Aeolus Developer Conference Report. ACKS REQUIRED BY NOV 13.

Hugh Brock hbrock at redhat.com
Sat Nov 10 10:37:59 UTC 2012


Hello Aeolus community!

You've seen some mails over the past week about the Aeolus developer
conference. Hopefully you noticed my youtube links on #aeolus and so
on. We didn't do as good a job as I had hoped we would with sending
out periodic minutes, however, so I wanted to give a brief summary
here as well as detailing some important next steps that various folks
at the conference proposed we take throughout the week.

First I want to give a huge thank you to all the folks who put
together talks for the conference. In no particular order:

Richard Su -- Aeolus API project, Aeolus dev-tools setup, Aeolus
  upstream CI testing with Travis
Martyn Taylor -- Rails Engines, TIM (template and image management)
Martin Povolny -- HW Profile cost manager
Jiri Stransky -- Aeolus components and tech committee representation
Steven Hardy -- Heat overview and demo
Tomas Sedovic -- Conductor + Heat, how and why
Imre Farkas -- Automated provider selection
Ian Mcleod -- Image Factory overview
Jeremy Perry -- Conductor permissions UI
Jaromir Coufal -- Conductor navigation and list/detail/baseball-card views
Dmitri Dolguikh -- Katello architecture and UI
Tzu-mainn Chen, Jiri Tomasek, Matt Wagner, Francesco Vollero --
  community development
Scott Seago -- Alberich rails engine for authorization management,
  stateful instance support concepts
Steve Linabery -- Tagging and packaging with github, productizing with
  Brew 
Matt Booth -- V2V how and why, v2v future plans
Angus Thomas -- Rationalizing Aeolus object names, Winged Monkey concepts
Mike Orazi -- Organizing upstream projects with github, github issues
Michal Fojtik -- New developments in Deltacloud
Jan Provaznik -- Deltacloud State Tracker
Andy Smith -- Customer requirements and issues
Hugh Brock -- Fixing Aeolus upstream, Aeolus governance

As you can see, nearly the entire team presented on topics from
community building to permissions infrastructure. The presentations
were uniformly well done and really helped give the entire team a
better insight into what the various components actually do.

Second: My major agenda item for the week was Aeolus Project
governance and developing an upstream. I'm happy to report that the
group was able to agree on a set of proposals to take to the list for
comment and hopefully approval. To wit:

* We split into teams, each of which came back with a proposed Aeolus
  mission statement. After much discussion we refined the four
  proposals into one which reads:

  "Aeolus mission: to provide superior tools and workflows for flexible
  construction, management, and monitoring of multi-instance systems
  across clouds."

  Please ACK or NACK-with-suggestions the mission statement by
  November 13.

* Resolving direction and technical disagreements among the various
  Aeolus projects has become problematic and often leads to
  unresolved, festering problems. Therefore the group proposes that we
  establish a Technical Cabal, charged with having short, weekly
  meetings to set technical direction for the project and debate and
  vote on technical and architecture details (What kind of API should
  be used here? Does this requirement merit a separate component or
  can it be satisfied by project X? etc.). Jiri Stransky will be
  sending out a mail with a proposed membership list, which I would
  like the community to ACK or NACK this week. In particular, we've
  nominated at least a few folks who were not at the meeting; if
  you're not interested in serving, fine, register that on the list
  and we'll accept nominations for someone else. I will schedule and
  chair the first couple of meetings until the Cabal can pick its own
  Awesome (the term we, in a moment of insanity, picked to use instead
  of "chair"). If you were nominated for the Tech Cabal, please accept
  or decline by November 13. If you would like to nominate yourself or
  someone else, please do so by November 13.

* We agreed that three other Cabals would be useful: a Website Cabal,
  charged with keeping the website up to date; a Publicity Cabal,
  charged with managing our appearances at conferences, social media,
  slides, papers, etc.; and a Release Cabal, charged with getting
  upstream releases pushed out and documenting the required processes
  for the rest of the team. Jiri Stransky will have proposed
  membership lists for those Cabals out on the list shortly. If anyone
  else would like to join one, just ask. We nominated acting
  Awesomes for each Cabal to get things rolling until a permanent
  Awesome is chosen.

* We agreed that in the future the membership of each Cabal would be
  determined by nomination and on-list ACKs, and that terms of
  membership should align roughly with release boundaries. Members are
  free to serve for as many terms as they like however if re-elected.

* We agreed to propose a roughly six-month release cycle, subject to
  adjustments as needed by the Release Cabal. The next two releases
  will be on a bit shorter cycle to get us better aligned with
  downstream release boundaries: the Release Cabal will propose a
  release integrating Factory 2.0 and Deltacloud 1.0.x by the end of
  2012 (i.e. two sprints away), followed by another release around the
  end of March which will hopefully include a complete Aeolus API. The
  first couple of meetings of the Technical Cabal will finalize the
  feature priorities for those two releases.

* Each Cabal should have its first meeting the week of November
  12. Meetings can be via hangout, IRC, or whatever the Cabal chooses,
  but minutes should be published to aeolus-devel. 

I'd like everyone with commit access to an Aeolus project to respond
with either ACK or NACK + suggestions to the above proposals, and to
the proposed Cabal membership lists, by November 13. I will consider
failure to respond by November 13 an ACK, as well as an indication
that you're not paying attention.

Third: Matt Wagner, the acting Awesome for the Publicity Cabal,
has volunteered to collect all the presentations given and make them
available on aeolusproject.org. I'll take charge of getting him the
ones I already have, but if you didn't send your slides to me before,
make sure you send them to him now please.

Thanks once more for all the hard work that went into the week. I am
very excited to see what comes next as we begin pushing planning and
feature development upstream and making our releases available to a
wider audience.

Take care,
--Hugh


-- 
== Hugh Brock, hbrock at redhat.com                                   ==
== Senior Engineering Manager, Cloud BU                            ==
== Aeolus Project: Manage virtual infrastructure across clouds.    ==
== http://aeolusproject.org                                        ==

"I know that you believe you understand what you think I said, but I’m
not sure you realize that what you heard is not what I meant."
--Robert McCloskey



More information about the aeolus-devel mailing list