. I don't know enough to say whether any of these variants is generally preferred within FESCo.
2 features which would have changed that have been proposed over time: https://fedoraproject.org/wiki/Features/KDE_Plasma_Desktop_by_default https://fedoraproject.org/wiki/Features/Cinnamon_as_Default_Desktop Both have been rejected by FESCo.
Repeating myself, "I don't know enough to say whether any of these variants is generally preferred within FESCo.".
Looking back at these two proposals in particular, they have one thing in common - they haven't been even supported by maintainers of the respective desktops. That made both proposals pretty much impossible to vote for. Mirek
Miloslav Trmač (mitr@volny.cz) said:
On Sat, Oct 19, 2013 at 9:27 PM, Kevin Kofler kevin.kofler@chello.at wrote:
Miloslav Trmač wrote:
No, the intent was very much to change what the resulting desktop prioritizes. Quite a few FESCo members would be rather disappointed if the new Workstation ended up just an unchanged GNOME[1].
[snip]
[1] As opposed to any of 1) non-GNOME, 2) GNOME changed by Fedora, 3) GNOME upstream changing. I don't know enough to say whether any of these variants is generally preferred within FESCo.
2 features which would have changed that have been proposed over time: https://fedoraproject.org/wiki/Features/KDE_Plasma_Desktop_by_default https://fedoraproject.org/wiki/Features/Cinnamon_as_Default_Desktop Both have been rejected by FESCo.
Repeating myself, "I don't know enough to say whether any of these variants is generally preferred within FESCo.".
And I would argue that having the user interface swing wildly in design & implementation based on "the current composition of an elected board that is refreshed in part every six months" is not the sort of situation that Fedora would want to be in anyway.
(Of course, if it was, that would add an entirely different feel to the elections. Vote now in next month's elections to bring on people to completely change the proposed product split? Candidates running on 23 products instead of 3?)
Bill
Bill Nottingham wrote:
And I would argue that having the user interface swing wildly in design & implementation based on "the current composition of an elected board that is refreshed in part every six months" is not the sort of situation that Fedora would want to be in anyway.
That's a very lame excuse for sticking with an awful desktop environment as the default just because it is the status quo.
Kevin Kofler
On Fri, 2013-11-01 at 04:40 +0100, Kevin Kofler wrote:
Bill Nottingham wrote:
And I would argue that having the user interface swing wildly in design & implementation based on "the current composition of an elected board that is refreshed in part every six months" is not the sort of situation that Fedora would want to be in anyway.
That's a very lame excuse for sticking with an awful desktop environment as the default just because it is the status quo.
Not excellent...please avoid insulting other people's hard work just because you are working on something else.
Matthias Clasen wrote:
On Fri, 2013-11-01 at 04:40 +0100, Kevin Kofler wrote:
Bill Nottingham wrote:
And I would argue that having the user interface swing wildly in design & implementation based on "the current composition of an elected board that is refreshed in part every six months" is not the sort of situation that Fedora would want to be in anyway.
That's a very lame excuse for sticking with an awful desktop environment as the default just because it is the status quo.
Not excellent...please avoid insulting other people's hard work just because you are working on something else.
You got the causality wrong: It's not that I call GNOME "awful" because I work on KDE, but that I work on KDE because GNOME does not fulfill my needs and expectations at all. (Even GNOME 2 didn't, but GNOME 3 went even further away from it.)
Kevin Kofler
On Sat, 2013-11-02 at 16:17 +0100, Kevin Kofler wrote:
Matthias Clasen wrote:
On Fri, 2013-11-01 at 04:40 +0100, Kevin Kofler wrote:
That's a very lame excuse for sticking with an awful desktop environment as the default just because it is the status quo.
Not excellent...please avoid insulting other people's hard work just because you are working on something else.
You got the causality wrong: It's not that I call GNOME "awful" because I work on KDE, but that I work on KDE because GNOME does not fulfill my needs and expectations at all.
There's a long way between « doesn't fulfill my needs and expectations at all » and « awful ».
I think Matthias was complaining that you used the latter, when the former is absolutely fine, and perfectly understandable.
Hello, Two updates arising from today's FESCo meeting:
On Wed, Sep 11, 2013 at 9:31 PM, Matthew Miller mattdm@fedoraproject.org wrote:
The nomination period will be at least one month from this announcement.
The nomination period will end on Oct 14 0:00 UTC.
Membership in a Working Group is a significant investment in time. We strongly recommend against volunteering to serve on more than one, lest you get your wish. Mirek
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 10/02/2013 05:05 PM, Miloslav Trmač wrote:
Hello, Two updates arising from today's FESCo meeting:
On Wed, Sep 11, 2013 at 9:31 PM, Matthew Miller mattdm@fedoraproject.org wrote:
The nomination period will be at least one month from this announcement.
The nomination period will end on Oct 14 0:00 UTC.
Just to circle back around on this, the self-nomination period is now officially over.
On 11.09.2013 21:31, Matthew Miller wrote:
Introduction
Based on discussions at and around Flock, the Fedora Project Board has approved a proposal for a big change in the way we put Fedora together. Rather than presenting one Fedora with multiple slightly-different install options, future Fedora will be designed, developed, and promoted as three separate products built around a common core.
To take that idea from talk to reality, we're making a corresponding change to Fedora leadership. Each product will be guided by a Working Group, which will function as an independent subcommittee of FESCo, (the Fedora Engineering Steering Committee). FESCo will resolve issues which impact multiple working groups, and the Fedora Board will continue to set overall strategic direction, but the working groups will be largely autonomous within their own areas.
The Groups
We are creating a group for each of the three initial products the Board has approved:
- Fedora Workstation
- Fedora Server
- Fedora Cloud
What about Fedora Embedded? Do you plan to drop ARM support on Fedora? I can't match small credit card size devices with either Workstation, Server or Cloud group. Is this group list fixed or could be extended and on what basis?
Mateusz Marzantowicz
On 10/03/2013 05:42 PM, Mateusz Marzantowicz wrote:
On 11.09.2013 21:31, Matthew Miller wrote:
Introduction
Based on discussions at and around Flock, the Fedora Project Board has approved a proposal for a big change in the way we put Fedora together. Rather than presenting one Fedora with multiple slightly-different install options, future Fedora will be designed, developed, and promoted as three separate products built around a common core.
To take that idea from talk to reality, we're making a corresponding change to Fedora leadership. Each product will be guided by a Working Group, which will function as an independent subcommittee of FESCo, (the Fedora Engineering Steering Committee). FESCo will resolve issues which impact multiple working groups, and the Fedora Board will continue to set overall strategic direction, but the working groups will be largely autonomous within their own areas.
The Groups
We are creating a group for each of the three initial products the Board has approved:
- Fedora Workstation
- Fedora Server
- Fedora Cloud
What about Fedora Embedded? Do you plan to drop ARM support on Fedora? I can't match small credit card size devices with either Workstation, Server or Cloud group. Is this group list fixed or could be extended and on what basis?
Mateusz Marzantowicz
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Marcela
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 10/04/2013 04:31 AM, Marcela Mašláňová wrote:
On 10/03/2013 05:42 PM, Mateusz Marzantowicz wrote:
On 11.09.2013 21:31, Matthew Miller wrote:
Introduction ------------
Based on discussions at and around Flock, the Fedora Project Board has approved a proposal for a big change in the way we put Fedora together. Rather than presenting one Fedora with multiple slightly-different install options, future Fedora will be designed, developed, and promoted as three separate products built around a common core.
To take that idea from talk to reality, we're making a corresponding change to Fedora leadership. Each product will be guided by a Working Group, which will function as an independent subcommittee of FESCo, (the Fedora Engineering Steering Committee). FESCo will resolve issues which impact multiple working groups, and the Fedora Board will continue to set overall strategic direction, but the working groups will be largely autonomous within their own areas.
