We interviewed Frederic Peters, the new Gnome Release Manager, to understand the future plans for Gnome 3.x, his opinion about Unity and Cinnamon and much more.
Swapnil: Please tell us more about yourself. How and when you came in contact with computers?
Peters: I am 32, living in Brussels, Belgium, working as a software developer in a small cooperative company (entr'ouvert); in my free time, apart my work in GNOME and Debian, I am also quite involved in a local radio (radio panik) and an association offering hosting to various persons and collectives (domainepublic.net). I encountered my first computer in the late 80's, in the small period where computer classes were about programming, and not about using Word; I got hooked and got my own computer a few years later.
Swapnil: How and when did you come in contact with Free Software? What attracted you towards Free Software?
Peters: I got a modem and connected to various BBS in my local area, one of them was running on GNU/Linux (I can't remember its name but it was managed by Pablo Saratxaga, who did quite some work on locales) and I got curious. Of course there were technical reasons, new things to discover, but it also arrived at a time I was a teenager thinking and reading quite a lot about society and politics, full of ideals, and Free Software just felt like the right thing to do. 15 years later I am quite happy to still hold those ideas.
Swapnil: Which distribution do you use and why?
Peters: Debian. As I had my first steps in the free software universe, I understood it as the closest to the ideals, a universal operating system, collectively built by dozens of persons, and growing. After less than a year of using it I felt I got so much from it that I went to contribute my part, and became a Debian Developer, that was in the month of June, in '98.
Swapnil: You recently took over the role of Gnome Release Manager at the time when Gnome is going through a major transition. What challenges are there for the Gnome team?
Peters: Well, the transition people talk about is mostly technical, from metacity/gnome-panel to gnome-shell, and while it was quite something, I think we are handling it fine, and that's because at it heart, we didn't change, the people of GNOME are still the same, and they still aspire to the same thing; building great, and free, softwares for our users.
But still, talking about transitions, a major challenge is not to get insulated developing for standalone computers as people are more and more using online services (and too often losing sight of the freedoms they deserve). In this aspect I have to applaud the work initiated by our KDE friends with the ownCloud project, not only it is good software, on par with closed ones, but it gives back control to the users, while giving the possibility for companies to offer hosting of those services. I certainly hope we will get to extend gnome-online-accounts to such services, being limited to Google and Windows Live should only be a momentary sadness.
Swapnil: What is the audience Gnome is targeting with 3.x series? How satisfied are you with 3.x?
Peters: I'm still a bit jealous of the Ubuntu tag-line, targeting human beings, what else? I am quite happy with 3.x, sure some responses were rough but many were really positive. Whenever we get to manage booths, after years of people passing before us and nodding, we are now back at getting passionate users, inspired by joyful passions.
Also I am happy about the way it kept on improving, 3.2 was much better, and 3.4 even more so. Also, while continuing the work on the core shell, we are now getting applications redesigned around the same principles. Ideally we will get updated Human Interface Guidelines published soon, to communicate those principles in a clear and effective manner.
Swapnil: Gnome 3 has received mixed responses, mainly about lack of customization. A lot or features have been taken away and users are looking for alternatives? What is your general opinion about it beyond the fact that people like status quo?
Swapnil: Gnome's multi-monitor set-up seems to be very bad. On the contrary KDE offers a lot of customization for a user of multi-monitor set-up. Unity is also developing well and listening to user's feedback. What plans are there for multi-monitor set-up?
Peters: I am not used to multi-monitor, the only time I use it is when plugging beamer, and I really appreciate the way it works, with the static secondary display. I'd have to test it but I am pretty sure it's possible to drag the top bar from one screen to the other in the display preferences panel. As for the application menu, it's still quite new and not used much, I am sure things will get ironed in time. As I have my desktop configured with sloppy focus, this is something I am quite interested in.
Swapnil:. Gnome does have road-maps for future releases, yet how much does change based on customer feedback or complaints?
Peters: I don't like to think of our users as customers. We listen to them, that doesn't always mean doing it their way, sometimes the proposals are just opposite of what we want to achieve, other times we get valid usecases but the proposed solutions are lacking in some ways, we then have to think about them with regards to the big picture, existing and planned code; this takes time.
Oh, and we have some opinionated developers and designers, of course.
