Meeting:Community Architecture 2008-09-29

Community Architecture Meeting :: Monday 2008-09-29

Roll call

Present: MaxSpevack, GregDeKoenigsberg, PaulWFrields, KarstenWade, JackAboutboul, HarishPillay, MichaelTiemann

Respins

Paul brought us up to speed on the remix questions.

Greg and Paul need to chat about Fedora trademarks and OLPC.

Goals

Max indicates the pressing need of getting our goals updated.

Tiemann suggests articulating both in qualitative and quantitative terms the value that we provides. Separate that which we can define and measure from that which we must accomplish. Provide the organic context from which our other products and capabilities are derived.

Jack ran-down the University/Education status, and Max took notes directly on the goals page.

Additionally, work is progressing in APAC with university ties to OSS. (Maybe Harish can add some more details? I can only type so fast.) What would be a best-cost open courseware strategy? We need to underpin open courseware. We need to simplify the nature of the relationship to educational institutions around an open courseware system.

Seneca:

• Whatever work comes directly out of Seneca students in the Fedora Project.
  • One of the Seneca students is working on Java packaging.
• FSOSS will be where some "consortium for open source and education" is founded.
  • Metric for success at FSOSS is to have the foundation started, which will lead all this open source education stuff.

Tiemann: if and when local universities are priming the local government pumps with open source software so that every GIS, statistics, etc. question can be answered with open source infrastructure, that is a win. We can measure specific contributions to specific projects, but if a school with a $2M endowment for teaching OSS rewires their state's government with open source, that is a major win condition. Collaboration between education and government. Priming the pump is just as important as the specific technical achievements that are made. How our actions today enable or allow the possibility of changes that will still be relevant in 20 years. Things happen on different cycles -- 6 months to 20 years.

Fedora as a "technical election process". How it leads Red Hat in the right direction, and sometimes leads others in the wrong direction. If we are acquiring companies for millions of dollars at a time, the analytical function that Fedora provides is incredibly important. Fedora/Puppet relationship. Led to Matt Hicks and Everest. Fedora itself functions as a metric on the community. "Cost of election and voting machines" analogy.