The Groups ----------
We are creating a group for each of the three initial products the Board has approved:
- Fedora Workstation * Fedora Server * Fedora Cloud
What about Fedora Embedded? Do you plan to drop ARM support on Fedora? I can't match small credit card size devices with either Workstation, Server or Cloud group. Is this group list fixed or could be extended and on what basis?
Mateusz Marzantowicz
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Yes, we're probably not going to be offering a direct embedded product right from the get-go. However, the Fedora ARM project will definitely have a place in the Server and Cloud variants, as several hardware manufacturers have been announcing ARM-based servers.
We certainly don't want to suggest that ARM-based hobby projects like Pidora will be going away. ARM folks are encouraged to work on the Base Design functionality so we can meet their needs in the core OS and then we'll be providing tools similar to the current Spins to allow them to continue building their own installable images.
The main change being made here is with how we are presenting *The Fedora Project* to the world. In the past, we've tried to be all things to all people, but going forward we want to pick a few specific areas that we will focus on (and market) particularly. The wide world of Fedora software and packages will always be there.
On 10/04/2013 02:49 PM, Stephen Gallagher wrote:
What about Fedora Embedded? Do you plan to drop ARM support on Fedora? I can't match small credit card size devices with either Workstation, Server or Cloud group. Is this group list fixed or could be extended and on what basis?
Mateusz Marzantowicz
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Yes, we're probably not going to be offering a direct embedded product right from the get-go. However, the Fedora ARM project will definitely have a place in the Server and Cloud variants, as several hardware manufacturers have been announcing ARM-based servers.
Also remember that ARM are not only for smell devices: http://goo.gl/qpDzy
On Fri, Oct 4, 2013 at 8:30 AM, Mihamina RKTMB mihamina@rktmb.org wrote:
On 10/04/2013 02:49 PM, Stephen Gallagher wrote:
What about Fedora Embedded? Do you plan to drop ARM support on Fedora? I can't match small credit card size devices with either Workstation, Server or Cloud group. Is this group list fixed or could be extended and on what basis?
Mateusz Marzantowicz
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Yes, we're probably not going to be offering a direct embedded product right from the get-go. However, the Fedora ARM project will definitely have a place in the Server and Cloud variants, as several hardware manufacturers have been announcing ARM-based servers.
Also remember that ARM are not only for smell devices: http://goo.gl/qpDzy
That article is almost 2 years old, and an ARM based HP Moonshot server is still not available. While the intent of your message is certainly true, you might do better by picking a better example like the Calxeda ARM servers. Fedora is already using those as the builders in koji.
josh
On 10/04/2013 11:49 AM, Stephen Gallagher wrote:
The main change being made here is with how we are presenting *The Fedora Project* to the world. In the past, we've tried to be all things to all people, but going forward we want to pick a few specific areas that we will focus on (and market) particularly. The wide world of Fedora software and packages will always be there.
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
" basically if whatever they come up with doesn't work for RH, it's a non-starter."
Sending a clear cut message to the community and people wanting to participate in it so in a sense Red Hat has already presentedFedora Project* to the world or at least it's view on it's community.
"Here come and join Fedora we will allow you to do everything but influence or participate in anykind of direction that project might take but we do our best effort making you feel like you actually are making a difference"
I can understand why people get the notion of that we are nothing more the a freaking test bed for RH due to a disgrace and disgusting corporate behaviour towards the community like was shown on the last FESCO meeting when the true intention of RH was shown with the direction of Fedora and how much community gets to be involved in that decision.
What a disgrace and disgusting corporate behaviour towards the community that took place on that meeting putting Red Hat on par with Canonical.
Truly a dark day in Fedora history that took place there on that meeting.
But I think Stephen for trying to propose to certain extent fair proposal on how to handle the working group nominations
JBG
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 10/04/2013 11:14 AM, "Jóhann B. Guðmundsson" wrote:
On 10/04/2013 11:49 AM, Stephen Gallagher wrote:
The main change being made here is with how we are presenting *The Fedora Project* to the world. In the past, we've tried to be all things to all people, but going forward we want to pick a few specific areas that we will focus on (and market) particularly. The wide world of Fedora software and packages will always be there.
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
" basically if whatever they come up with doesn't work for RH, it's a non-starter."
Sending a clear cut message to the community and people wanting to participate in it so in a sense Red Hat has already presentedFedora Project* to the world or at least it's view on it's community.
I'm going to attempt to respond to this here, but I do realize it's going to be difficult to communicate my point clearly, especially amongst strong emotions.
The phrasing used above was unfortunate and open to the wrong interpretation. Red Hat is not attempting to force its vision into the world. Red Hat has gotten as far as it has not by being a dictator but by being a facilitator.
Red Hat as an organization recognizes that individual contributors in the greater community are an asset both to that community and to Red Hat and we don't want to alienate anyone with good ideas. Furthermore, our general philosophy is heavily centered around the idea that the community will sometimes come up with dissenting ideas that prove out to be better than Red Hat's current plans. Historically, the company's approach in those situations has been to re-target, often by hiring those individuals so that they can work on their idea full-time, rather than as a side-project.
Now, what was really intended by that statement that you quoted above (and I acknowledge I'm putting words in people's mouths a bit) is that Red Hat *may* flex its muscles a bit if the community were to do something extremely unlikely that would be in direct opposition to the needs of Red Hat. And when I say "extreme", I'm talking in the neighborhood of "Fedora should drop the ix86 line and focus only on embedded ARM" or "Fedora should give up producing an OS entirely and become a package repository for CentOS".
You are absolutely *not* going to see Red Hat micromanaging the creation of a Fedora-to-Red-Hat-specification because that would actively *degrade* the value of Fedora to Red Hat. Fedora is not just a testbed for RHEL, it's a proving ground for technologies, and those may eventually get included RHEL *from any source*, not just Red Hat.
I cannot reiterate this statement enough "Red Hat cannot dictate Fedora because that action will actively remove the real benefit that Red Hat gets from Fedora".
"Here come and join Fedora we will allow you to do everything but influence or participate in anykind of direction that project might take but we do our best effort making you feel like you actually are making a difference"
The Fedora Community *does* make a difference. We (Red Hat) could not do what we do without you. If that weren't the case, Red Hat would never spend the kind of money it does on funding Fedora rel-eng, FUDCons, Flock, FADs, representation at other conferences, swag, etc. Red Hat would spend that money *on Red Hat* if it didn't see value in the enabling these communities.
I can understand why people get the notion of that we are nothing more the a freaking test bed for RH due to a disgrace and disgusting corporate behaviour towards the community like was shown on the last FESCO meeting when the true intention of RH was shown with the direction of Fedora and how much community gets to be involved in that decision.
What a disgrace and disgusting corporate behaviour towards the community that took place on that meeting putting Red Hat on par with Canonical.
Truly a dark day in Fedora history that took place there on that meeting.
But I think Stephen for trying to propose to certain extent fair proposal on how to handle the working group nominations
On 10/04/2013 03:44 PM, Stephen Gallagher wrote:
Now, what was really intended by that statement that you quoted above (and I acknowledge I'm putting words in people's mouths a bit) is that Red Hat*may* flex its muscles a bit if the community were to do something extremely unlikely that would be in direct opposition to the needs of Red Hat. And when I say "extreme", I'm talking in the neighborhood of "Fedora should drop the ix86 line and focus only on embedded ARM" or "Fedora should give up producing an OS entirely and become a package repository for CentOS".
You are absolutely*not* going to see Red Hat micromanaging the creation of a Fedora-to-Red-Hat-specification because that would actively*degrade* the value of Fedora to Red Hat. Fedora is not just a testbed for RHEL, it's a proving ground for technologies, and those may eventually get included RHEL*from any source*, not just Red Hat.
Oh please cut the crap and stop beating that dead parrot [1] this is anything but *may*.
