A GNOME dev enters an Akademy and…

Felipe Borges and me recently went to Akademy after a conversation between the two of us three days before the KDE conference that went like:

  • “Ugh, seems there is no GNOMEr going to Akademy…”
  • “Let’s go?”
  • “Let’s go”

And so three days later we traveled to Wien to meet with the KDE community. On arrival, we were pleased by a friendly and joyful ambient on the pre-registration party, which had no registration at all! We were happy to know these issues don’t happen only at GUADEC.

I only knew from before Aleix Pol, the vice-president, Lydia Pintscher, the president, and Alex Fiestas, a KDE dev I would have paid to had him as flatmate when I was at university 😜. Fortunately, we started meeting all the people around in no time .

Shortly after revealing I was part of GNOME I was tagged widely as the “GNOME spy”, and that is not far from the reality! I was there to learn what they do better, and of course finding ways that we can collaborate.

What KDE & Akademy does better

  • CoC and photo policy. They have all of this already sorted out, we are being a little late at GNOME. However, with my board hat on, we (mostly others at the board) have worked a lot to make this a reality and I’m confident the online CoC will be created relatively soon.
  • Non tech talks. I felt Akademy had some more interested talks than GUADEC, at least for someone that is already more or less aware of the technical happenings at GNOME. Things like 2 talks on newcomers experience, a talk on running events locally, privacy software, different board/community reports, etc.
  • Trainings. They had training in public speech and also training on how to behave constructively in a community. Both sound like excellent ideas to be done at GNOME.
  • A process to include 3rd party software. It’s called “Incubator” and it’s being handled by Aleix Pol. This is something we lack, and there is no easy way to convert a 3rd party project to be a GNOME project, therefore losing potential new GNOME developers on the way. The “World” group at GitLab partially alleviates this, but doesn’t fix it.
  • Marketing contractors. They now have marketing people available for the community. It’s something we lack, although the engagement team does a lot of work to improve that I have the feeling it would be useful to have someone with this background working on certain things at GNOME.
  • Board turnover. They have a 3 year 1/3 turnover of the board. This is quite good to build long term goals.
  • Compartmentalization of projects. Not sure how useful it would be for us, they have separated plasma apps, rest of apps, etc. So they have a defined product called Plasma and can be provided as is to OEMs as a branded product done by KDE.
  • Gender and age diversity. I was gladly surprised by the amount of women and age diversity at KDE. Things like the LGBT dinner are good ways to help the situation, and I hope it keeps improving at GNOME.

There are other areas KDE does better because of not-easily-fixable handicaps at GNOME such as QT documentation being better than gtk+, etc. But! I also realized how much GNOME does better in some other areas, something you don’t realize until you see what other projects are doing.

What GNOME does better

There are three main areas that struck me as game changers.

  • Newcomers experience. Seriously, the difference is huge. Thanks to Flatpak, Builder, GitLab and GitLab CI we stepped up the game massively. Not only compared to KDE but also compared to any other complex software or software organization I have seen so far.
  • GitLab. KDE uses a mix of tools such as Phabricator, Bugzilla, CLI tools, Travis for CI, etc. For us, switching to GitLab (or any tool that would integrate everything into a single tool as good as GitLab) has been a massive win both in sysadmin maintenance and in development workflows.
  • Focus.  Historically KDE has had a quite broad focus, several projects with a variety of designs and technical practices has been part of the product. GNOME puts a lot of effort in making a clear focus in order to improve quality, reduce the UX variety  and make sure we walk more or less in the same design direction, technical practices, etc. Seems KDE is trying to improve this now too, which is nice to hear.

There are other areas that GNOME does better just because of handicaps KDE might has such as not being default in most of well known distributions or being less enterprise-ready (work we do a lot at Red Hat for GNOME in order to comply with US regulations, etc.).

I tried to give my advice and offered my help on these areas from now on, as I believe what GNOME uses would benefit KDE and in the same way would benefit GNOME by using the same tooling as our colleagues at KDE.

What KDE wants to work on

Flatpak. Yes! Looks like KDE has a wide interest on use more and more Flatpak. Some aspects I could gather from them is that the community driven nature and FOSS stack of Flatpak aligns with the values of the KDE community. Seems they also appreciate the technical thought and technology behind that was put into Flatpak. Alex Larsson is truly an amazing dev.

However, they would appreciate the Flatpak community to proactively approaching KDE to help use Flatpak more, and it’s something I would like to see more too. I proposed to do a hackfest together to collaborate improving the usage of Flatpak and collaborate with the community as a first step. So hopefully there will be one soon!

Wayland. They really want to improve the Wayland status at KDE. At GNOME we have some (1.5?) developers working on it and at KDE they have one working on it. So I proposed to meet with the GNOME developers to share advice and see how the collaboration in Wayland freedesktop for new APIs such as screenshot, screen sharing, color picking, etc. can be improved. Hopefully soon we can all meet to discuss these out.

And yes, it was pretty interesting

I  also met and chat with Albert Vaca, the creator of KDE connect. He actually uses GNOME but only attends KDE events because he loves the community, which is quite interesting to me.

KWin maintainer Roman Gilg who I chatted with about Wayland and further collaboration between GNOME and KDE. Roman also tried to save me on Saturday for the lack of parties (I hope next Akademy there is a party on Saturday. GUADEC social events were amazing this year thanks to Ismael).

Slimbook CEO Alejandro López who I chatted with about hardware, OEMs and opportunities around that. Excellent and fun as always.

Ben Cooksley who is this amazing free time sysadmin contributor that runs the whole KDE infrastructure and we chatted about CI setup and funding.

The newcomers experience leader Neofytos Kolokotronis who is passionate about improving the newcomers on-boarding.

