We took a post-Flock survey of attendees to get some feedback on what you like and what we could do better next year. Most of those offering feedback, 71%, were attending Flock for the first time in 2014. Given the continental switch, this isn’t too surprising as a likely indication of overall attendance. Because of the proximity to Brno (where there’s a large office full of Red Hat engineers) and the large number of Fedora fans in the Czech Republic in general, we saw a significant audience from within the country, with around 250 attendees total.

One of the most significant changes we made from last year was the number of sessions held. For the first Flock, we accepted nearly every submission and worked them all into the schedule. This year, we held fewer concurrent sessions, and 70% of respondents thought we had just the right amount. A few of the comments noted oddities about the schedule. In a few cases, these were intentional, but when you saw strange blank spots, this was usually due to late drop-outs.

As for the most popular sessions, the most frequently mentioned was Sean Cross’ keynote on the Novena. The following talks received more than 2 mentions in the survey (total mentions in parentheses) as an attendees’ favorites:

  • Fedora.next Joint Session, Matthew Miller (6)
  • Governance of Fedora under Fedora.next, Haïkel Guémar and Toshio Kuratomi (5)
  • New features of rpm and dnf, Jan Zeleny (5)
  • Fedora for Developers, Langdon White (4)
  • State of the Fedora Kernel, Josh Boyer (4)
  • Advocating Fedora.next, Christoph Wickert (4)
  • Where’s Wayland?, Matthias Clasen (4)
  • Taskotron and Me, Tim Flink (3)
  • Release Engineering and You, Dennis Gilmore (3)
  • Fedora Server Role-ing Along, Stephen Gallagher (3)
  • Review Server: Package Reviews without Bugzilla, Pierre-Yves Chibon/Stanislav Ochotnicky (3)
  • Python 3 as Default, Slavek Kabrda (3)
  • Through a Wire Fence: Fedora and RHEL Paul Frields (3)
  • Fedora Infrastructure present to future, Kevin Fenzi (3)
  • ARM Server Update, Jon Masters (3)
  • Evolving the Fedora updates process, Luke Macken (3)
  • Building an application installer from the ground up, Richard Hughes (3)

The rest of the survey’s questions were open-ended, but I’ll answer a few that were mentioned repeatedly:

  • Speaker preparedness: Several commenters noted that they attended talks where speakers seemed uncomfortable giving a talk or ill-prepared on the topic. We’re interested in hearing your suggestions on the best ways to bring you better-prepared speakers next year.
  • Names on voting: Related to the above, one person suggested having submitters’ names on topics when talk proposals are open for voting. The Flock committee chose not to do this intentionally to make the conference more open as well as to prevent discrimination. We believe that the benefits of making submission voting anonymous outweigh the benefits of adding names. There are other ways of ensuring expertise in an area. (Voting is the most significant but not the sole factor in whether a talk makes the schedule.)
  • Scheduling choices: There are only so many ways to make a schedule, and we have to consider many factors. We try not to schedule things that are of interest to the same general community (e.g. not having two kernel talks at the same time), but space availability and speaker availability don’t always allow it. Sometimes at a conference, you have to make hard choices. It’s a sign of a good conference that you want to attend too many things. It’s a sign of a bad conference when you spend too many sessions sitting in a hallway because there was nothing you wanted to see.

Thanks to everyone who filled out the survey. If you have other specific concerns or things you’d like to see next year, please don’t hesitate to let me know. And remember, you have only 11 more days to finish your bids for Flock 2015!