Yesterday we turned in our mentoring organization application[1] for the 2010 Google Summer of Code[2]. We won't know until next week if our umbrella organization is accepted, but we'd like to get ahead on preparing a press release. This idea came from a side discussion with Kara, Mel, Paul, and myself. The idea is to have a press release to out via the Red Hat press blog (http://press.redhat.com), and have it be usable by Fedora and JBoss.org. Below is the background, and here are some relevant pages. This is our plan: https://fedoraproject.org/wiki/GSoC_2010_plan This is the in-progress page with information for students, mentors, sub-projects, upstreams, and admins: https://fedoraproject.org/wiki/GSoC_2010 Fedora Project and JBoss.org have submitted together as a single umbrella organization, which was Google's preference in the past. We've worked up some new ideas and plans to fix problems from the past, and our goal is to increase both the quantity and quality of student projects. Bottom line is this: more quality proposals means more that we like and rank, which increases the quantity of student slots from Google. The better we can simplify and amplify our message, the more of that goodness we'll all get to enjoy. == Background == I'm working on this because last fall I wrote up a report[3] that revealed some surprising and pleasing results. After that, our team[4] agreed it was worth a portion of my time ongoing if we could accomplish several goals: improve 2010 GSoC as a proving ground; help push out the framework/methods to other projects (such as what RIT is doing with our assistance); take lessons and apply them to a larger idea of improving mentoring and our ability to take in new contributors in a similar way to students. For the last few GSoC years Fedora Project has been paired with JBoss.org under a single "Red Hat umbrella"; this was Google's choice, and reasonable from the perspective of their watching the program's size and budget. Our experience was mixed for two main reasons: no previous communication channels between Fedora and JBoss that didn't go through Red Hat; different community types and styles. JBoss.org is more like the Apache project - a series of upstreams, loosely or tightly coupled, that all sit on top of a large number of OSes. For this year, the already collected mentors began last fall in working out differences and laying plans for resolving previous problems. One example is that we are having mentors and associated sub-projects compete directly for student slots based on the strength of the student proposal. I.e., there is no direct attempt to divide the slots, some to Fedora, some to JBoss, to create an artificial fairness. A few of our mentors, Toshio Kuratomi and Yaakov Neemoy, attended the Mentor Summit last fall, and brought back ideas and specific plans from other umbrella organizations that have had greater success than ours has. For example, KDE has managed around 40 student projects for the last few years, compared to our average of 10. All of that has brought us to this point, where we need to broadly explain to various target audiences the value of working with our umbrella organization. - Karsten [1] https://fedoraproject.org/wiki/Community_Architecture [2] http://socghop.appspot.com/ [3] https://fedoraproject.org/wiki/GSoC_report_2009 [4] https://fedoraproject.org/wiki/Community_Architecture -- name: Karsten 'quaid' Wade, Sr. Community Gardener team: Red Hat Community Architecture uri: http://TheOpenSourceWay.org/wiki gpg: AD0E0C41
Attachment:
pgpfeSu5TjOUb.pgp
Description: PGP signature
-- marketing mailing list marketing@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/marketing
