Monday, May 27, 2013

Interview: 'We want to build a strong product that is stable and people can rely on'

Interview: 'We want to build a strong product that is stable and people can rely on'

Interview: 'We want to build a strong product that is stable and people can rely on'

Vincent Untz is the chairman of the OpenSUSE board, Gnome hacker and all-round nice guy. He's also part of the team at Attachmate working on SUSE's cloud product. So, in short, he's seen free software from almost every angle.

Linux Format magazine caught up with him at FOSDEM, in Brussels, where he was due to give a talk titled "Have the Gnome community turned crazy?"

LXF: How did you get started with open source software?

VU: Actually, I usually say free software, not open source. I got started around 2002, I think. At the beginning of the year, I was just curious about it. I was looking at what was happening, and that was the point when the Gnome team was working on Gnome 2, just before 2.0, so there was a lot of activity on the mailing list.

I was really looking at that, and at some point I thought 'maybe I could try to help'. So I started doing various things, like looking at the bugs, helping with the membership committee and stuff like that. Over the year, I just did more and more.

LXF: You've been very active in the Gnome community. How did you end up with OpenSUSE, which is more commonly known for KDE?

VU: Back in 2005, I was trying to help on Ubuntu actually. I contributed a few patches, and after a while I said I couldn't do both Gnome and Ubuntu. I had to pick one, and it was Gnome.

During that time, I was still doing my studies. In 2007, I finished my PhD thesis, and so I started looking for a job. Back then, it was Novell, and they had a pretty good position doing stuff for Gnome, and that was my dream job. I joined, and part of the job was to work on OpenSUSE.

Half upstream and half OpenSUSE. So that's how I joined the OpenSUSE community. That was the time we were opening the development to the community, so it was quite exciting.

Vincent Untz

LXF: Currently, you're chairman of the board of OpenSUSE. What does that involve?

VU: Meetings! A lot of meetings. The role of the board is to help organise the community and handle the financial and legal side of things. We try not to get involved in the technical discussions. We leave the technical things to the technical leadership and the release team.

All the non-technical stuff depends on the board. The chairman is appointed by SUSE, the other five members are elected by the community. I'm not really sure it's a goo d thing to be appointed by the company, but it's for historical reasons. The chairman is really there to make sure the board stays active, getting focused, and moves forward, but otherwise he's a normal board member.

LXF: Do you still have time to work on code?

VU: Actually, no. I still have a job, which is not being the chairman - that comes on top of my normal job. I work on the cloud, which is on OpenStack, but other than that I don't have much time for technical matters anymore.

I try to help the Gnome team every now and then, because I have accumulated a lot of knowledge and I'm trying to share that with the team, but they're doing quite well without me. It looks like I'm not needed for the Gnome team. It's quite good to see that we've built a community, so I can go away from the Gnome team and it goes well.

LXF: How is OpenSUSE used with SUSE. Do they use it as the base of their commercial offering?

VU: The way we see it, OpenSUSE is the upstream for SUSE Enterprise. We take OpenSUSE at one point. We stabilise it. We add certain features that are requested by our customers. We make sure that it's well tested.

What's interesting is when you compare it to other projects, like Fedora. Fedora is usually seen as a test bed for RHEL (one of the goals of Fedora is to be at the bleeding edge of software development, which is pretty good for free software as a whole in my opinion), but we don't do that for OpenSUSE. We want to build a strong product that is stable and people can rely on, so it's a bit different.

LXF: You've been involved with Gnome since before Gnome 2. Gnome 3 has been a bit controversial.

VU: Really? I didn't notice!

LXF: Were you surprised by how people reacted to it?

VU: Yes and no. When we first started, we didn't really kn ow what we were doing, but after a few months we had a plan and we knew that people would have a strong emotional reaction to that. Some people would love it, some people would hate it. We were expecting stuff like that, definitely.

What's interesting is that I was expecting much more negative feedback than we got, and we got quite a few positive comments quite early, which was really encouraging for us, and helped us keep moving. What I didn't expect is that people would keep, after like two years now, keep saying stuff like "I hate Gnome 3", "the Gnome community just hates the world", "they're evil", and stuff a bit crazy.

I mean, it's perfectly OK to not like Gnome 3; that's fine, it's an opinion, but you can see all the emotion inside such comments, and I didn't expect people to react like that in the long-term.

LXF: Your talk is entitled 'Have the Gnome community gone crazy'. In a word…

VU: Well, if crazy is meant as a negative word, then no. If it's positive, then yes. It depends how you want to understand it, but the talk is really about explaining the point of view of the Gnome community. We've been extremely bad at communicating our decisions. And also debunking some of the myths that keep going around, and keep spreading despite being obviously wrong.

LXF: Would you like to point some out now?

VU: I love the one that the Gnome community doesn't accept feedback at all, and that we don't listen to users. When we built Gnome, we built something that we believe in; and there are a lot of strong decisions, strong choices that we put in there. Sometimes we feel that we've got everything right, and then we realise that it's wrong. And so sometimes we do step back and realise we have to reverse.

There are some concrete examples, like the shutdown menu item which we didn't have in the first few releases of Gnome 3, and that we released back in Gnome 3.6. We have the notification area, which was an improvement, but which was still not really usable in the beginning - people had the issue that they didn't really see when there was a new notification - so we worked a lot on usability there.

LXF: The one we keep hearing a lot is that Gnome 3 is designed for touchscreens.

VU: Yes. That's a common one. People say that we make something that's for tablets. No, that's not our goal. Our target is really the traditional computers, but also the new computers that we are seeing that are touch enabled, like the new devices for Windows 8. We know that building something for tablets, or even phones, is completely different. We would like to do that at some point, probably, but we're not ready for that.

Right now, we're still about traditional computers. We're really about traditional computers, even though we try to make sure Gnome i s touch-aware, so you can use it if you have a touchscreen.

Vincent Untz

LXF: Just getting back to your work at SUSE, where you work in the cloud. Is the cloud seen as the future of SUSE?

VU: Obviously, I don't know the whole strategy for SUSE, and I cannot comment officially on that, but it's clear that the cloud is the direction of the whole industry right now. If you look at OpenStack, which is what we adopt for our product on the cloud, it's supported by nearly all the industry leaders. The OpenStack Foundation is a huge success, and has a lot of sponsors; SUSE is a platinum member of the Foundation, and it's simply incredible.

I cannot analyse that number, but the OpenStack Foundation has a budget, which is around ten million dollars, and the Foundation is le ss than one year old. The project is around two years old. So it's just completely insane.

When you look at the way OpenStack is moving, it's going so fast. The contributions going in are just huge in terms of number. Right now, we are catching up with what Amazon is doing with the web services, but quite soon we will become the top innovating. I think cloud is an important part of anybody innovating in the industry right now, including SUSE I guess.

    


No comments:

Post a Comment

//PART 2