Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions: 1) On an internal mailing list in RedHat I have seen some comments about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed. 2) Another solution might be to organize the elections once more after some internal gaining in the Env&Stacks WG, to raise awareness of people about the work this WG is doing.
I am open for any other ideas, helping to solve the current situation. Please comment.
[1] https://fedoraproject.org/wiki/Env_and_Stacks/Nominations
Regards, Jan
On 24 November 2015 at 02:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed. 2) Another solution might be to organize the elections once more after some internal gaining in the Env&Stacks WG, to raise awareness of people about the work this WG is doing.
I am open for any other ideas, helping to solve the current situation. Please comment.
These were my comments previously:
Maybe the problem with not a lot of nominations is that no one knows what Env & Stacks really does? When I talk with a lot of people their opinion is that it is mostly a talking versus doing group. It doesn't have any "power", it doesn't have any real "responsibility", and it doesn't have a known scope.
Now those may all be false assumptions, but those are the ones that I see on IRC from people. Maybe explaining clearly what the group really does, really has scope over and what it is supposed to deliver may get people interested in it?
Looking through the history of the list.. there are 2 people who post regularly to this list: Jan and Honza with no others posting since maybe June or August. I don't know the exact number of people on the list, but that would indicate that most people are working on these problems in other locations.. possibly other Fedora lists and definitely other CentOS lists. With that being the case, are there reasons for keeping this group?
Hi Stephen,
First of all I would like to thank you for your open mail.
I will try to react on it a bit. On 11/24/2015 08:24 PM, Stephen John Smoogen wrote:
On 24 November 2015 at 02:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed. 2) Another solution might be to organize the elections once more after some internal gaining in the Env&Stacks WG, to raise awareness of people about the work this WG is doing.
I am open for any other ideas, helping to solve the current situation. Please comment.
These were my comments previously:
Maybe the problem with not a lot of nominations is that no one knows what Env & Stacks really does? When I talk with a lot of people their opinion is that it is mostly a talking versus doing group. It doesn't have any "power", it doesn't have any real "responsibility", and it doesn't have a known scope.
I guess, that you are right. But see on Docker. It is part of Fedora and it works.
There are many projects around env&stack like here [1]. But If I can tell we don't have an aim for each election period, Fedora release or real target which we are focusing on. This could be a problem.
From my last election period, I think that we discuss and discuss a bit more. We should work on a new things or a visions for Fedora or what should be an aim for E&S. I know that there is a bunch of projects like Software Component Pipeline, DevPI, DevAssistant, Fedora Developer Portal, Docker, Copr and another. We are working on it. But they are not visible so much, I guess. We should do a meeting where should be an update about projects.
But Stephen, on the other hand. Many folks maintain several packages or they are busy with another work. I guess, that we can not concentrate fully on E&S projects. Correct me if I am wrong.
Now those may all be false assumptions, but those are the ones that I see on IRC from people. Maybe explaining clearly what the group really does, really has scope over and what it is supposed to deliver may get people interested in it?
Looking through the history of the list.. there are 2 people who post regularly to this list: Jan and Honza with no others posting since maybe June or August. I don't know the exact number of people on the list, but that would indicate that most people are working on these problems in other locations.. possibly other Fedora lists and definitely other CentOS lists. With that being the case, are there reasons for keeping this group?
Group is important, but we should define aims. Like Fedora modularization. Docker, .... I think, other WG like Cloud, Workstation, Base have to work closely with us.
But as I said, thanks for you opened email.
[1] https://fedoraproject.org/wiki/Env_and_Stacks/Tasklist
On 25 November 2015 at 02:24, Petr Hracek phracek@redhat.com wrote:
Hi Stephen,
First of all I would like to thank you for your open mail.
I will try to react on it a bit. On 11/24/2015 08:24 PM, Stephen John Smoogen wrote:
On 24 November 2015 at 02:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed. 2) Another solution might be to organize the elections once more after some internal gaining in the Env&Stacks WG, to raise awareness of people about the work this WG is doing.
I am open for any other ideas, helping to solve the current situation. Please comment.
These were my comments previously:
Maybe the problem with not a lot of nominations is that no one knows what Env & Stacks really does? When I talk with a lot of people their opinion is that it is mostly a talking versus doing group. It doesn't have any "power", it doesn't have any real "responsibility", and it doesn't have a known scope.
I guess, that you are right. But see on Docker. It is part of Fedora and it works.
I don't see what that has to do with Env and Stacks. Yes it is an environment and it is a stack but there is no connection to where docker got into Fedora because of this group. It doesn't mean there isn't a connection but it isn't obvious from emails on this list that the work was because of this group or that it got in faster because there was a committee.
There are many projects around env&stack like here [1]. But If I can tell we don't have an aim for each election period, Fedora release or real target which we are focusing on. This could be a problem.
From my last election period, I think that we discuss and discuss a bit more. We should work on a new things or a visions for Fedora or what should be an aim for E&S. I know that there is a bunch of projects like Software Component Pipeline, DevPI, DevAssistant, Fedora Developer Portal, Docker, Copr and another. We are working on it. But they are not visible so much, I guess. We should do a meeting where should be an update about projects.
But Stephen, on the other hand. Many folks maintain several packages or they are busy with another work. I guess, that we can not concentrate fully on E&S projects. Correct me if I am wrong.
That is true, but if the work is getting done without the committee then why is there a committee? If we treat this like a layer in the kernel or a library.. if the layer does not add anything or that you are finding that you are just calling the lower layers anyway.. why is that layer there? Should it be removed, should it be altered? What things rely on it and would they be better with a different framework?
Now those may all be false assumptions, but those are the ones that I see on IRC from people. Maybe explaining clearly what the group really does, really has scope over and what it is supposed to deliver may get people interested in it?
Looking through the history of the list.. there are 2 people who post regularly to this list: Jan and Honza with no others posting since maybe June or August. I don't know the exact number of people on the list, but that would indicate that most people are working on these problems in other locations.. possibly other Fedora lists and definitely other CentOS lists. With that being the case, are there reasons for keeping this group?
Group is important, but we should define aims. Like Fedora modularization. Docker, .... I think, other WG like Cloud, Workstation, Base have to work closely with us.
But as I said, thanks for you opened email.
[1] https://fedoraproject.org/wiki/Env_and_Stacks/Tasklist
-- Petr Hracek Software Engineer Developer Experience Red Hat, Inc Mob: +420777056169 email: phracek@redhat.com
env-and-stacks mailing list env-and-stacks@lists.fedoraproject.org http://lists.fedoraproject.org/admin/lists/env-and-stacks@lists.fedoraprojec...
On 11/25/2015 08:18 PM, Stephen John Smoogen wrote:
On 25 November 2015 at 02:24, Petr Hracek phracek@redhat.com wrote:
Hi Stephen,
First of all I would like to thank you for your open mail.
I will try to react on it a bit. On 11/24/2015 08:24 PM, Stephen John Smoogen wrote:
On 24 November 2015 at 02:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed. 2) Another solution might be to organize the elections once more after some internal gaining in the Env&Stacks WG, to raise awareness of people about the work this WG is doing.
I am open for any other ideas, helping to solve the current situation. Please comment.
These were my comments previously:
Maybe the problem with not a lot of nominations is that no one knows what Env & Stacks really does? When I talk with a lot of people their opinion is that it is mostly a talking versus doing group. It doesn't have any "power", it doesn't have any real "responsibility", and it doesn't have a known scope.
I guess, that you are right. But see on Docker. It is part of Fedora and it works.
I don't see what that has to do with Env and Stacks. Yes it is an environment and it is a stack but there is no connection to where docker got into Fedora because of this group. It doesn't mean there isn't a connection but it isn't obvious from emails on this list that the work was because of this group or that it got in faster because there was a committee.
