This post was inspired by the pre-event interviews that exist on opensource.com.
Flock is a Fedora-defining event of the year. In the lines below, a few speakers share what the conference means for them and what they’re looking forward to the most.
Stephen Gallagher is a long-term Fedora contributor, member of the Fedora Engineering Steering Committee, and a UNIX/Linux developer. His current focus project is OpenShift, which is a Platform-as-a-Service (PaaS) cloud product that lets developers quickly develop, host, scale, and deliver apps in the cloud. He has a workshop on Friday, August 5th.
1) What would you like to achieve at Flock 2016? In other words, what outcome would you like to see?
In my view, the primary purpose of Flock is to energize the Fedora community for the upcoming year. We gather our flock together in one place, pat ourselves on the back for what we achieved over the previous year and then get hard to work on the next steps.
In the case of me personally, I’m hoping to see renewed interest in the modularization efforts of Fedora, particularly within the Server Edition. There are several sessions scheduled to focus on this and I think this will be a make-or-break opportunity for it: either the community will come together and start making it happen or we’ll poke holes in it and come up with something better. Either way, Fedora wins!
2) What have you learned during your membership in the Fedora Engineering Steering Committee?
FESCo has taught me a great deal about the wider open-source community. One of the historical issues with open-source is that it tends to develop very insular “silos” around particular technology. This in turn leads to projects that are generally very well-thought-out internally but are very difficult to combine into a larger solution. (As a contrived example, a web server developer is probably going to be most concerned with the performance and options of the server for all cases and therefore customizing it for a specific workload may be very complicated.)
Working on FESCo, I’m required to always be thinking about how new technology fits into the overall mission of Fedora (or whether it fails to do so). In the course of my service, I’ve had the opportunity to learn about a great many projects (and how they interact) that if I had stayed working on only my own day-job tasks I would never have known about.
3) You were one of the organizers of the first Flock (right?). What has changed over the 3 years?
I was not one of the original organizers, though I have been to every Flock thus far. I’m also planning to prepare a bid for 2017.
In truth, I can’t think of much that has changed about Flock itself over its three years, except that the organizers have gotten practice and now it feels more professionally put-together than it did at first. What I mean by that is that in its early days, people were still trying to figure out how much of FUDCon to retain in the process.
I think our first Flock (at Charleston) had one obvious problem that nearly every attendee remarked upon: too much content. That sounds like a good problem to have, but what it ultimately meant was that with five or six concurrent sessions running at all times, most of them were poorly-attended or attended only by people who were already familiar with the topic being discussed. This basically amounted to a failure to use the conference to advertise and expand these topics to new people. In subsequent years, Flock has done a much better job of scheduling.
4) Besides Modularity, what do you think are other hot topics at the upcoming Flock?
I think it’s pretty clear that there’s going to be a lot of coverage of Project Atomic, ostree and the upcoming Atomic Workstation efforts. That’s exciting stuff and I think we’ll see a lot of attendance there.
Docker topics will be common as well, though I personally wonder if that’s reaching peak publicity. I think many of our attendees have already drunk that Kool-Aid, so I’m not sure how attendance will be there.
In terms of non-technical topics, I think there’s going to be an opportunity to really focus on our outreach efforts: everything from minority and gender outreach to university and high-school programs.
All in all, there’s a lot of great content coming up at Flock (if you were to look at my sched.org, you’d see there are no time-slots that I don’t have a session scheduled that I want to see!)
5) Could you give us a teaser of your talk (that doesn’t give it away, of course ? (unless the abstract is enough)
Actually, no 🙂
While I will be actively participating in several sessions (notably the “State of the Fedora Server Union” and the “PRD Workshop”); a change in my day-job responsibilities at Red Hat meant that I ended up passing the sessions I was originally going to lead on to another person.
However, I can talk about them anyway (since I collaborated with Michael Wolf on the content).
The “State of the Fedora Server Union” is a talk I gave at the last two Flock conferences. (Fedora Server Edition as a concept was actually born at the first Flock, along with Workstation Edition and Cloud Edition). It’s essentially a two-part, 50 minute talk: the first twenty minutes are a recap of all the cool stuff that the Fedora Server SIG has produced since the last Flock. The remaining 30 minutes are an interactive session where we will put together a 50,000 foot view of what we want to accomplish between now and the next Flock.
Later at the conference we will have the “Fedora Server Pow-Wow”, which is a two-hour working session to kick off the new ideas that come up during the State of the Union session. This will be a place for people to come and start talking about implementation of these new ideas, ideally throwing together a first jab at a proof-of-concept for it.
6) What do you think is a good way to attract more contributors to Fedora?
Kidding aside, I think one of the things that we’re finally starting to understand is that people don’t find the sausage factory very attractive. The next generation of developers and open-source ambassadors no longer look at the platform and say “I want to be a part of that!”. Instead, they look at the interesting things being done on top of those platforms.
A few pieces like Docker have brought us back to understanding that the platform has value, so that’s been helpful (and is probably a big piece of why every conference — Fedora or otherwise — apparently has to dedicate at least two sessions to Docker).
However for the most part, what attracts people is doing something cool with a technology. This is where I think we need to be doing more to advertise. I’ve been glad to see the recent series of “How do you Fedora?” on the Magazine. That’s a great first step, but unfortunately so far it hasn’t been tremendously “sexy”.
One of the things that Matthew Miller brings up in a lot of his talks that I feel like we aren’t capitalizing on is that Fawkes Robotics is using Fedora to compete in world championship robotics competitions. The fact that this isn’t general knowledge about Fedora is a huge oversight, IMHO. We need to get Tim Niemueller to do one of those “How do you Fedora?” articles, stat.
Brian Exelbierd is an Open Source proponent who does software engineering in the area of container technologies. He’ll be speaking on August 3rd and 4th.
What would you like to achieve at Flock 2016? In other words, what outcome would you like to see?
I’d like to see if there is interest in adding the ADB as a Fedora project, probably in Fedora Cloud as a way of advancing open source and attracting new users. I’d like to get more people interested in contributing to Fedora Docs as a lot of work is going on to make that even easier. Lastly, I’d like to solidify friendships and make new connections to parts of the project I have no familiarity with.
Honza Horak has been a contributor to Fedora for over 5 years. He has put a lot of effort in getting the Software Collections into Fedora. He will deliver a talk about Application containers and System Services on August 2nd.
Container technologies evolve so rapidly that there is always something better on the TODO list than writing a blog post or documentation where I could share the recent findings we experienced during application containers development in Red Hat.
On the other hand, sharing those experiences, ideas how to do things correctly and how not to do things because we failed, is something that other developers will benefit from once they turn to the dark side of containers development as well. And many will eventually do, even if they don’t think so yet.
Anyway, that’s basically why I’m going to talk at Flock – I want to share some lessons we learned during our journey from having Software Collection packages ready to having application images based on those packages running the same in Open Shift and also on bare metal. Since the journey has been full of questions and not every-time clear answers, I’d like to hear opinions of others about the way we chose, whether it is something we can do the same in Fedora (which I really believe in).
That brings me to the most important part of my Flock visit: I want to talk to people interested in containers development in Fedora about ways to improve the containers maintenance experience and how to improve Fedora-based container images made available by Fedora contributors.
One specific aspect is that all application containers produced by Fedora contributors should run on OpenShift smoothly if possible, which might sound weird initially, but it is actually the correct platform where such images make sense in the end. Anyway, as I mentioned earlier, I want to hear the opinions of others in the first place.