What became clear on that meeting is that Red Hat with FESCO as it's enforcer was taking the necessary precautionary steps to ensure the community would never even have the slightest chance and the freedom to venture off it's beaten path.
It was an action and decisions based on distrust rather than openness with Red Hat *already* tightening it's muscle on it's community leash.
And that three product proposal is nothing more then Red Hat micromanaging the creation of Fedora products through the false notion of the community being part of that through the working groups .
We already have an existing cloud community and SIG which is perfectly capable of delivering product(s).
We already have an existing Server community and SIG perfectly capable to deliver several products.
We already have several *DE which could have had the equal opportunity of engaging in a specific "Workstation SIG" but as we all know the decision has already been made within Red Hat to s/default desktop/Fedora workstation/" which is probably what this whole shenanigan is about in the first place some marketing and documentation stunt while migrating the default desktop to workstation product status.
It's an outright disgusting and disgraceful corporate behavior towards the community what Red Hat has done here and one can just image all the ideas that have started to spring alive in the community which Red Hat has killed in birth so it would not interfere with it's corporate vision or disrupt it's internal processes and workflows.
JBG
1. http://www.davidpbrown.co.uk/jokes/monty-python-parrot.html
On Fri, Oct 04, 2013 at 03:14:27PM +0000, "Jóhann B. Guðmundsson" wrote:
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
Jóhann, you're taking one out-of-context quote from one FESCo member, reading too much into it, and building an alarmist story around it.
This is absolutely a real community process. Red Hat members of the working groups can make their merit-based cases the same way as anyone else, and if they can't show that merit to the community, they don't get a special trump card. They will have to find another way to advance their cause.
You may think that this is just talk, but I promise you it isn't. Fedora provides value to Red Hat in many different ways, but genuine community voice is among the most crucial. If that voice tells us one thing and we can't listen, that's our failure, our loss -- and not what's going to happen here.
It's completely fair for Red Hat -- and Red Hatters -- to talk about what directions in Fedora we think would be most beneficial to the company, and about the resources -- time, money, people, and so on -- that we could bring to bear in certain directions (and probably won't in other directions). If we clearly talk about that, and about the technical merit of directions proposed, and we can't be convincing, and can't adapt what we're proposing to become convincing... well, we have some soul-searching to do.
On Fri, Oct 4, 2013 at 2:23 PM, Matthew Miller mattdm@fedoraproject.orgwrote:
On Fri, Oct 04, 2013 at 03:14:27PM +0000, "Jóhann B. Guðmundsson" wrote:
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
Jóhann, you're taking one out-of-context quote from one FESCo member, reading too much into it, and building an alarmist story around it.
This is absolutely a real community process. Red Hat members of the working groups can make their merit-based cases the same way as anyone else, and if they can't show that merit to the community, they don't get a special trump card. They will have to find another way to advance their cause.
Let me add a few words here as well. I'm of the same opinion as Matthew here -- I think Jóhann is reading too much into a unfortunately worded quote. (And, based on Jóhann's recent behavior, he seems to have an axe to grind with Red Hat.)
I'd like to state for the record that while I was the Fedora Project Leader, Red Hat never once told me what to do as the FPL or exercised any undue influence on what Fedora should or shouldn't be doing. Of course, they watched with interest to see what was happening in Fedora, and various Red Hat engineers added new features to Fedora along the way, and quite a few Red Hat employees took part on the Fedora Board and FESCo and FAmSCo and various other SIGs -- but I can state unequivocally that I never tried to force Fedora's hand, or did I see any sort of underhanded behavior or grand conspiracy to which Jóhann refers.
I'm sorry Jóhann, but I can't sit here and watch you make those kinds of accusations without sharing what I saw and experienced while I was an insider at Red Hat. It's not helpful to the Fedora community to continue with these baseless accusations.
Let me even be a little more blunt here: I don't think Fedora could thrive without the support and help that Red Hat (and, by extension, it's employees) provide. It could probably survive, but it would only be limping along. In that same manner, I don't think Red Hat could thrive the way it has without the great work that Fedora does. For better or worse, the Fedora community and Red Hat need each other. I don't see any easy way for them to go their separate ways without damaging both sides.
(For the record, I no longer work for Red Hat, have nothing tangible to gain by Red Hat's success, but still hold them in high esteem based on my time working there.)
-- Jared Smith
On 10/04/2013 09:50 PM, Jared K. Smith wrote:
On Fri, Oct 4, 2013 at 2:23 PM, Matthew Miller <mattdm@fedoraproject.org mailto:mattdm@fedoraproject.org> wrote:
I'm sorry Jóhann, but I can't sit here and watch you make those kinds of accusations without sharing what I saw and experienced while I was an insider at Red Hat. It's not helpful to the Fedora community to continue with these baseless accusations.
Jared are accusing me of baseless accusation after what is clear cut and took place at that meeting?
Do you really want to head down this road?
Did you not read the meeting long as well as Stephens response?
Did you look at the history who signed up to the WG page before it even got announced to the community?
Want to take this further shall we start pointing out individuals that Red Hat invented job positions for within our communities then planted individuals outside the community in those positions to satisify it's compulseve corporate need for control?
Shall we talk about how Red Hat employees have been granted all kinds of privileges within our community without as even bother to introduce themselves to the community even to the extent that fesco is now judging people if they are "socially ready" for proven packagers while Red Hat employees walk around and are granted those privileges freely?
Back of your words Jared dishonour me to my face and tell I'm wrong or if I'm lying!!!
JBG
On Saturday, October 05, 2013 11:29 AM, "Jóhann B. Guðmundsson" wrote:
Did you look at the history who signed up to the WG page before it even got announced to the community?
Who? People who were at Flock and/or had read the meeting logs and knew about the WG about to be announced.
I knew about that page before it was announced.
And I don't work for Red Hat, I just happen to read the FESCo meeting logs every week.
Want to take this further shall we start pointing out individuals that Red Hat invented job positions for within our communities then planted individuals outside the community in those positions to satisify it's compulseve corporate need for control?
So every time you say that, I can't help thinking you're just jealous they took someone « outside the community » instead of you.
It might not be what you're thinking, but it's really how you sound.
Maybe those « individuals » are just more competent? Or you know, maybe Red Hat prefers hiring people who don't spend their time vomiting their hatred on mailing-lists?
On Sat, Oct 5, 2013 at 8:08 AM, Mathieu Bridon bochecha@fedoraproject.orgwrote:
On Saturday, October 05, 2013 11:29 AM, "Jóhann B. Guðmundsson" wrote:
Did you look at the history who signed up to the WG page before it even got announced to the community?
Who? People who were at Flock and/or had read the meeting logs and knew about the WG about to be announced.
I knew about that page before it was announced.
And I don't work for Red Hat, I just happen to read the FESCo meeting logs every week.
Want to take this further shall we start pointing out individuals that
Red Hat invented job positions for within our communities then planted individuals outside the community in those positions to satisify it's compulseve corporate need for control?
So every time you say that, I can't help thinking you're just jealous they took someone « outside the community » instead of you.
It might not be what you're thinking, but it's really how you sound.
Maybe those « individuals » are just more competent? Or you know, maybe Red Hat prefers hiring people who don't spend their time vomiting their hatred on mailing-lists?
It sounds like this is an issue from way back from the tone of this email.
Come on guys we cant keep going back and forth on this. find it sad that we are. My opinion is this. we get funding from these guys. So to some extent I think they ca put in engineers to assist us when they see a feature that they want in. I am not commenting on the methods they use. This is because I actually do not care.
It is open source. Unhappy with something, just fork it and move on. That is the beauty of it. No one will come after you cause that is the beauty of the whole thing.
I think we also need to look at deliverables. I usually am not the person looking at the small details. For me it is more of deliver and we are all good. I mean if this needs to be revised I am sure we can have a look at it and ask them to change it for the community.
this is just my $0.02
thanks.
