With the Google Summer of Code season upon us, I’ve decided to stick my nose right in to the situation and help run a GSoC for the CentOS Project. I was co-admin for Fedora for 6 GSoC sessions, a mentor for a few specific projects including the kick-off for Transifex, and ran the one-off Fedora Summer Coding 2010 the year Fedora didn’t get in to GSoC. I would LOVE co-admin assistance, this is a fun effort that is well-worth sharing with other contributors. Let me know if you want to help in any way. Note that a few admins and mentors from each project are invited each Fall to Google headquarters for the Mentor Summit. I’ve attended a handful of times, it’s an absolutely fantastic event.
So far my understanding and history lesson for the CentOS Project is …
- CentOS applied at least one time in the past;
- We were not accepted;
- That is all.
Please let me know any details, subtleties, or differences to my history-so-far. I have seen URLs to some wiki pages but haven’t got the #acl to read them yet.
My plan is to do the usual what-works that I learned as an admin for Fedora for 6 years, as well as what I’ve learned of the open source way. This means:
- We need an ideas page, with an idea paired to a mentor. For example, the 2013 Fedora ideas page.
- We need students to be dreaming up ideas — especially students who are already users or contributors to the project — and get those ideas out to see if there are likely mentors amongst other contributors
- It doesn’t hurt to ask about your idea!
- We need to be thinking in a realistic and useful way about the GSoC potential — what can it really mean, how can we use it to the project’s advantage.
- Then we need to fix the ideas page on the discussion from 2) and 3).
- Do all the rest required by deadline to apply as a mentoring organization (I can do that.)
- Magic! (If we’re lucky; I’ll monitor the process and shepherd it through.)
- Magic requires us to find and engage with students.That’s what is next … first let’s get accepted!
The only time my organization didn’t get in to GSoC … we hadn’t finalized the ideas page (partially because of my mistaken thinking.) If the CentOS Project can do the basics, and we have good ideas, CentOS has a fair chance of getting in.
Where it comes to “what the heck can students do that helps CentOS?!?”, I have this small list to spark ideas; I’ll be opening a new thread on centos-devel to discuss all ideas in full:
- Anything a variant SIG needs to get ready, such as packaging or working through a Software Collections setup.
- Coding work related to QA.
- Debranding from the RHEL upstream.
- Any set of “does it work on/at/with $X” testing.
- Test harnesses such as from the Beaker project.
- Connecting documentation back up to Red Hat Enterprise Linux docs.
- Additions to the Core SIG work.
- Work on the new open community build system (possibly including Koji, COPRs, and Software Collections upstream work that CentOS absorbs.)
- Note that Transifex was started to solve a Fedora Project problem, and is now an entire project and commercial entity.
- Any tool that is built for a specific purpose to help the CentOS Project.
- And … what can you think of?
If you are a student with an idea looking for a mentor, please let me know. I will post it and help you look for a mentor. Ideas from students have been very successful in the past!
I’ll also note that existing community members who are students are qualified to enter the program, and your existing relationships may be useful for your efforts. Email me if you have any questions about this.
This should be another great GSoC year, and I hope the CentOS Project gets a chance to participate as a mentor organization!