There is something very corporate about Canonical’s London offices, looking down on the Houses of Parliament. It’s a long way from the grass roots up to the 27th floor of Millbank Tower. The relationship between the corporate backers of Ubuntu Linux and the open source communities with which they interact has rarely been smooth. And yet Ubuntu remains one of, if not the, most popular desktop Linux distributions around.
That said, Canonical’s CEO, Jane Silber, seems sure that primacy of the desktop in computing is over. Canonical is looking to push Ubuntu out onto converged internet TVs, into smartphones and cars, and vying to become a player in the highly corporate world of ‘continuous computing’…
It’s almost exactly two years since Jane Silber took over the role of CEO from Canonical’s founder Mark Shuttleworth. Silber joined the company shortly after it was founded in 2004. Even as its chief operating officer (COO), she attracted the headline ‘Jane Silber runs the company while Mark Shuttleworth gallivants’. So we start by asking her if she feels this is still a fair characterisation of their respective roles?
“I know exactly where that came from. That particular headline came about prior to me becoming CEO, where it was really becoming clear what my role was and how much I was involved. In terms of our roles now, Mark is certainly not off gallivanting around. He is still heavily involved in the company. It’s his full-time day-to-day job as our product strategist.”
While Shuttleworth also has his eponymous charitable foundation, Silber maintains that his calendar is still filled by Canonical. She laughs at the notion that Mark is now just a figurehead, with her getting all the work done.
“Mark, in an interesting way, reports to me on product strategy. Ultimately, I am also responsible for that. I’m responsible for everything that Canonical does. I’m responsible for ensuring our success at a financial level, on a commercial level, at a strategic and partnership level. We have a really smart team; I obviously don’t do it all alone, but ultimately I am the CEO.”
We put it to Silber that if you look at the other Linux companies with large in-house development teams, they seem to split the distros into community and enterprise editions. Red Hat has Fedora, SUSE has openSUSE, but Canonical doesn’t do this. What is the reasoning?
“That is a very explicit vision that we took early on. It is a key part of our strategy, to not do that. The reasoning is that we thought our business model shouldn’t be around bits of the software but around the services that we provide. And so we made an early commitment that our best work would always be free and that there wouldn’t be a less-good version for community and a better version for enterprise users or paying users. Rather than selling security updates or more features in the software, we sell our services.”
Quite whether Red Hat would be comfortable being indirectly positioned as selling security updates, or even open core, is questionable. However, we continued to press Silber on the challenges Canonical itself faces with a single distribution. We ask Silber whether she finds that this model can cause problems, especially within Ubuntu’s community relations…
Canonical often gets a reputation of being very arrogant, because it has got the professional team and the interests of the company to put into the distro at the same time as it has the community sometimes wanting to do very different things.
“Yes, it’s true. There are challenges when we sometimes want to do things for commercial reasons and we can’t reveal it as early as the community would like. There are challenges around the fact that most enterprises would actually like some non-free software in there. We have made an exception for hardware enablement and drivers, but we have made a commitment that the applications must be free.
“There is constant tension and trade-offs that we make, but I think that that is a healthy tension. It keeps us paying attention to the entire set of stakeholders and interested parties in Ubuntu and not just one thread. But it does create certain challenges.”
Perhaps Red Hat and SUSE are doing the ‘community thing’ more through the Linux community itself, while their commercial distro is more separated off, giving them more freedom to do exactly what they want?
“Right,” she concurs. “We’re trying to do it in a much more cohesive, single collaborative effort. With mutual respect and understanding for someone who is much more interested in the community effort and the very important commercial drivers.”
Moving the conversation along to talk about some more specifics, we ask Silber for her take on the whole issue of Unity, Shell and GNOME 3. Last year, Canonical took the highly controversial decision to drop the latest full implementation of GNOME from Ubuntu. Instead, the company has moved ahead with its own Unity user interface, which, while still being built on the core of GNOME, discards the GNOME Shell user experience…
Naturally, Canonical’s decision to rip one of the largest single subsets of the GNOME user base away from the Foundation was not entirely well received. So how are current relations with GNOME?
“I think it’s a pretty complex dynamic, and what you see on email lists and on message boards is a sign of the passion, the level of belief and the level of caring that people have. But I also think its dangerous to read too much into a single thread or a single viewpoint there.
“If you take a step back, several years ago, we started looking at the state of open source. Particularly desktop open source, the state of Ubuntu, and of the competition in terms of the proprietary operating systems and offerings. We took a very long-term strategic view of all the factors in the ecosystem, and we decided that one of the really important things for open source to be successful in general, and for Ubuntu to be successful in particular, was that we had to raise the level of design and user experience.
“We called it then as the key strategic threat. Mark Shuttleworth covered it in keynotes, including OSCON. He planted the flag and said, we collectively need to address this, and it’s a thing that Canonical is going to invest in. At that time we didn’t have Unity in mind, but our vision was that, as an ecosystem, the open source world needed to raise its game. And that we would help lead that, and help rally people around that cause.
