Release process. We've talked to the community about the release process from DevOps and community stand points. What's it like from production?

With production it's a lot of communication between the US and UK, who are the primary live release studios. For 2.5.0 there was a long list of blockers that Chris normally says 'we can't release with these'. They identify those, if it's not in Jira they put it in, then they find resources for the bugs. Lots of Jake's job is making sure the blockers get a high focus from the resources assigned to them. At the end of each day there's a hand off, saying these are the remaining blockers, these are what're being handed off, etc… Everything gets handed around between the studios, so there's no real lapse in communication, no situations where a bug will sit for days without getting looked at. It's always being picked up and the highest priority issues get called out to focus on most. As the blocker list gets whittled down, they look at criticals. Then they look at 'high' bugs, etc… At some point they see they've got most of the bugs fixed, and they pick a release date. Then they keep working, and inevitably thing go wrong, and fixing one bug causes more, so they push the date back because blockers have to be gone, and you can't always schedule for things. At CIG they're a lot more forthcoming with that information than normal, they tell us lots of what's keeping them from releasing things. The word of the day is communication. They all talk to each other, all the time. Jake gets tons of emails, QA has 'flair' threads, where they point out the new blocks, Jake hands those off to the UK, etc… lots of communication to get releases ready for live.