And last but not least, I interchanged with Aleix and Lydia valuable advice on how the board, community and funding is ran in both GNOME and KDE. I also ended having a passionate argument with Aleix (as always 😜) that ended up with a hug.

We have already some email threads and discussion on these ideas on the work, going to the conference is already paying off.

Overall, nice conference, nice people, and I’m thankful to all of you at KDE!


Thanks the GNOME Foundation for sponsoring both Felipe’s and my trip.



15 thoughts on “A GNOME dev enters an Akademy and…

    1. Hey Martin, actually I don’t know if there is something written, I was just informed about the “red=no pics, blue=pics are fine” at the conference.

  1. I would love to see a lot more collaborations like these, it’s Extremely healthy for the overall future of Linux in every possible area. By being as constructive as this, learning from eachothers camps, and collaborating wherever possible (and there’s a million of possibilities just waiting for being discovered) it will be possible to supercharge the efficiency for everything from onboarding to development. Linuxers are already doing this a lot more than others, but far-far-far from enough. Let’s discover and achieve the next milestone!

      1. Definitely.
        People of these communities are just passionate people developing something cool they believe in. Starting from this, caring about other projects is really limited to whether they find cool what others do and if possible share workload and collaborate in common needs.

  2. Around a decade back i was really excited about Wayland. There was such strong universal vibe going on and it seemed like things will advance fast and the outcome will be great. Unfortunately things didn’t turn out to be like that at all. We are nowhere near the era Wayland will start to dominate. And unfortunately still not much direct advantage for the end users can be observed. Still no real problems were resolved. Compared to the problems that existed a decade back.

    As for the Flatpack. I detected much less universal enthusiasm when it was conceived. It is still a young project. There was much older alternative available called AppImage and Snappy was conceived about that time. Therefore compared to Wayland there was more fragmentation involved from the beginning. In theory it sounds nice but in practice you quickly realize things are not all that mature yet. And when trying to package an application having a bit more dependencies. And things like that. You quickly need to become an expert in packaging. Or it won’t happen. In my opinion Flatpack likely could end up being like Wayland. A decade of talk and development. But still far away from replacing the other packaging solutions it was conceived for to replace.

    Likely such projects are too ambitious and they turn out to take more effort and resolve less (and add a few new ones) existing problems than originally imagined. And the pace turns out to be simply too slow to make any real difference.

    Such projects would need to replace what they are trying to replace in 3 to 5 years. If that doesn’t happen it likely won’t happen at all.

    1. Aprecciate your insight. However I would just say (in a good way) that if we all had your opinion/mindset we wouldn’t have GNOME or any Linux desktop at all 🙂

      We need more energy and people like you pushing for it, not less!

      1. I am curious but much more cautious. When it comes to the Flatpack. Compared to the Wayland. I was pushing for Wayland roughly for about first 5 years. Did that with a big dose of enthusiasm involved. Unfortunately slowly more and more i started to realize it likely ain’t going to happen anytime soon. And now almost 5 years after that epiphany unfortunately that is still true. In the end a decade is just too long. X11 is still dominating the display area. Wayland (at least until now) didn’t resolve any oppressing issue worth mentioning.

        And as i simply can’t imagine Flatpack in its current form could replace .rpm or .deb packages in a decade. Counting from when it was conceived and based on how things stand ATM. That is why i can’t be pushing for it as aggressive as i was pushing for Wayland. At least for the first few years.

        And to be honest i don’t know if such mindset is hurting things like Linux desktop. Linux packaging …

        What i do feel is hurting it is projects that get all the spotlight and can’t deliver. I am not saying Flatpack won’t deliver. Hopefully it will do just that. But based on the Wayland experience i am more cautious.

      2. There are definitely major things already solved both by Wayland and Flatpak (security, delivery of apps cross system, reproducibility, lower barrier for development, tear-free graphics, portals for interoperatibility, etc.), and the current status is that they already work. I don’t know what else would expect to compare a technology developed over more than 20 years with one developed over 5 though.
        It’s fine being cautious, but if that means “just give up, new shifts on technologies are hurting the ecosystem” then again, we would be nowhere now or 20 yeara ago. These big changes always require years of development, expecting otherwise is unrealistic.

  3. The problem with security is i in general trust FOSS apps. Therefore i don’t need sandboxing and permission control. When trying to pack a FOSS app. All of that only gets in the way. Therefore inherently we are making this packaging solutions for non-FOSS apps. As we can’t just blindly trust them. Tear-free only works if everything is in optimal condition. And a lot of GPU drivers aren’t there yet. Therefore tear-free experience is only available on a limited scope of hardware for now.

    Don’t read this as having a negative mindset. It’s just that i expected general improvements will be measurable in less then a decade (Wayland). Based on how things started. And if solutions like Flatpack/Snappy/AppImage won’t replace solutions like .deb/.rpm packages in a decade. There is still much work left to be done in Linux packaging area in the future. Going beyond .deb/.rpm/Flatpack/Snappy/AppImage efforts …

    Therefore maybe on the next GUADEC and Akademy some discussions about that would be great. As currently the spotlight is on for example Flatpack. Claiming “The future of apps on Linux” is here. But maybe a future is something else entirely. And we are not there yet.

    Anyway thanks for all the hard work. And fingers crossed we are there already.

    1. > The problem with security is i in general trust FOSS apps. Therefore i don’t need sandboxing and permission control.

      OSS software may be more trustworthy, but that doesn’t mean it’s immune to the benefits of sandboxing.

      Imagine you have your favorite secure, open chat app. Then, a vulnerability is discovered that lets a rogue message execute arbitrary code on your system. It seems rather scary, no?

      With Flatpak, it’s not really much of an issue. Why? Because the app is sandboxed!! Arbitrary code execution is still bad, but they wouldn’t be able to access your filesystem or the full message bus.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: