Hi Manu, On Thursday 28 November 2013 22:18:42 Manu Gupta wrote:
Hi Agustin,
Let me try to answer your queries and I have a few questions of mine as well.
I find nothing new in a lot of things you have enumerated here. Please have a look at it http://en.opensuse.org/openSUSE:Strategy
It is how it should be since many the proposal comes from analyzing also the community documentation, interviewing community members, talking to them in events, in the office..... The proposal is about increasing focus, not about spread it by pursuing a new goals. Among the several possibilities we came through, we chose the closest and most compatible with who we are. We looked to base the proposal on our strengths and already shared values, not new ones. This is key to succeed, in my opinion. The lack of ideas is not a problem in openSUSE. So it should not surprise you that there are not many new ideas.
1) Add focus to increase alignment among contributors. What kind of focus are you talking about? I am not sure Can you explain a bit?
3) Catalyze openSUSE maturity process.
- openSUSE has an interesting number of contributors. Now we think it is
time to reinforce our structures.
- Having more solid structures/groups will allow openSUSE to assume more
responsibilities and deliver.
- In general, we think we need few rules but good ones, easy to follow and
analyze.
* Rolling distribution based on release early/release often principle.
Is it not conflicting with the below points, longer release cycles?
What we are proposing is decoupling our distribution, so instead of delivering a compromise between what distro developers and users need, we provide a better solution for both of them. And we propose to do this is an achievable way, in my opinion.
* Using OBS repositories to cover further use cases.
- The idea is to initially target developers that need a stable base and
tools on top.
- Another interesting use case are derivatives. There could be more.
Derivatives are usually community owned, in past we publicized SUSE Studio. Also, do we really have the man power (being realistic) to churn out more derivatives from within the community?
Derivatives are the consequence of having a solid base others can build solution for their needs upon. Our main role would be to create that base. With our tools and our values, we would have a better chance to attract people interested in creating them. The proposal is not about providing solutions directly as much as creating the conditions for other to come and create them.
Principles that would drive our efforts: * Stability and quality as core values.
- We can make the more stable distribution for users based on Green values.
This has been discussed numerous times and everyone agrees on it. Why do we need to reestablish it?
Because it is so important for the proposal that I did not wanted to take it for granted.
* Longer release cycle.
Is there really a very good reason for a longer release cycle. I mean what kind of problems will it solve? What is openSUSE team really expecting out of it?
The Release cycle is not a goal, is a tool.... to achieve users and production needs. It is also a relevant reference check point for each milestone of the development, integration, stabilization and promotion process of any distro release. What is the Release cycle that our users need? Among all the possibilities we have, which one also addresses our development needs and constrains? If we want to adapt our distribution to developers we need to bring them in different layers/areas of the distribution new software, which means that we need to update the Soft. versions often. If we want to adapt to "professional users" (not corporate users), we need to bring them a solid system, with well proven technologies, putting extra effort in other areas like deployments, usability, translations, etc. So, since in the "new release" we propose to target users that work with openSUSE, we need to bring them values that, in some areas, require time to deliver. But there is also other elements to consider. Developers or power users are capable of consuming the software very fast. The time between the software is released by us (project) and they have it installed is low. So releasing frequently is a "feature". The time that users that work with openSUSE (non devs) requires to consume the software is longer. In some cases so long that they skip releases. How much effort from our side is "not consumed" by a user that require more that 8 months to make an update worth it for his particular "use case"? I would say..... a lot. There are more well known factors associated to cost, maintenance, scalability, etc. So we do not try to solve any problem with the Release cycle. We simply propose to adapt our release cycle to our users need, taking in consideration internal factors and constrains. The exact "cadence" needs to come as a consequence of the definition of the Action Plan for the Release, not now.
Open Governance Model =====================
Our Governance model has improved over the years. We would like to work with the community the following years in taking some steps further.
* Technical governance model adapted to our new development processes: very
few but clearer rules. Mentoring ecosystem.
Again how do you plan to have a mentoring ecosystem. I am not sure again. Mentoring takes a lot of time and few people have been willing to put in that kind of effort.
Mentoring is a culture you deliver by creating policies and processes that requires and/or promote training to improve or increase your skills and responsibilities within the project. Two examples we are trying to push... 1.- Increase the training hours in oSC. This action would benefit from "packaging these training sessions" to make them replicable in some other events we are attending to. It also needs a little more focus on needs. Right now the focus in on the offer side. To simplify..... the content offer is based on the trainers desires more than on the contributors demands/needs. We can have a better balance. This is a point in which SUSE can add value in our community. 2.- As you can see in the proposal for factory, we have defined several roles with different duties and levels of expertise required. By defining policies that promote that those who assume a certain role train others to also assume that same role, we are giving steps toward promoting that culture. 3.- The openSUSE Team has become in the last months part of SUSE training program. We have trainees with us. We will publish a blog post with more details about this. This is an internal action but with external impact also.
There are some questions we would like you to answer:
1) Do you agree with the proposed goals?
We are bringing topics that have already been discussed, we just need to move forward and I believe we are doing so again and again.
I understand your frustration. I counted on it in every step of the process that have led us here. This is one of the reasons I wanted to make sure we came here with something solid enough to be credible and open enough to leave room for "chefs". It is a very hard compromise to achieve. Being open to adapt the proposal to our (project) needs and desires can balance this compromise. I would also like to tell you that is our plan to support these ideas we are presenting with executing effort. The proposal is not only about defining direction, it is also about defining actions. This makes the proposal more interesting, I hope. If the process comes to a successful ending, we will define the execution plans (sprints) and begin to work asap. What I want is to make sure we is that, in general, the openSUSE Team can define a sustained effort for a few months instead of reacting to what comes. I think we can spread that vision to other areas of the project. Many of you participate upstream. This idea could be the "equivalent" of a major release approach/design. Translated to our context, we are proposing a possible next "major release".
What I believe is : 1. We need to stop discussing things what have been already discussed
If you change the context, the picture that comes from discussed ideas might change. We need to revisit some of them.
2. If we all know that something is missing, then let us take them and discuss one by one.
Can you elaborate this further? I do not understand you.
2) Are there any other aspects relevant in this discussion? Can you summarize> them?
3) Which are the major risks you see in this view?
4) How do you think we should proceed in order to go from these ideas to real> actions?
5) What suggestions do you have for this "New Factory" and "New Release"?
None so far, but really is the New Release really worth it, will it actually solve a lot of problems ( I am not a factory contributor, so my views can be discarded ).
Please add any other comments or ideas you have in this or a separate thread.
Saludos -- Agustin Benito Bethencourt openSUSE Team Lead at SUSE abebe@suse.com -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org
Saludos -- Agustin Benito Bethencourt openSUSE Team Lead at SUSE abebe@suse.com -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org