Just for the connection, one thing we talked through few weeks back about docker was the Fedora dockerfiles that require much more love on https://github.com/fedora-cloud/Fedora-Dockerfiles and as a consequence Slavek became one of the maintainer on github.
However, even that doesn't invalidate anything said above -- such small steps are not visible and the steps are too small to be taken as big contribution.
There are many projects around env&stack like here [1]. But If I can tell we don't have an aim for each election period, Fedora release or real target which we are focusing on. This could be a problem.
From my last election period, I think that we discuss and discuss a bit more. We should work on a new things or a visions for Fedora or what should be an aim for E&S. I know that there is a bunch of projects like Software Component Pipeline, DevPI, DevAssistant, Fedora Developer Portal, Docker, Copr and another. We are working on it. But they are not visible so much, I guess. We should do a meeting where should be an update about projects.
But Stephen, on the other hand. Many folks maintain several packages or they are busy with another work. I guess, that we can not concentrate fully on E&S projects. Correct me if I am wrong.
That is true, but if the work is getting done without the committee then why is there a committee? If we treat this like a layer in the kernel or a library.. if the layer does not add anything or that you are finding that you are just calling the lower layers anyway.. why is that layer there? Should it be removed, should it be altered? What things rely on it and would they be better with a different framework?
Valid points. From my PoV, having a real committee is needed only when there are some questionable issues where we need to dive deep, explain several solutions and vote for one of them. During last year, many discussions about future of Fedora happened during E&S meetings, but I don't remember any voting, it was basically not needed. And without need to be a decision body I don't see reasons to have E&S elections either, maybe not even official membership.
Honza
On 24 November 2015 at 19:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed.
That internal post was from me, and I think the idea of downgrading Environments & Stacks from a governance body to a discussion group is one we should seriously consider.
As part of explaining why I've started to think that, I'd like to point out some relevant historical timelines:
* October 2013: Matthew Miller presented the "Fedora Rings" concept at the inaugural Flock conference * January 2014: CentOS becomes a Red Hat sponsored community project * April 2014: Project Atomic was launched * August 2014: OpenShift Origin v3 rearchitecture was announced
The reason I think that timeline is relevant is that at the time the Environments & Stacks Working Group was first conceived:
1. CentOS was not yet sponsored directly by Red hat 2. Project Atomic didn't exist as a community 3. OpenShift Origin had yet to rebase their architecture on Docker & Kubernetes
What this means is that initiatives that may otherwise have been Fedora Environments & Stacks projects have instead found homes in other communities - the governing body for softwarecollections.org is a CentOS SIG, the Docker integration work in Fedora is largely being driven through the Cloud WG and Fedora Release Engineering by way of initiatives like "Two Week Atomic", and the various pieces of the Atomic Developer Bundle can be consumed directly from Project Atomic by the Workstation WG, without needing additional input from Environments & Stacks.
Containerising components of the operating system itself is a question to be addressed by the Base WG, while authority over the package review process rests with the Packaging Committee.
Individual pieces like the Developer Portal, DevAssistant, and COPR are handled by the people working on them, without needing additional oversight from Envs & Stacks.
The Software Component Pipeline idea is something I've been looking at more recently, but like Software Collections before it, I'm starting to think CentOS may be a better upstream home for that, since it's aimed primarily at folks for whom the operating system is just "the place where my code runs", rather than something they're interested in helping to define. If the SCP ends up looking like it may also be useful in developing Fedora itself, that would likely be a decision for FESCo rather than us.
I think Environments & Stacks still has a useful role to play as a discussion and advocacy group, the only part I'm questioning is the need to be formally constituted as a working group, with elections and formal voting procedures.
Regards, Nick.
On 11/25/2015 11:33 AM, Nick Coghlan wrote:
On 24 November 2015 at 19:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed.
That internal post was from me, and I think the idea of downgrading Environments & Stacks from a governance body to a discussion group is one we should seriously consider.
As part of explaining why I've started to think that, I'd like to point out some relevant historical timelines:
- October 2013: Matthew Miller presented the "Fedora Rings" concept at
the inaugural Flock conference
- January 2014: CentOS becomes a Red Hat sponsored community project
- April 2014: Project Atomic was launched
- August 2014: OpenShift Origin v3 rearchitecture was announced
The reason I think that timeline is relevant is that at the time the Environments & Stacks Working Group was first conceived:
- CentOS was not yet sponsored directly by Red hat
- Project Atomic didn't exist as a community
- OpenShift Origin had yet to rebase their architecture on Docker & Kubernetes
What this means is that initiatives that may otherwise have been Fedora Environments & Stacks projects have instead found homes in other communities - the governing body for softwarecollections.org is a CentOS SIG, the Docker integration work in Fedora is largely being driven through the Cloud WG and Fedora Release Engineering by way of initiatives like "Two Week Atomic", and the various pieces of the Atomic Developer Bundle can be consumed directly from Project Atomic by the Workstation WG, without needing additional input from Environments & Stacks.
Containerising components of the operating system itself is a question to be addressed by the Base WG, while authority over the package review process rests with the Packaging Committee.
Individual pieces like the Developer Portal, DevAssistant, and COPR are handled by the people working on them, without needing additional oversight from Envs & Stacks.
The Software Component Pipeline idea is something I've been looking at more recently, but like Software Collections before it, I'm starting to think CentOS may be a better upstream home for that, since it's aimed primarily at folks for whom the operating system is just "the place where my code runs", rather than something they're interested in helping to define. If the SCP ends up looking like it may also be useful in developing Fedora itself, that would likely be a decision for FESCo rather than us.
I think Environments & Stacks still has a useful role to play as a discussion and advocacy group, the only part I'm questioning is the need to be formally constituted as a working group, with elections and formal voting procedures.
Regards, Nick.
As a previous member I agree. Maybe it's time to work on the agenda without elected body.
Best regards, Marcela
On 11/25/2015 11:33 AM, Nick Coghlan wrote:
On 24 November 2015 at 19:32, Jan Kurik jkurik@redhat.com wrote:
Hi Env & Stacks Working Group,
during the Nomination period of the Nov/Dec 2015 Fedora Elections, this working group has not got enough nominees to cover all the open seats. We have currently 4 open seats but only 2 nominees [1]. I would like to agree on a plan, how to deal with this situation.
From my point of view there might be two solutions:
- On an internal mailing list in RedHat I have seen some comments
about handover of responsibilities from Env&Stacks WG to other teams (like Atomic, Cloud, etc.). If this is the case, the WG might revisit its purpose and adapt the resources to it, so the Elections will not be needed.
That internal post was from me, and I think the idea of downgrading Environments & Stacks from a governance body to a discussion group is one we should seriously consider.
As part of explaining why I've started to think that, I'd like to point out some relevant historical timelines:
- October 2013: Matthew Miller presented the "Fedora Rings" concept at
the inaugural Flock conference
- January 2014: CentOS becomes a Red Hat sponsored community project
- April 2014: Project Atomic was launched
- August 2014: OpenShift Origin v3 rearchitecture was announced
The reason I think that timeline is relevant is that at the time the Environments & Stacks Working Group was first conceived:
- CentOS was not yet sponsored directly by Red hat
- Project Atomic didn't exist as a community
- OpenShift Origin had yet to rebase their architecture on Docker & Kubernetes
What this means is that initiatives that may otherwise have been Fedora Environments & Stacks projects have instead found homes in other communities - the governing body for softwarecollections.org is a CentOS SIG, the Docker integration work in Fedora is largely being driven through the Cloud WG and Fedora Release Engineering by way of initiatives like "Two Week Atomic", and the various pieces of the Atomic Developer Bundle can be consumed directly from Project Atomic by the Workstation WG, without needing additional input from Environments & Stacks.
