This is a part of the FESCo Elections Interviews series. Voting is open to all Fedora contributors. The voting period starts on Thursday, 19 November and closes promptly at 23:59:59 UTC on Thursday, 3 December 2020.

Interview with Dan Čermák

  • Fedora Account: defolos
  • IRC: defolos (found in fedora-devel, #fedora-i3, #fedora-join, #fedora-meeting, #fedora-meeting-1, #fedora-meeting-2, #fedora-meeting-3, #fedora-meeting-4, #fedora-modularity, #fedora-python, #fedora-qa, #fedora-ruby, #fedora-sway, #fedora-arm, #fedora-iot, #opensuse-factory, #opensuse-buildservice, #rpm, #silverblue)
  • Fedora User Wiki Page


Why do you want to be a member of FESCo and how do you expect to help steer the direction of Fedora?

Fedora has been my favorite distribution for the better part of a decade and has pulled me further and further into its community. After becoming a package maintainer and getting more and more involved, I wish to give back more and shape the future of the project that is so dear to many of us.
I would especially like to apply my experience gained from my dayjob and my involvement with the openSUSE community, the Open Build Service and openQA. These have given me valuable insights into the tooling of other distributions and how they compare to ours. I would like to help improve our current tools, especially with the focus on more automation (where applicable) and especially to streamline and simplify the packaging workflow, so that Fedora will become the distribution with the best packaging experience.

How do you currently contribute to Fedora? How does that contribution benefit the community?

I am currently mainly involved in the creation of the i3 spin of Fedora as part of the i3 SIG, where I contribute as the maintainer of i3 & related utilities and test automation with openQA. With this we hope to create a polished spin of Fedora that will help increase the projects reach as other Linux distributions have done with their respective editions.

I also maintain a bunch of ocaml packages with the end goal to package Facebook’s static analyzer infer for Fedora (an initiative that was kicked off and is mostly driven by Jerry James) which will benefit developers who just want to install and use infer without having to resort to using brew or pulling down docker containers.

Besides maintaining and co-maintaining a few other packages (Emacs, a few python packages and some sway related utilities), I also contribute to Fedora’s automated image testing infrastructure (openQA). This is at the moment mostly limited to the i3 spin, but once we have the initial test coverage setup, I intend to contribute to other test cases as well and help expand it (preferably more into the ARM & IoT world).

How should we handle cases where Fedora’s and Red Hat Enterprise Linux’s needs conflict in an incompatible way?

I don’t think that there is a one-size-fits all answer to this problem. My short answer answer is: we all should try to talk to each other more and strive to find a common ground as the relationship between Fedora and RHEL is mutually beneficial and should not be brought out of balance. However, I will strongly object to any changes that will have a mostly negative impact on the Fedora contributors, if such a balance cannot be found.

However, the past has in my humble opinion shown that even controversial cases (such as the push for Modularity and the Git Forge decision) could have had a less negative aftertaste if we all talked more to each other. The Modularity initiative is imo extremely useful and has a real value for EPEL users: it provides them with means to have a stable distribution while also having a few up to date packages. But the initial issues resulted in a lot of pain for Fedora packagers, who do not benefit a lot from modularizing their packages. I believe that we should have spent more time at the drawing board albeit a lot of problems did not appear until after modularity was deployed. This would have also helped RHEL, which is currently stuck with a sub-optimal implementation of the modularity concept.

Another case from the past year is the controversial git forge decision to drop pagure. While I personally don’t think that going with GitLab is the best way forward for Fedora, I can on the other hand not blame the CPE team: they are already stretched too thin to also maintain the code base of pagure. My main gripe with the decision is the initial poor communication from both sides (I have put in enough oil into the fire myself, for which I am sorry now). The reasons for the decision are understandable and at the end of the day, we as the community must make the decision: is this important enough for us that we will take over maintenance of pagure?

All my ramblings aside, I do believe that if conflicts should appear, that we can work together to minimize their negative impacts. If that be impossible, then my choice will be the one where the life of all Fedora contributors is affected in the least negative way.

What else should community members know about you or your positions?

I’m an open source enthusiast and have become very passionate about testing and automation, which are the areas where I would love to bring Fedora forward. I am also quite involved in the openSUSE project and would like to collaborate further with the openSUSE community as there is already a lot of common ground due to our common heritage. My dream is that we will at some point have a build system as good as the Open Build Service (without all of its complexities and quirks).

Besides that I am also very interested in the ARM side of the world and am looking very much forward to running Fedora on my Pinebook Pro and eventually on a Pinephone or some other Linux compatible smartphone.