-- Mathieu
-- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.**org/mailman/listinfo/develhttps://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-**of-conducthttp://fedoraproject.org/code-of-conduct
On 10/05/2013 07:08 AM, Mathieu Bridon wrote:
So every time you say that, I can't help thinking you're just jealous they took someone « outside the community » instead of you.
It might not be what you're thinking, but it's really how you sound.
Maybe those « individuals » are just more competent? Or you know, maybe Red Hat prefers hiring people who don't spend their time vomiting their hatred on mailing-lists?
You are making the assumption that Red Hat has not already offered me a job as well as the fact that I would associate my name to a company when behaves like this.
Good for you...
JBG
Am 06.10.2013 22:18, schrieb Jóhann B. Guðmundsson:
On 10/05/2013 07:08 AM, Mathieu Bridon wrote:
So every time you say that, I can't help thinking you're just jealous they took someone « outside the community » instead of you.
It might not be what you're thinking, but it's really how you sound.
Maybe those « individuals » are just more competent? Or you know, maybe Red Hat prefers hiring people who don't spend their time vomiting their hatred on mailing-lists?
You are making the assumption that Red Hat has not already offered me a job as well as the fact that I would associate my name to a company when behaves like this.
Good for you...
*what* exactly is your problem with Redhat?
i am always the bad-ass because my hard opinions about wrong technical decisions, well so it may be, but what you are doing all the time is fight against a company because it is a company without realize what Redhat is doing for the open source ecosystem over many years
do you not realize that without Redhat Fedora would not exist at all and *nobody* but you is interested in Fedora without Redhat?
Le 06/10/2013 22:18, "Jóhann B. Guðmundsson" a écrit :
You are making the assumption that Red Hat has not already offered me a job as well as the fact that I would associate my name to a company when behaves like this.
Good for you...
JBG
I don't think that's what Mathieu meant here and I know he *immediately* regretted that part. Everyone here recognize your skills and your hard work, you're a worthy contributor and i don't give a shit of what Red Hat thinks about you or anyone else here.
I understand your concerns about Fedora, most us want to keep Fedora a truly community-driven project. But Matthew did an outstanding work in pointing out Fedora weaknesses and proposed a good roadmap to make Fedora a great platform to build products upon. Who cares if he did it on his paid time ? We -the community- discussed his proposal at Flock, then all the process has been lead in the open and there are logs and mails that prove it.
The only sad thing here is to see contributors whom i highly esteem disputing like brats instead of discussing it, keeping a cool head.
H.
PS: everyone should listen to Jared, Fedora Community voice of reason :o)
----- Original Message -----
On 10/05/2013 07:08 AM, Mathieu Bridon wrote:
So every time you say that, I can't help thinking you're just jealous they took someone « outside the community » instead of you.
It might not be what you're thinking, but it's really how you sound.
Maybe those « individuals » are just more competent? Or you know, maybe Red Hat prefers hiring people who don't spend their time vomiting their hatred on mailing-lists?
You are making the assumption that Red Hat has not already offered me a job as well as the fact that I would associate my name to a company when behaves like this.
I'd say most of Fedora (and even most of Red Hatters) would quit immediately in case the company starts behave like you think it behaves. And I'm saying it as a guy who signed mortgage week ago.
And that means I'd still like to see you join us and work on Fedora full time ;-).
R.
Good for you...
JBG
devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
On 10/07/2013 08:20 AM, Jaroslav Reznik wrote:
----- Original Message -----
On 10/05/2013 07:08 AM, Mathieu Bridon wrote:
I'd say most of Fedora (and even most of Red Hatters) would quit immediately in case the company starts behave like you think it behaves. And I'm saying it as a guy who signed mortgage week ago.
Or people turn a blind eye to the facts on what's actually taking place.
- It places distrust in the community ( as came completely clear on last FESCO meeting )
- It puts the community to disadvantage compare to it employees which now as stepped up to the level that community members are subjected to character and social scrutiny by FESCO ( look at Dan's pp request 3 meetings ago ) while Red Hat employees entirely bypass that and other ( privileged ) processes that community members have to go through.
- It elevates it's own "product(s)" above community's work either under the so called "defaults" ( or as we are heading now "3 products" ) or various strategically placed "recommendations" here and there putting competing community maintained products at disadvantage.
- It creates ( high ranking ) positions ( suddenly ) in communities, then recruits individuals outside the community and places them in those positions and in those communities ( people can just look through the internet archive's for advertised Fedora positions both for the title they give these individuals as well as the statement you will be working as as opposed to working with ).
etc...
So as you can see it already is behaving as I think it behaves and quite frankly this is an disgusting and unjust corporate behavior towards the community based on mistrust and misuse and sends mixed signals inside and outside of our community and labels our work as some kind of RH experiment and test bed.
All of the issues I have mentioned here before can be dealt with internally by Red Hat.
- It has to take a leap of faith and just let go and place trust in the community since it's highly unlikely that it will venture to far away from Red Hat interest at least I would be very surprised if it did.
- If it thinks that our processes are to complex for an new employee to walk through to gain the necessary access to be able to perform it's work, it needs to work with us improving those processes and workflows so that *everybody* Red Hat employees and community members alike will gain from it as opposed to be bypassing it altogether for it's employee while the community drowns in bureaucracy.
- It will need to understand that forcing everything under a single product ( default ) or three products as well as single audience ( or three different audience ) hinders growth in sub communities ( due to them not being equally presented ) as well as fair competition thus innovation between competing products applications or applications stack ( be it through better written code/compatibility/features/maintenance you know those little things that competing products implement or achieve over each other ) .
- It needs to understand that there is no need to invent ( high ranking ) position and try to elevate new employees to those positions within sub community since it will come naturally on it's own by the share time that employees has to work and dedicate to the sub community surrounding the component or group of components. ( An community member only has around 2 - 4 hours max each day to dedicate to the project unless he's unemployed or is being paid to work in it ).
So fourth and so on,
Red Hat has pretty smart managers and team leaders within their ranks which I'm pretty sure will straight these issues out and deal with the community on equal ground and in harmony which benefits us all.
JBG
2013/10/7 "Jóhann B. Guðmundsson" johannbg@gmail.com
Or people turn a blind eye to the facts on what's actually taking place.
- It places distrust in the community ( as came completely clear on last
FESCO meeting )
Fesco members are all elected by contributors (no nominated members by Red Hat), if you think they doesn't do their job properly, you're more than capable to step up at the next election.
- It puts the community to disadvantage compare to it employees which now
as stepped up to the level that community members are subjected to character and social scrutiny by FESCO ( look at Dan's pp request 3 meetings ago ) while Red Hat employees entirely bypass that and other ( privileged ) processes that community members have to go through.
I don't remember being employed by Red Hat, neither are the other members who were worried that FESCo rushed that pp request. You're completely rewriting history here, because, FESCo approved this pp request without respecting its own guidelines, many voices from the community had intervened to restore legality here.
This incident was handled by community sponsors in respect to the process defined by the community, not by Red Hat employees.
- It elevates it's own "product(s)" above community's work either under
the so called "defaults" ( or as we are heading now "3 products" ) or various strategically placed "recommendations" here and there putting competing community maintained products at disadvantage.
One of Fedora's biggest success was to build a strong community that spear-headed the GNU/Linux efforts for years. But we're reaching our own limits and we have to set clear goals to keep this community together. By defining three products (and people are free to propose other products, it's a truly community-driven process), we are setting these goals that will make Fedora works in the future.
- It creates ( high ranking ) positions ( suddenly ) in communities, then
recruits individuals outside the community and places them in those positions and in those communities ( people can just look through the internet archive's for advertised Fedora positions both for the title they give these individuals as well as the statement you will be working as as opposed to working with ).
etc...
