We started on-time, but spent some unexpected time introducing ourselves. Some of have been around OpenStack for a few years, but we have many new contributors as well. The discussion was very active. We decided to combine the first two talks about long term plans and project tagging into a longer multi-hour discussion. Allison Randal and Bill Franklin were very helpful in providing background on separate discussions and open source history references. It was a very productive back and forth. We came out of it with a few simple objectives for what this group wants to accomplish.
- Helping OpenStack go faster
- Establishing a release roadmap
- Drive cross-project consistent processes
- Measuring success/progress on above
Rob Hirschfeld went over what is DefCore and RefStack. About half of the people there were familiar and the rest were hearing parts of this for the first time. Arkady Kanevsky reminded us that Tempest is not yet ready for production us and thus neither is RefStack. RefStack and thus Tempest production support is a requirement for using the OpenStack brand and base mark as we move forward. It will be a very strong message to the OpenStack community if we can mobilize more Tempest developers out of a release roadmap. Geoff Arnold and myself got into the debate on RefStack promoting interoperability and requiring code to be present. I hope to convince more people of the need to keep OpenStack more than just APIs.
We spent the rest of the afternoon re-debating how to run through an implementation of feature through the proposed release roadmap workflow. No real progress on project tagging. We will need to re-litigate tomorrow. It was ultimately very useful exercise as we were able to get all those present to recognize how the OpenStack technical governance operates today. In order to change hearts and minds, we need to start with the projects, their cores, and leads. If we can convince the developers of the benefits of a multi-release roadmap, then we can easily convince any other OpenStack related group like the Technical Committee or the Board of Directors.
Day Two we will review the status of the Neutron services split, the future planning for this group, and have breakout session on some smaller, but important topics. Expect another update soon.
We worked through the topics and came up with a revised list. Find the discussion here. I look forward to seeing you all 26-27 January, in Palo Alto.
google on-air https://plus.google.com/events/caeods2ad9ksksmphp767icgfhk
- 8:00 – 9:00 Breakfast
- 9:00 – 10:30 Talk one: (James Haselmaier will lead the discussion) establish a process by which longer term vision and product direction can emerge from within the community, http://lists.openstack.org/pipermail/product-wg/2014-December/000051.html along with a 2-3 architecture plan
- 10:30 – 10:45 break
- 10:45 – 12:15 Talk two: (Sean Robertss speaker unless a TC member could be found) Tagging approach to project maturity, http://ttx.re/the-way-forward.html, TC spec https://review.openstack.org/#/c/138504/5/resolutions/20141202-project-structure-reform-spec.rst
- 12:15 – 2:00 Lunch
- 2:00 – 3:30 Talk three: Rob Hirschfeld as speaker) defcore and refstack. havana capabilities and designated sections are set. What does this mean? What’s in store for icehouse and beyond?
- 3:30 – 3:45 break
- 3:45 – 5:00 Continue any from above
google on-air https://plus.google.com/events/cadr4ncpo1psbrkf7h2jiurh93c
- 8:00 – 9:00 Breakfast
- 9:00 – 10:30 Talk four: (Gary Kotton as speaker) Neutron services split https://github.com/openstack/neutron-specs/blob/master/specs/kilo/services-split.rst
- 10:30 – 10:45 break
- 10:45 – 12:15 Talk five: (Sean, Allison, Rob, Stefano will lead the discussion) discuss this group’s charter. Establish cadence and forum for regular meetings of this working group 12:15 – 2:00 Lunch
- 2:00 – 3:30 Breakout sessions:Crossproject specs and efforts this cycle
- Logging rationalization efforts/direction including https://review.openstack.org/#/c/132552/
- OS profiler including https://review.openstack.org/#/c/134839/
- Tempest and Grenade as deployment/upgrade validation tool
- Telco group and Public Cloud Providers architecture/strategy/use cases
- (Arkady lead) interleaving Enterprise readiness into themes, maturity and longer term architecture
- any others?
- 3:30 – 3:45 break
- 3:45 – 5:00 Continue any from above
Agenda details on etherpad https://etherpad.openstack.org/p/kilo-product-management-midcycle
You can still RSVP for this event here.
RSVP for the OpenStack Product Management Kilo Midcycle Meeting here.
It’s time to finalize the OpenStack Product Management topics for the Kilo midcycle meeting coming up. We will need speakers for each topic. I want to get the best authority possible for each talk, so locking down the topics is required. I will need to get their commitment as soon as possible.
