Where are all those projects? Not on the Fedora ideas page.

Folks, check this out – your idea doesn’t have to be complete, or even contain much more than a few sketchy use cases. Sometimes you just need to spark an idea in the student, something to become passionate about.

If you aren’t sure if you can mentor, put your idea up anyway. If a strong student or three shows up, mentors can be found.

You may wonder why this matters so much to me? Quite simply, we missed our chance at being in last year’s Summer of Code because we didn’t have an ideas page.

I look at the page now, there are great ideas there, but not very many of them. I know that the Fedora Project and JBoss.org folks have many more ideas in them. A strong ideas page is an essential part of the application.

If you are interested in helping Fedora/JBoss.org to get in to this year’s Summer of Code, or want to help run the program after that, join up to the cross-project mentor list and ask what you can do.

We kept a log (HTML log and HTML + plain text summary), had a plan, and completed the important bits.

Early the following week I did a crossdump for Mel Chua.  We are doing some work-swapping for a few weeks.  I’m helping get Fedora Insight to the next level (staging, ready for production), and Mel is working on the next stages of work the Summer Coding SIG needs.

Within the next week you should start to hear about what students, mentors, sub-projects, and upstreams need to be doing.  If you are involved in Fedora or the JBoss.org projects, help pass on the word.

I did some research, analysis, and reporting recently about the Fedora Project and (later) JBoss.org participation in Google’s Summer of Code project.  I’ve done various parts in our GSoC effort since 2006, the second year we participated, and the last two summers I’ve been focusing on strategic work for our overall summer coding efforts while working as an administrator.

Up to now, I was going off my instinct that it was a worthwhile way to spend my time.  In fact, until this last year, it really hasn’t been in my job role in anyway, just something I did because.  Now, as part of the Community Architecture team, my job is to think and act strategically.  Last few summers I put a lot of initial work in to attracting mentors and project ideas, with good results in 2008 that were largely repeated in 2009.

Based on the results to date, it’s clearly worth the time and effort.  In fact, it’s clear that with more of a focus on being an efficient umbrella organization, we can have possibly more students, working on more useful, interesting, and ultimately successful/useful projects.  If this translates to additional upstream efforts such as what has gone on before, this is a good benefit to the wider open communities.

I’m calling together a meeting on #fedora-meeting-1 sometime in the next few days, with some specific invites out to Toshio, Yaakov, and Sankarshan.  The former two were at the Mentor Summit a few weeks ago and came to camp out at my place on a nice October night.  That’s when we kicked off the idea that we needed to really step-up our efforts to be a better umbrella organization.  Sankarshan has been a mentor and mentor’s-mentor for a number of GSoC years; I think he has provided the most and very useful feedback to student proposals of any other mentor.

If you are interested in helping with this larger effort, let me know.

According to the timeline, 18 to 23 March is the time for students to talk about project ideas with the mentoring organization.  On 23 March, students can begin submitting proposals, which are due by 3 April.  Before that and until 15 April, mentors may ask students for more clarifications or fixes to the proposals.

So, we need mentors to review project proposals.  We need students with ideas that are not on  the Fedora list or JBoss.org list to bring those ideas to our discussion group, seek mentors, and prepare to propose.

One thing we can all be excited about is the new web app, Melange, which is being used for the program.  At last year’s Mentor Summit, Leslie Hawthorn showed us the internal application they had been using, and, oh, such pain.  Melange is built from the ground-up to support large scale projects doing work similar to the Summer of Code.  I expect we’ll start to see it appear in other locations that pair large numbers of students with mentors or mentoring organizations, such as open educational resource creation.  Or the pairing of developers with teachers in creating Sugar activities.

(Post updated with fixed link to JBoss.org ideas list – http://www.jboss.org/community/docs/DOC-13401.)

The call for papers deadline has been extended a week to 16 March.  If you’d like to get in front of potential contributors and help them to get to the next level in open source, then submit a talk in one of the ‘Summit or ‘World tracks.

Fedora has all the latest and greatest stuff that someone thought of in the shower in the morning and thought “this is the coolest thing ever”. [The] trouble is that you cannot support that for seven years. I think you mitigate the problem of a lot of crazy features by having the split model that we have at Red Hat.

Gavin, being interviewed about Seam by ZDNet (quote appears on page 2), makes other good points, but that is the main one I wanted to share with the Fedora planet today.   I call that “the HP moment” after a Hewlett-Packard commercial from years ago, where someone has an idea in the shower and goes to his HP laptop to write it up.

• 68 proposals, 45 mentors
• 12 accepted projects (student slots)
  • Approx 1 in 5 proposals rose to the top
• 25% of the accepted projects are about Transifex, which is a GSoC project from last year
• 16.7% are FUNC, which is funcing awesome
• 16.7% are JBoss.org projects, unsure on my reaction there …
• Lots of work to improve Fedora infrastructure and contributor interfaces

We’ve had a huge increase in the number of mentors. Where last year we were a little short handed to start, this year we have a very large pool, nearly eight times what I saw last year. Wow. My people are really showing themselves. I’m proud; I mean that.

We’ve had a slight increase in applications so far, maybe twice the number I recall from last year? These aren’t statistics, just a quick gut survey and recall. But definitely more students. We suppose there will continue to be new applications right up to the deadline. As I did in previous years, I am not going to review applications until the deadline. You can do as you wish. It is likely that we’ll divide up the proposals to have mentors focus on one or just a few applications.

If the student application deadline is extended, that would be good for Fedora and JBoss.org. We could use a bit more time to get the word out and attract more students. I suspect Fedora is a bit hard to categorize for students. It is downstream of many projects, yet there are many, many Fedora SIGs and other sub-projects that it’s hard to know where to start. JBoss.org is equally difficult in a different way. There are 42 projects currently on the JBoss.org projects page, and some dozen plus ideas on the SoC ideas page. It’s very focused as a project (application server and components), but it is a large project with hundreds of contributors creating a broad-range of components.

Thanks again, mentors, we’ll definitely have work for you in reviewing and helping to tune the proposals. Keep encouraging students to apply, especially if there is a week extension.

Fedora’s project ideas page is here, and JBoss.org’s is here.

This is the first year that the two organizations have worked together. This was partially at the direction of Google; with so many organizations vying for mentoring slots, it made sense for us to work under one organization. One of my jobs, since I sort-of have a foot in both worlds, is to help us work together effectively.

Now our work in both Fedora and JBoss.org is:

1. Fill up the ideas pages more
2. Talk with potential students
3. Answer questions and encourage people to make proposals

Interested students should be told to look at the ideas pages and the main Google Summer of Code pages.

One more tip: many of these students are new to open source projects. There is plenty of reading they can be doing to understand how the Google side works; make sure they do that reading and don’t get you to do it for them. You are looking for students who know how to ask smart questions and stuff like that. Or at least show the possibility of learning those skills quickly.

More to come via the Fedora planet, especially since Bob McWhirter, the JBoss.org lead, is aggregated there. Also, watch the Fedora Summer Coding 2008 pages. Mentors and those interested in mentoring should begin discussions on fedora-mentors-list.

http://exitcondition.alrubinger.com/2008/01/15/wanna-do-my-homework/

I just started reading Andrew Lee Rubinger’s blog, or maybe he just began blogging? Not sure, but he has a polished, funny, irreverent, and clean writing style. In particular, I appreciate his thematic connections between other parts of life and the funny pool of sweat, blood, and tears that is open source software. Write on!

(He’s also a JBoss guy working on the Enterprise Java Beans (EJB) 3.0 project.)