Containerising components of the operating system itself is a question to be addressed by the Base WG, while authority over the package review process rests with the Packaging Committee.
Individual pieces like the Developer Portal, DevAssistant, and COPR are handled by the people working on them, without needing additional oversight from Envs & Stacks.
The Software Component Pipeline idea is something I've been looking at more recently, but like Software Collections before it, I'm starting to think CentOS may be a better upstream home for that, since it's aimed primarily at folks for whom the operating system is just "the place where my code runs", rather than something they're interested in helping to define. If the SCP ends up looking like it may also be useful in developing Fedora itself, that would likely be a decision for FESCo rather than us.
I think Environments & Stacks still has a useful role to play as a discussion and advocacy group, the only part I'm questioning is the need to be formally constituted as a working group, with elections and formal voting procedures.
I have to admit I haven't thought about the changes above as something that would influence E&S, but it does make sense to me after reading it and giving a day to think more about it. And it's good that such things happen, no matter where it is.
From my PoV there are still same directions where we want to go with Fedora distro -- being more flexible, lowering barriers, etc. Even the Component Pipeline is something I see valid for Fedora, maybe just piece of it, not whole. But having that, it would be something that would bring the missing flexibility.
That said, I mostly agree with the opinions here about the elections and stuff.
honza
On 26 November 2015 at 17:07, Honza Horak hhorak@redhat.com wrote:
From my PoV there are still same directions where we want to go with Fedora distro -- being more flexible, lowering barriers, etc. Even the Component Pipeline is something I see valid for Fedora, maybe just piece of it, not whole. But having that, it would be something that would bring the missing flexibility.
Right, I still see value in Envs & Stacks as a public discussion forum to kick around ideas like "How might the Software Component Pipeline concept be useful to Fedora?". There are a range of folks working on, or at least interested in, various developer experience and modularisation related projects that have the potential to influence Fedora's future direction, so it's good to have a place to be able to raise those ideas that don't already have a home and ask "Hey, is this something we think is worth pursuing further?".
However, that "preliminary sounding board" role doesn't require the formal governance structure we have now - the mailing list and the space on the wiki are all we need, with one of the things we discuss being where and how to pitch ideas we decide to pursue.
Cheers, Nick.
----- Original Message -----
On 26 November 2015 at 17:07, Honza Horak hhorak@redhat.com wrote:
From my PoV there are still same directions where we want to go with Fedora distro -- being more flexible, lowering barriers, etc. Even the Component Pipeline is something I see valid for Fedora, maybe just piece of it, not whole. But having that, it would be something that would bring the missing flexibility.
Right, I still see value in Envs & Stacks as a public discussion forum to kick around ideas like "How might the Software Component Pipeline concept be useful to Fedora?". There are a range of folks working on, or at least interested in, various developer experience and modularisation related projects that have the potential to influence Fedora's future direction, so it's good to have a place to be able to raise those ideas that don't already have a home and ask "Hey, is this something we think is worth pursuing further?".
However, that "preliminary sounding board" role doesn't require the formal governance structure we have now - the mailing list and the space on the wiki are all we need, with one of the things we discuss being where and how to pitch ideas we decide to pursue.
I totally agree. I also think we should be a group of people that encourages new ideas and new ways to improve Fedora - and also tries to give guidance and helping hand to those that come up with the ideas. IOW, if you have a crazy idea and are afraid/unsure what the broader community would say, come talk to us first and we'll work with you to form a solid proposal/POC/"marketing strategy" for it.
Cheers, Nick.
-- Nick Coghlan | ncoghlan@gmail.com | Brisbane, Australia
env-and-stacks@lists.fedoraproject.org
-
Bohuslav Kabrda -
Honza Horak -
Jan Kurik -
Marcela Mašláňová -
Nick Coghlan -
Petr Hracek -
Stephen John Smoogen