Red Hat is a synonym for "Upstream first", right? Let's have a look at a QA project that was developed behind closed doors, but was ultimately upstreamed due to the effort of two young students. After all the effort they went through to upstream the project, later on they found that other teams were using forks of the project, keeping fixes and even similar tests to themselves. Once again it was time to convince people that upstreaming code has compelling advantages.
This path began for me around the year 2009 and is still ongoing. I'd like to share our story to let you learn from our mistakes as well as successes.
The project I'm talking about was called Autotest, but evolved via KVM-autotest, run, virttest all the way up to Avocado/Avocado-vt. But don't let that distract you. How to deal with upstream code in a community and how to handle collaboration is the true subject of this talk.
Anyone interested the process of going from downstream to upstream, learn what worked, what needed to be tweaked or even reworked and whether it is worth it. Sometimes even an intern/student can be the last drop that pulls the avalanche.
This is a talk of 60 mins.
Metadata Update from @bex: - Issue tagged with: Talk In Consideration
I regret that we are unable to accept this talk this year. There will be lightning talk sessions and I encourage you to consider this option.
Metadata Update from @bex: - Issue status updated to: Closed (was: Open)
Login to comment on this ticket.