The topic options are on the etherpad here. My choices for topics are multi-release strategy, project tagging, neutron services split, DefCore and RefStack, and cross project specifications. I will post this call to the mailing list. I need you, the attendees, to respond with your priority topics and any speaking offers. Respond by mailing list, tweet, comment on this post, or by the etherpad.
Review the Latest OpenStack ABCs at the Product Management Midcycle
The group called the OpenStack Ambassadors was created to recognize the active user group leaders worldwide and promote their leadership for other user groups. More details on the group can be found here. We met during the OpenStack Kilo Summit in Paris and developed a plan for 2015. Find that plan in detail here. We decided that our highest purpose is simply to mentor OpenStack user groups.
That’s me, second from the end on the right.
As part of the Foundation support, Martin Kiss is creating a new site dedicated to the support of the user groups. It should be ready by the Vancouver Summit. We also plan to have user group resources such as starter packs. Going forward we are meeting regularly around building and improving the OpenStack user groups. This should and could be the easiest way for the Ambassadors to help other user groups through communicating. We are meeting alternating Weekly on first, third, fifth Tuesdays at 08:00 GMT and second and fourth Fridays at 18:00 GMT on the chat.freenode.net IRC channel:
#openstack-meeting-alt. The meeting details can be found here.
If you are part of the OpenStack community, jump on the train and join us!
There are three important bylaws changes that require 25% of the 13K plus OpenStack Foundation members to vote in order to pass. These changes are around the efforts to update the OpenStack trademark process, adjust the size of the Legal Committee, and change the percentage of members required to vote in an election. Find the full breakdown of the proposed bylaws changes here. I voted yes on all three.
We also have 8 open OpenStack Foundation Individual Director seats. It was tough for me to reduce the vote to just 8, as there are many valuable voices running this time. It is a good problem to have. Find the full list of candidates here. I will keep my vote on this private, as I am a current Director, and I want represent everyone in the community as equally as possible.
Check out Rob Hirschfeld’s post on the same topic. He is also running for one of the director seats.
If you are an OpenStack Foundation member, check your email, and VOTE!
Link to the previous training post here
Community training has always been an effort centered around giving the user groups tools to teach. As our efforts progressed, our content has evolved. My involvement with the OpenStack Ambassador program has helped by given us more visibility into what the other user groups are looking for as well. We have changed the way we looked at training and so the training team and content is changing with it.
The training team got together over the Kilo summit in Paris. Here is the result https://etherpad.openstack.org/p/training-guides-kilo-summit. From our summit meetings, we boiled our goals down to a few:
- Most of the content in HTML slides using example scenarios and use cases
- The Upstream Training content will largely become the Developer Training Guide
- Use OpenStack training clusters built by an automated local installation on VirtualBox (via osbash scripts) which follows the official OpenStack install-guide
- Developing a one node training cluster version (all-in-one) using public cloud
- Online Quizzes
- Our target audience is the OpenStack User Groups, Upstream Training/OpenStack Summit training, Universities, and Self-paced training
- We will continue to support the OpenStack Documentation Program projects by cross posting content
As part of the newly revived effort, I created a new team called the Trainers team. This team is meant to be made of the people that donate there time teaching at the user group events. More on this in another post coming soon.
All the rest of the training details can be found here in the Training Guides wiki.
Look to the previous post for background on OpenStack Product Management
OpenStack is a very active community. Bursts of change happen quite often and it can be difficult to keep up if you are not immersed. We need manage the flow of critical information and decision making just like any other engineering organization. The people that represent the Product Management of OpenStack are a critical group within the community. The next OpenStack Product Management meeting is scheduled to happen 26-27 January 2015 at VMware, Creekside A, 900 Arastradero Road, Palo Alto, CA, 94304. Directions to VMware Palo Alto are here.
Anyone interested in participating can RSVP here.
There are many things to discuss. We have pulled a few important topics from the summit forward into this list. Some new issues to discuss, have bubbled to the top. The best topics are going to be about something we can do something about. e.g. Focusing more developers on a key blueprint for Kilo release, Major changes to the way the TC deals with projects. We have hour and half time slots for five major topics and one two hour breakout session. Let’s fine tune this list down to the five must important.
Add comments to this post and/or on the OpenStack Product Management mailing list
The information above is pulled from the working agenda here.