Swapnil:. Ubuntu burnt its fingers with Global Menu. They are now giving the option to disable Global Menu. Despite the backlash that Unity faced, Gnome introduced Application Menu with 3.4 (without any option to disable it). Is Gnome team keeping an eye on Unity's backlash and learn from it or continuing on its own road-map? Will Application Menus become default or will there be options to disable them and use traditional menus instead. What purpose does Application Menu solve?
Peters: Ubuntu global menu and the application menu are not the same thing; the global menu is about deporting the existing menu, designed to be part of the window, to the top panel; on the other hand the application menu is all new, to be used by applications that specifically design for it, they are not created to hold the enormous menubars we have in some applications.
Swapnil: The default theme and font of Gnome 3 also meets with criticism. Adwaita has a very thick border (which makes it bad for netbooks or small screen devices), are you planning to change the theme to something different? There is also a criticism of default fonts used in Gnome 3. Are you planning to change the fonts to something more aesthetic such a Droid?
Peters: I am typing this on a netbook and I can certainly understand some complaints, but it got better in 3.2, and 3.4, and will continue to improve. There is no drastic theme change planned but Adwaita will continue to evolve, but it requires a lot of attention to details, testing it with many applications, tailoring it in places, and quite often the designers have to get back to GTK+ hackers to add new features.
Swapnil: Gnome 3's power is extensions. However, there is no mechanism to install, enable or remove extensions. A user has to use a browser to install extensions. Tweak Tool does allow to enable/disable extensions but you can't install/remove. In addition, Tweak Tool is not part of Gnome, some distributions don't pre-install it. Is there any plan to provide a tool where users can easily manage (install, remove, enable) extensions, themes and other such things? If there are plans when will it be implemented?
Peters: It was decided the primary mean of handling extension would be the extension.gnome.org website; but still I was pretty sure the GNOME Tweak Tool allowed to install extensions, and if it doesn't yet, it will probably happen.
Swapnil: From where is the majority of Gnome contribution coming from, can you give some data?
Peters: It's really difficult, two years ago Dave Neary did some work centered on module owners, but this is clearly not enough; I'd love to have enough time to collect data about contributors, demographics, employment, and study whatever this gives. Even better if this could be done for previous years, it would be really nice to get past anecdotical evidences and point out that a program like GWOP really did a good job.
Swapnil: What is your opinion about Unity?
Peters: To be honest, it's not packaged for Debian, so I didn't get to use it for real. Of course there were tensions around Unity and the relation to GNOME Shell but I think this is over, and nowadays people are happy acknowledging they are just two different projects, with their own goals.
Swapnil: How does Gnome project look at Cinnamon?
Peters: Even more so than Unity I didn't get to use Cinnamon. From my distant position I thought the previous approach, installing GNOME Shell as well as a serie of extensions, was nice and probably easier to maintain in the long run, so I hope the developers tried to discuss extension points with the shell team before forking.
As for the fork itself, and it's a thing they have a common with MATE, it's a real loss that they didn't import the git repositories in the proper manner, as it makes exchanging patches much harder.
Swapnil: Do you have a "standard set" of reference hardware, or do you just test on a variety of machines?
Peters: Not really, developers use their computers, and that's about it, hopefully this brings enough diversity, and then there are more structured tests done by distributions. Also we get hardware support though lower layers of the stack, so we get video cards, and bluetooth mices working thanks to the hard work of other developers.
Swapnil: I would love to hear the plans and roadmap for Exchange support in Evolution. I have seen the Evolution Exchange source code on the Evolution page but don't have an idea as to the stability and when it will be packaged up (rpm/deb) for broad use. Exchange support has long been the one thing holding me back from using Linux 100% at my work place(s).
Peters: I don't know much, you'd have to ask Evolution developers for more details, I know there are plans to integrate Exchange support with GNOME Online Accounts.
Swapnil: Just want to say that I love the changes done in Gnome 3.x. With so many linux distributions out there now, almost all of them implement their own changes to the UI. I personally love a clean Gnome Shell without any changes (which generally slow things down). Will we ever see a distro coming directly from the Gnome team? I’ve tried building via jhbuild, but have never been successful (hint). Would be great to see a gnome distro with its own package management.
Peters: I don't think GNOME 3 gets customised by distributions as much as GNOME 2 was; thinking about major distributions like Fedora, OpenSUSE, Debian, (and Ubuntu even), they don't divert much. Still there are efforts underway to make testing the development versions of GNOME much easier, some of them in jhbuild, others in a new project called 'ostree' (developed by Colin Walters), and collectively we try to fix build failures as fast as possible.