I consider the whole Red Hat as a particular contributor, i don't give a rat's ass about anyone position (community manager, cloud architect or whatever). What i consider is the work done by individuals. Off course, Red Hat wants to drive Fedora where are their own interests, but they have as much power as their contributions are worth to the community. I'm not supporting Matthew's proposal because he works at Red Hat but as fellow contributor who did a great job.
So as you can see it already is behaving as I think it behaves and quite frankly this is an disgusting and unjust corporate behavior towards the community based on mistrust and misuse and sends mixed signals inside and outside of our community and labels our work as some kind of RH experiment and test bed.
As you, i still resent what Brian Steven said about Fedora being only RHEL sandbox, but we have to keep our heads cool. You have valid arguments (mixed signals sent to the community for instance), please, don't mix everything with unfair arguments.
All of the issues I have mentioned here before can be dealt with internally by Red Hat.
- It has to take a leap of faith and just let go and place trust in the
community since it's highly unlikely that it will venture to far away from Red Hat interest at least I would be very surprised if it did.
I'm pretty sure that most RH employees involved in Fedora are thinking the same way.
- If it thinks that our processes are to complex for an new employee to
walk through to gain the necessary access to be able to perform it's work, it needs to work with us improving those processes and workflows so that *everybody* Red Hat employees and community members alike will gain from it as opposed to be bypassing it altogether for it's employee while the community drowns in bureaucracy.
+1
- It will need to understand that forcing everything under a single
product ( default ) or three products as well as single audience ( or three different audience ) hinders growth in sub communities ( due to them not being equally presented ) as well as fair competition thus innovation between competing products applications or applications stack ( be it through better written code/compatibility/features/**maintenance you know those little things that competing products implement or achieve over each other ) .
You're being unfair, this decision has been discussed in the open and has been approved by a fully community process, you were THERE at Flock when we discussed this face to face.
- It needs to understand that there is no need to invent ( high ranking )
position and try to elevate new employees to those positions within sub community since it will come naturally on it's own by the share time that employees has to work and dedicate to the sub community surrounding the component or group of components. ( An community member only has around 2 - 4 hours max each day to dedicate to the project unless he's unemployed or is being paid to work in it ).
People hired at these "high ranking" positions have as much influence as our community wants to give them. Matthew is no exception. As long as Red Hat doesn't interfere in our community defined process as a corporate beast, i'm glad that there are people who are paid working on Fedora because i don't have as much time as i wish i could give to *OUR* project.
So fourth and so on,
Red Hat has pretty smart managers and team leaders within their ranks which I'm pretty sure will straight these issues out and deal with the community on equal ground and in harmony which benefits us all.
JBG
devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.**org/mailman/listinfo/develhttps://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-**of-conducthttp://fedoraproject.org/code-of-conduct
This is a gesture of good faith, and i think that you're only worried for our (aka community) best interests
@+ H.
Hi Jóhann, I do agree with you that the interaction between Red Hat and Fedora needs to be clearer, and that currently it is a bit vaguely defined and thus it gives ground to conspiracy theories and feelings of disenfranchisement.
That said I think you too need to be open to that Red Hat, like yourself and any other participant in the Fedora project does so because there is a sense of self interest. That self interest may wary from enjoyment of community, to skills building, or to Fedora providing a solution to a technical problem you have. Red Hat is not investing heavily in Fedora in terms of infrastructure and development resources just because we as a company needs a place to spend money, if that was the case I am sure you would agree we should instead donate the money to the Red Cross or similar. The reason Red Hat invests in Fedora is because Fedora plays an important part in both product development and in innovating new technologies.
So if Fedora ends up not being interesting or useful to you personally anymore I assume you would leave Fedora behind, the same is true for Red Hat.
So I think part of the reason we end up having these kinds of argument it is because for a long time maybe both inside and outside Red Hat there has been a pretense that Red Hat as a company has no direct interest in Fedora and that Red Hats resources and contributions to the project is a given, no matter what. Red Hats involvement in Fedora has somehow become the unspoken of elephant in the room. Maybe what we need to do is instead start speaking openly of why Red Hat wants to be involved with Fedora.
So you mention that some Red Hat employees have bypassed processes, and I am sure this has happened, but that is a direct consequence of that Fedora not being a 'random' distro for Red Hat, but an integral part of our product development. I mean there is no secret that RHEL is built from Fedora. The tools used to build Fedora overlap and intermingle with the tools used for building RHEL. So I am not saying that makes everything ok, but what I want to say is that we need to accept that these things doesn't happen out of malice, and work together to find solutions for how they can be handled better going forward in a way that is mutually beneficial and acceptable to all.
So there are two solutions to the challenge faced with Red Hat and Fedora. The first option is a decision that Red Hat withdraws from Fedora and tries to build replacements for Fedora current role in our product development. Or that the Fedora community and Red Hat agrees that the current involvement from Red Hat is beneficial to Fedora overall, despite that it comes with some strings attached and that the rules of Fedora might at times collide with the practical concerns of Red Hat, who needs to build products for our customers. And I don't think (almost) anyone inside or outside Red Hat wants solution 1.
So maybe everyone involved needs to take a deep breath and accept that there is no 'clean' solution here. There is no rule that can be made that somehow resolves all the complexity of Fedora both being a community project and at the same time a core part of the Red Hat product development workflow and overall market strategy. Sometimes this weird duality will create friction, but we need to discuss and talk calmly about these issues and try to find solutions, instead of assuming bad things of each other.
And often if a change ends up being good or bad is a lot up to the participants. If you go into something only looking for reasons why it is bad, then there is a good chance you will end up making it bad, at least for yourself. And at the same time if you approach something as an opportunity to do something positive, your chances of doing that is greatly increased. And often the good solutions is about thinking outside the box a bit.
And as a sidenote, I think there is a tendency to brand any discussion about Fedora inside Red Hat as some kind of backroom dealings and skulduggery, but I think this is silly and unfair. Red Hat like any other participant sometimes need to figure out what is the Red Hat position on issues and challenges, a position which might not align with every Fedora community member or every individual Red Hat employee, and Red Hat being a company and not an individual can only reach such positions by discussing them internally first. And to me this is actually beneficial to the Fedora community as it can provide the community with a clear sense of what the official company position is on a given subject, as opposed to trying to somehow extract it from the buzz of various individual Red Hat employees stating a mix of company positions and their private opinions.
The real challenge here is to avoid the need to build company positions lead into a default of doing discussions internally that can be just as fine be done in the public with full community involvement. This is a challenge that any project with a big corporate sponsor has and I been involved with such challenges many times before in jobs prior to Red Hat. And that is going to be an ongoing challenge for Red Hat and the Fedora community that I am sure we will need to keep working on for as long as Red Hat and the Fedora community exists, as it comes out of the soft challenges like human nature and culture development.
Christian
On Mon, 2013-10-07 at 10:10 +0000, "Jóhann B. Guðmundsson" wrote:
On 10/07/2013 08:20 AM, Jaroslav Reznik wrote:
----- Original Message -----
On 10/05/2013 07:08 AM, Mathieu Bridon wrote:
I'd say most of Fedora (and even most of Red Hatters) would quit immediately in case the company starts behave like you think it behaves. And I'm saying it as a guy who signed mortgage week ago.
Or people turn a blind eye to the facts on what's actually taking place.
- It places distrust in the community ( as came completely clear on last
FESCO meeting )
- It puts the community to disadvantage compare to it employees which
now as stepped up to the level that community members are subjected to character and social scrutiny by FESCO ( look at Dan's pp request 3 meetings ago ) while Red Hat employees entirely bypass that and other ( privileged ) processes that community members have to go through.
- It elevates it's own "product(s)" above community's work either under
the so called "defaults" ( or as we are heading now "3 products" ) or various strategically placed "recommendations" here and there putting competing community maintained products at disadvantage.