“Unity evolved out of that, but it wasn’t an intention. We didn’t have it formulated in our mind when we said ‘This is the next frontier’. We knew that design was the next frontier, but we didn’t know how we were going to navigate our way through that. Over time, our vision for Unity became clear. Some of that was done publicly and some of that was not. The fact that we are now committed to Unity and GNOME is committed to another path, this has been difficult for many people, but I think it’s not a disaster. One of the things that people in open source have always said is that choice is good. Competition is good and people will choose what is the best for them.
“The fact that GNOME and other projects now value design,” Silber stops, perhaps to reconsider the boldness of what she is about to say. “If you go back three years ago nobody was talking about design, nobody was doing user research. It is actually something we have had great influence on, by calling attention to it and putting our efforts there. I think, whether you like Unity or not, its existence has helped raise the bar across a number of projects. That is something that we feel good about; you can attribute that to our leadership in that area, even if it’s not our code and our design.”
Canonical has recently announced its plans for expansion into some major new areas. The firm announced Ubuntu TV at the CES in Las Vegas at the beginning of the year, and Ubuntu for Android was launched at the Mobile World Congress. We finish up the interview by asking Silber to expand on the bigger, strategic picture and where the company is headed in 2012 and beyond.
“Ubuntu essentially is a platform, and as the industry moves to more of a multi-device, continuous computing, mobile world, we think that the way people interact with their computers and their digital lives in the future is on much more of a continual spectrum of form factors and devices. In the future, the desktop will no longer be the sole or even the primary interaction point.
“Ubuntu is particularly well suited to support that spectrum and it is an explicit strategy of ours to do that.”
Intrigued and slightly confused by the concept of ‘Ubuntu for Android’, we ask if this in not two operating systems on top of each other…
“It is two operating systems next to each other. When you have an Android phone and you are interacting with another Android phone, it’s stock Android, nothing changed about it at all. If you dock it with a keyboard and monitor, it recognises it’s in a different form factor and Ubuntu will take over, providing a full Ubuntu desktop.
“Importantly, [the Ubuntu instance] still has good integration points with the Android side: common contacts, common Wi-Fi configuration, calendaring, even to the point of telephony functionality. So, if you receive a text message while you are docked, it shows up on your Ubuntu screen and you can reply to it. If a call comes in, you can answer it in desktop mode, you don’t have to switch back to phone mode.
“If you are browsing on your phone and have tabs open, and then dock your phone, the desktop browser will have those same tabs open. It delivers this continual experience.”
Silber is clearly passionate about the technology. She sees this notion that the devices converge, and flow in and out of one another is one of the coming trends of 2012.
“We see unification at the OS level already from some of the big proprietary players. It’s what Apple is doing, its what Microsoft is doing and it’s part of the Ubuntu vision as well.”
We put it to Silber that despite her prediction of demotion of the desktop, even in the word of ‘continuous computing’, office applications still play a big role. They are vital for cracking the corporate part of this interconnected market. Microsoft has an obvious advantage here. Apple might be looking increasingly uncertain, but Google has potentially cracked the problem with the growing ubiquity of a Google Docs login among corporate users. What is Ubuntu’s strategy here? Silber rightly argues that Google Docs can run on any platform and that it is not exclusive to Chrome OS or Android. However, the point at which a user logs into Google Docs on Ubuntu it becomes logical for them to use all of Google’s cloud apps, as they all tie in together…
So why would we use Ubuntu One for any of our calendaring, mail, contacts or anything else that we want to have unified across all our devices? Does Ubuntu have a master plan in there?
“Clearly it’s competition, we need to compete. One of the advantages we have, from an industry ecosystem perspective, is that there is great interest from OEMs and handset manufacturers in being able to provide, frankly, non-Google services. And I think we have yet to see how it plays out.
“It’s certainly a battlefield, but one of the reasons that Ubuntu is attractive is that we are not selling handsets, we are not making hardware. We are a good partner in many ways and continue to play to that.”
Silber refuses to be drawn on specific partners for either Ubuntu’s TV or Android solution. “We are in conversations with a number of people. Our go-to-market strategy is through those OEMs, handset manufacturers or mobile network operators. So it’s really their product roadmap and their product to announce rather than ours.”
We ask if Ubuntu will still be recognisable within these OEM incarnations and if they will still interact with a standard Ubuntu desktop and Ubuntu’s own-brand cloud services.
“I need to be careful about not committing another company to something,” Silber responds cautiously. “But it is our vision and our intention that it is an identifiably Ubuntu experience, solution and platform, and that there is value in that.
“There are some key characteristics of a good platform in this area. Obviously you have to have good, credible, high-quality, feature-full products. For us, one of the key selling points is that it is an open platform, which allows for collaborative development with community and with companies.
“Having a unified experience across form factors is important. That doesn’t mean identical. Unity runs across desktop and TV for example, but it shows a different face. Obviously the 10ft lean-back experience is very different from interacting with a laptop or desktop.
“That unified platform and that unified experience is important in terms of the platform strategy and it also lends itself to a good developer ecosystem. The platform is necessary, but you also need a strong developer ecosystem, which means people developing applications but also a store to expose
“This is the rationale for the work we have been doing recently around the Software Centre and ‘developer.ubuntu.com’, which have received a lot of attention in the last year or so. We are trying to make it much smoother for ISVs and independent developers to deliver their applications to Ubuntu users.”
You can read Jane Silber’s full six-page interview in issue 112 of Linux User & Developer.