- It creates ( high ranking ) positions ( suddenly ) in communities,
then recruits individuals outside the community and places them in those positions and in those communities ( people can just look through the internet archive's for advertised Fedora positions both for the title they give these individuals as well as the statement you will be working as as opposed to working with ).
etc...
So as you can see it already is behaving as I think it behaves and quite frankly this is an disgusting and unjust corporate behavior towards the community based on mistrust and misuse and sends mixed signals inside and outside of our community and labels our work as some kind of RH experiment and test bed.
All of the issues I have mentioned here before can be dealt with internally by Red Hat.
- It has to take a leap of faith and just let go and place trust in the
community since it's highly unlikely that it will venture to far away from Red Hat interest at least I would be very surprised if it did.
- If it thinks that our processes are to complex for an new employee to
walk through to gain the necessary access to be able to perform it's work, it needs to work with us improving those processes and workflows so that *everybody* Red Hat employees and community members alike will gain from it as opposed to be bypassing it altogether for it's employee while the community drowns in bureaucracy.
- It will need to understand that forcing everything under a single
product ( default ) or three products as well as single audience ( or three different audience ) hinders growth in sub communities ( due to them not being equally presented ) as well as fair competition thus innovation between competing products applications or applications stack ( be it through better written code/compatibility/features/maintenance you know those little things that competing products implement or achieve over each other ) .
- It needs to understand that there is no need to invent ( high ranking
) position and try to elevate new employees to those positions within sub community since it will come naturally on it's own by the share time that employees has to work and dedicate to the sub community surrounding the component or group of components. ( An community member only has around 2 - 4 hours max each day to dedicate to the project unless he's unemployed or is being paid to work in it ).
So fourth and so on,
Red Hat has pretty smart managers and team leaders within their ranks which I'm pretty sure will straight these issues out and deal with the community on equal ground and in harmony which benefits us all.
JBG
On 2013-10-07 13:56, Christian Fredrik Kalager Schaller wrote:
Hi Jóhann, I do agree with you that the interaction between Red Hat and Fedora needs to be clearer, and that currently it is a bit vaguely defined and thus it gives ground to conspiracy theories and feelings of disenfranchisement.
That said I think you too need to be open to that Red Hat, like yourself and any other participant in the Fedora project does so because there is a sense of self interest. That self interest may wary from enjoyment of community, to skills building, or to Fedora providing a solution to a technical problem you have. Red Hat is not investing heavily in Fedora in terms of infrastructure and development resources just because we as a company needs a place to spend money, if that was the case I am sure you would agree we should instead donate the money to the Red Cross or similar. The reason Red Hat invests in Fedora is because Fedora plays an important part in both product development and in innovating new technologies.
So if Fedora ends up not being interesting or useful to you personally anymore I assume you would leave Fedora behind, the same is true for Red Hat.
So I think part of the reason we end up having these kinds of argument it is because for a long time maybe both inside and outside Red Hat there has been a pretense that Red Hat as a company has no direct interest in Fedora and that Red Hats resources and contributions to the project is a given, no matter what. Red Hats involvement in Fedora has somehow become the unspoken of elephant in the room. Maybe what we need to do is instead start speaking openly of why Red Hat wants to be involved with Fedora.
So you mention that some Red Hat employees have bypassed processes, and I am sure this has happened, but that is a direct consequence of that Fedora not being a 'random' distro for Red Hat, but an integral part of our product development. I mean there is no secret that RHEL is built from Fedora. The tools used to build Fedora overlap and intermingle with the tools used for building RHEL. So I am not saying that makes everything ok, but what I want to say is that we need to accept that these things doesn't happen out of malice, and work together to find solutions for how they can be handled better going forward in a way that is mutually beneficial and acceptable to all.
So there are two solutions to the challenge faced with Red Hat and Fedora. The first option is a decision that Red Hat withdraws from Fedora and tries to build replacements for Fedora current role in our product development. Or that the Fedora community and Red Hat agrees that the current involvement from Red Hat is beneficial to Fedora overall, despite that it comes with some strings attached and that the rules of Fedora might at times collide with the practical concerns of Red Hat, who needs to build products for our customers. And I don't think (almost) anyone inside or outside Red Hat wants solution 1.
So maybe everyone involved needs to take a deep breath and accept that there is no 'clean' solution here. There is no rule that can be made that somehow resolves all the complexity of Fedora both being a community project and at the same time a core part of the Red Hat product development workflow and overall market strategy. Sometimes this weird duality will create friction, but we need to discuss and talk calmly about these issues and try to find solutions, instead of assuming bad things of each other.
And often if a change ends up being good or bad is a lot up to the participants. If you go into something only looking for reasons why it is bad, then there is a good chance you will end up making it bad, at least for yourself. And at the same time if you approach something as an opportunity to do something positive, your chances of doing that is greatly increased. And often the good solutions is about thinking outside the box a bit.
And as a sidenote, I think there is a tendency to brand any discussion about Fedora inside Red Hat as some kind of backroom dealings and skulduggery, but I think this is silly and unfair. Red Hat like any other participant sometimes need to figure out what is the Red Hat position on issues and challenges, a position which might not align with every Fedora community member or every individual Red Hat employee, and Red Hat being a company and not an individual can only reach such positions by discussing them internally first. And to me this is actually beneficial to the Fedora community as it can provide the community with a clear sense of what the official company position is on a given subject, as opposed to trying to somehow extract it from the buzz of various individual Red Hat employees stating a mix of company positions and their private opinions.
The real challenge here is to avoid the need to build company positions lead into a default of doing discussions internally that can be just as fine be done in the public with full community involvement. This is a challenge that any project with a big corporate sponsor has and I been involved with such challenges many times before in jobs prior to Red Hat. And that is going to be an ongoing challenge for Red Hat and the Fedora community that I am sure we will need to keep working on for as long as Red Hat and the Fedora community exists, as it comes out of the soft challenges like human nature and culture development.
Christian
[cut]
Thanks for an extremely well written reply - after reading this I feel like I understand what this is all about. I have not had that feeling before.
That said, please don't top-post: [1]
--alec
[1] http://fedoraproject.org/wiki/Mailing_list_guidelines#If_You_Are_Replying_to...
That said, please don't top-post: [1]
Also, please trim irrelevant material [1]
[1] http://fedoraproject.org/wiki/Mailing_list_guidelines#If_You_Are_Replying_to...
Quoting Jóhann B. Guðmundsson (2013-10-05 05:29:57)
Shall we talk about how Red Hat employees have been granted all kinds of privileges within our community without as even bother to introduce themselves to the community even to the extent that fesco is now judging people if they are "socially ready" for proven packagers while Red Hat employees walk around and are granted those privileges freely?
Your vitriol is not appreciated
I want you to point out the person who got provenpackager privileges without going through normal provenpackager process. I have personally seen (and voted against) a few colleagues who wanted to get provenpackager (or sponsor) permissions but didn't have enough experience IMO. As far as I know they are not provenpackagers/sponsors.
If this really was a problem I know a lot of Red Hat employees would be as unhappy about this inequality as you seem to be. But it's not...
You are actually insulting and attacking integrity of every sponsor (i.e. people who actually vote for/against new provenpackagers).
I like Simon Phipps's quote "Corporations are not people". If you have a problem with specific action taken by specific Red Hat employees: point it out! Do not be generic or people will most likely write you off as another troll.
On Sat, Oct 05, 2013 at 03:29:57AM +0000, "Jóhann B. Guðmundsson" wrote:
Shall we talk about how Red Hat employees have been granted all kinds of privileges within our community without as even bother to introduce themselves to the community even to the extent that fesco is now judging people if they are "socially ready" for proven packagers while Red Hat employees walk around and are granted those privileges freely?
That is an utter fabrication. Red Hat packagers have to go exactly through the same process to became packagers as anyone else (well, it may be easier for them to find a sponsor, but sponsored they must be); they have to go through the same process to became proven packagers etc.
I respectfully suggest that you be silent if you do not know the facts. Your credibility is diminishing rapidly with every untrue statement you put forth.
D.
On 10/08/2013 06:02 AM, David Tardon wrote:
I respectfully suggest that you be silent if you do not know the facts. Your credibility is diminishing rapidly with every untrue statement you put forth.
You did comprehend I was not only talking about PP right?
JBG
On Tue, Oct 08, 2013 at 08:27:18AM +0000, "Jóhann B. Guðmundsson" wrote:
On 10/08/2013 06:02 AM, David Tardon wrote:
I respectfully suggest that you be silent if you do not know the facts. Your credibility is diminishing rapidly with every untrue statement you put forth.
You did comprehend I was not only talking about PP right?
You did read the paragraph you cut out from your reply, right?
So what are these special privileges that all red hat packagers do have, as you claim? I am very interested to hear, because I sure as hell do not have them (or never heard about them from anyone, anyway) and want to remedy the omission .-)
D.
On 10/04/2013 06:23 PM, Matthew Miller wrote:
On Fri, Oct 04, 2013 at 03:14:27PM +0000, "Jóhann B. Guðmundsson" wrote:
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
Jóhann, you're taking one out-of-context quote from one FESCo member, reading too much into it, and building an alarmist story around it.
This is absolutely a real community process. Red Hat members of the working groups can make their merit-based cases the same way as anyone else, and if they can't show that merit to the community, they don't get a special trump card. They will have to find another way to advance their cause.
You may think that this is just talk, but I promise you it isn't. Fedora provides value to Red Hat in many different ways, but genuine community voice is among the most crucial. If that voice tells us one thing and we can't listen, that's our failure, our loss -- and not what's going to happen here.
It's completely fair for Red Hat -- and Red Hatters -- to talk about what directions in Fedora we think would be most beneficial to the company, and about the resources -- time, money, people, and so on -- that we could bring to bear in certain directions (and probably won't in other directions). If we clearly talk about that, and about the technical merit of directions proposed, and we can't be convincing, and can't adapt what we're proposing to become convincing... well, we have some soul-searching to do.
And those words coming from a man who just back stabbed a man he went into feature process with and left him hanging ( Lennart ).
Am I and the rest of the community supposed trust what you suddenly say and claim now?
JBG
On 10/05/2013 05:34 AM, "Jóhann B. Guðmundsson" wrote:
On 10/04/2013 06:23 PM, Matthew Miller wrote:
On Fri, Oct 04, 2013 at 03:14:27PM +0000, "Jóhann B. Guðmundsson" wrote:
Why should the community participate in this when it turns out that the the whole WG and the next proposal is nothing but an utter and total sheninagan on RH behalf as came apparent on last FESCO meeting
Jóhann, you're taking one out-of-context quote from one FESCo member, reading too much into it, and building an alarmist story around it. This is absolutely a real community process. Red Hat members of the working groups can make their merit-based cases the same way as anyone else, and if they can't show that merit to the community, they don't get a special trump card. They will have to find another way to advance their cause.
You may think that this is just talk, but I promise you it isn't. Fedora provides value to Red Hat in many different ways, but genuine community voice is among the most crucial. If that voice tells us one thing and we can't listen, that's our failure, our loss -- and not what's going to happen here.
It's completely fair for Red Hat -- and Red Hatters -- to talk about what directions in Fedora we think would be most beneficial to the company, and about the resources -- time, money, people, and so on -- that we could bring to bear in certain directions (and probably won't in other directions). If we clearly talk about that, and about the technical merit of directions proposed, and we can't be convincing, and can't adapt what we're proposing to become convincing... well, we have some soul-searching to do.
And those words coming from a man who just back stabbed a man he went into feature process with and left him hanging ( Lennart ).
Am I and the rest of the community supposed trust what you suddenly say and claim now?
Hi Johann, you use the word *the community* in your emails a lot, but I don't see many others supporting your opinion, so can you please share with us who is *the community* you're talking about? For one, it's definitely not me and I think of my self as part of the community..(yes, I'm working for the community even outside my RH paid job). So far it looks like you're only hiding behind the term *community* because there's only you (or just a few of you), but it's better to say community than *all four of us*.
Thanks, Jirka
JBG
On 10/08/2013 07:00 AM, Jiri Moskovcak wrote:
Hi Johann, you use the word *the community* in your emails a lot, but I don't see many others supporting your opinion,
Hi Jiri
There are other ways to than being visible to show support and sometimes it's not the best strategy to do so.
When dealing with an overwhelming entity like fortune 500 company you have to be organized mobilized and smart when engaging it to ensuring for example if it manage to silences one voice there is another voice to replace it in the community it but I dont see many outside Red Hat employee contributing to this thread either many of those just want to find a mutual path to solve this ( which ofcourse can be found ).
People are supporting me plenty privately ( if that's what you are wondering ) even asking why I left the big elephant out of this discussion as in one of more real conflict between Fedora's growth and Red Hat's goals being money ( as in the project funds ) with several suggestion how to collect money to fund various for and in the project ( which this thread is not about ).
But before community members start popping up various crowd funding projects to help the aspect of the project that they think are being left out by Red Hat ( by funding or resources ) or find more sponsor or other ways to sponsor it ( manpower hosting what not ), we need to be able to ensure that the various work flows,policy's and other bits can handle a single sponsor and does so well.
JBG
On Oct 8, 2013, at 3:24 AM, Jóhann B. Guðmundsson johannbg@gmail.com wrote:
People are supporting me plenty privately
This answer does not hold up to scrutiny as a response to Jirka's inquiry. Such an arrangement requires publicly visible proxies to be credible. An alternative arrangement is for your mission statement to be presented to the community for non-binding vote. But on the face of it, the above assertion is a non-sequitur consider your implied lack of transparency in the Fedora-Red Hat relationship. It's simply an inappropriate suggestion that more lack of transparency, that's merely in opposition with another, is the way forward.
Chris Murphy
On Fri, 2013-10-04 at 07:49 -0400, Stephen Gallagher wrote:
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Yes, we're probably not going to be offering a direct embedded product right from the get-go. However, the Fedora ARM project will definitely have a place in the Server and Cloud variants, as several hardware manufacturers have been announcing ARM-based servers.
Apologies for the necro, I'm catching up on this whole discussion.
I really think whatever the New Way Of Doing Things turns out to be, it needs to include a minimal network install image much like the current netinst.iso, built for all primary arches, as a primary deliverable. Whether that's considered part of one of the 'products' or the 'base os' or whatever I don't know, but I really think it would be a huge mistake not to ship something along those lines. It is likely to be what a lot of ARM users want. I think having that, and ARM builds of the products so far proposed, should cover the bases - but so far as the Workstation product goes, remember that GNOME does not currently work well on most supported ARM platforms, so our 'ARM Workstation' is currently KDE.
We've done a lot of work over the last few cycles to really bump ARM up to 'first class citizen' status, and a lot of that is coming together - I think reasonably successfully - in F19 and F20. It would be rather odd to go with a change for F21 or F22 which goes in the opposite direction.
On Thu, Oct 24, 2013 at 02:29:25PM -0700, Adam Williamson wrote:
I really think whatever the New Way Of Doing Things turns out to be, it needs to include a minimal network install image much like the current netinst.iso, built for all primary arches, as a primary deliverable. Whether that's considered part of one of the 'products' or the 'base os' or whatever I don't know, but I really think it would be a huge mistake
That seems reasonable, and I don't know the answer to the followup yet either.
On Thu, Oct 24, 2013 at 5:29 PM, Adam Williamson awilliam@redhat.com wrote:
On Fri, 2013-10-04 at 07:49 -0400, Stephen Gallagher wrote:
We just started to support ARM, I don't think we want to drop it. I guess those three products are currently most important and other products like Embedded should go into Spins category. At least for now.
Yes, we're probably not going to be offering a direct embedded product right from the get-go. However, the Fedora ARM project will definitely have a place in the Server and Cloud variants, as several hardware manufacturers have been announcing ARM-based servers.
Apologies for the necro, I'm catching up on this whole discussion.
I really think whatever the New Way Of Doing Things turns out to be, it needs to include a minimal network install image much like the current netinst.iso, built for all primary arches, as a primary deliverable. Whether that's considered part of one of the 'products' or the 'base os' or whatever I don't know, but I really think it would be a huge mistake not to ship something along those lines. It is likely to be what a lot of ARM users want. I think having that, and ARM builds of the products so far proposed, should cover the bases - but so far as the Workstation product goes, remember that GNOME does not currently work well on most supported ARM platforms, so our 'ARM Workstation' is currently KDE.
It is up to each WG to determine their product requirements. That includes which architectures and target users they are trying to produce a product for.
We've done a lot of work over the last few cycles to really bump ARM up to 'first class citizen' status, and a lot of that is coming together - I think reasonably successfully - in F19 and F20. It would be rather odd to go with a change for F21 or F22 which goes in the opposite direction.
ARM is important long term, yes. I don't necessarily think that ARM is equally important across all of the existing products. I find it more likely that ARM is important enough to have it's own WG and it's own product, which may or may not have commonality with the other products.
josh
On Thu, 2013-10-24 at 19:56 -0400, Josh Boyer wrote:
It is up to each WG to determine their product requirements. That includes which architectures and target users they are trying to produce a product for.
We've done a lot of work over the last few cycles to really bump ARM up to 'first class citizen' status, and a lot of that is coming together - I think reasonably successfully - in F19 and F20. It would be rather odd to go with a change for F21 or F22 which goes in the opposite direction.
ARM is important long term, yes. I don't necessarily think that ARM is equally important across all of the existing products. I find it more likely that ARM is important enough to have it's own WG and it's own product, which may or may not have commonality with the other products.
I'm not entirely sure that makes sense; it seems to be a conceptual error. ARM is an architecture. In practice, at present, the ARM-architecture based hardware we support mostly falls into a certain category that kind of naturally lends itself to a particular kind of product, but that seems a transient scenario, not a permanent one. Looked at conceptually, it doesn't make any more sense for there to be an 'ARM working group' and an 'ARM product' than it does for there to be an 'x86_64 working group' and an 'x86_64 product', but those are, I think, prima facie absurd. The concepts of 'working group' and 'product' have been drawn up along broadly _functional_ lines, and a 'working group' or 'product' for a specific system architecture doesn't really line up with that design.
I think the approach I implied in my email - making sure the functional WGs and products we are inventing do not neglect any of our primary architectures and use cases - is the correct way to go.
On Thu, Oct 24, 2013 at 8:52 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2013-10-24 at 19:56 -0400, Josh Boyer wrote:
It is up to each WG to determine their product requirements. That includes which architectures and target users they are trying to produce a product for.
We've done a lot of work over the last few cycles to really bump ARM up to 'first class citizen' status, and a lot of that is coming together - I think reasonably successfully - in F19 and F20. It would be rather odd to go with a change for F21 or F22 which goes in the opposite direction.
ARM is important long term, yes. I don't necessarily think that ARM is equally important across all of the existing products. I find it more likely that ARM is important enough to have it's own WG and it's own product, which may or may not have commonality with the other products.
I'm not entirely sure that makes sense; it seems to be a conceptual error. ARM is an architecture. In practice, at present, the ARM-architecture based hardware we support mostly falls into a certain category that kind of naturally lends itself to a particular kind of product, but that seems a transient scenario, not a permanent one. Looked at conceptually, it doesn't make any more sense for there to be an 'ARM working group' and an 'ARM product' than it does for there to be an 'x86_64 working group' and an 'x86_64 product', but those are, I think, prima facie absurd. The concepts of 'working group' and 'product' have been drawn up along broadly _functional_ lines, and a 'working group' or 'product' for a specific system architecture doesn't really line up with that design.
I think the approach I implied in my email - making sure the functional WGs and products we are inventing do not neglect any of our primary architectures and use cases - is the correct way to go.
You make a good argument but we probably do need to account for the case that a particular product might not be suitable for every architecture. I was thinking more along the lines of every primary architecture has to have one or more of the core products but should not be required to have to them all. How to accomplish that is rather fuzzy.
John
On Thu, Oct 24, 2013 at 6:52 PM, Adam Williamson awilliam@redhat.com wrote:
On Thu, 2013-10-24 at 19:56 -0400, Josh Boyer wrote:
It is up to each WG to determine their product requirements. That includes which architectures and target users they are trying to produce a product for.
We've done a lot of work over the last few cycles to really bump ARM up to 'first class citizen' status, and a lot of that is coming together - I think reasonably successfully - in F19 and F20. It would be rather odd to go with a change for F21 or F22 which goes in the opposite direction.
ARM is important long term, yes. I don't necessarily think that ARM is equally important across all of the existing products. I find it more likely that ARM is important enough to have it's own WG and it's own product, which may or may not have commonality with the other products.
I'm not entirely sure that makes sense; it seems to be a conceptual error. ARM is an architecture. In practice, at present, the ARM-architecture based hardware we support mostly falls into a certain category that kind of naturally lends itself to a particular kind of product, but that seems a transient scenario, not a permanent one.
And if we think the WGs are unable to adapt to transient scenarios, then we've failed. Computing is an ever-changing world. We work with the situation we have today and for the short-term future, and adapt as changes pop up.
Looked at conceptually, it doesn't make any more sense for there to be an 'ARM working group' and an 'ARM product' than it does for there to be an 'x86_64 working group' and an 'x86_64 product', but those are, I think, prima facie absurd. The concepts of 'working group' and 'product'
Yes, x86_64 as a WG is absurd. It's already widely adopted, commonly availalbe, and used in all 3 of the mainly defined products. ARM is not _yet_.
have been drawn up along broadly _functional_ lines, and a 'working group' or 'product' for a specific system architecture doesn't really line up with that design.
I think the approach I implied in my email - making sure the functional WGs and products we are inventing do not neglect any of our primary architectures and use cases - is the correct way to go.
I think pretending the current class of readily available ARM hardware can fully support the products a WG wants to define is somewhat disingenuous. I'm not saying ARM won't soon, but it's simply not the case today. I fully agree that ARM should be kept in mind during the product creation and included if capable, but I do not think it should necessarily be something that _has_ to be targeted.
To get to specifics, I do not foresee the Workstation WG focusing on producing a product that will work equally on ARM. The dearth of workstation style ARM hardware and open graphics drivers would make it even harder to produce that product. This will not always be the case, but it is now and we need products NOW, not whenever ARM catches up. (Please note this is my opinion and does not necessarily reflect that of the WG.) Believe me, I would love to see ARM laptops that were just as functional as x86_64 laptops with full open source software support. Bring them on! We'll adapt as we go. Until then we need to focus on the best target for the products, as defined by the WG.
josh
devel@lists.stg.fedoraproject.org
-
Adam Jackson -
Adam Williamson -
Alec Leamas -
Bill Nottingham -
Chris Murphy -
Christian Schaller -
David Tardon -
DJ Delorie -
Frankie -
Haïkel Guémar -
inode0 -
Jared K. Smith -
Jaroslav Reznik -
Jens-Ulrik Petersen -
Jiri Moskovcak -
Josh Boyer -
Josh Boyer -
Jóhann B. Guðmundsson -
Kevin Kofler -
Marcela Mašláňová -
Mateusz Marzantowicz -
Mathieu Bridon -
Matthew Garrett -
Matthew Miller -
Matthias Clasen -
Mihamina Rakotomandimby -
Miloslav Trmač -
Pete Travis -
Reindl Harald -
Stanislav Ochotnicky -
Stephen Gallagher -
Tadej Janež