He is also, without a doubt, the open source leader who is most actively shaping the future of the Linux desktop. Shuttleworth’s company, Canonical, sponsors Ubuntu, which has done more than any other distribution to begin pushing the Linux desktop into the mainstream. In a controversial move, Ubuntu ships with a Restricted Drivers Manager, facilitating the use of proprietary technology. While this offends some purists, this decision – along with Ubuntu’s famously user-friendly interface – could enable adoption by a non-specialist mass audience.
Equally revolutionary was Canonical’s deal with Dell to sell desktops with Ubuntu pre-installed. Convincing a major hardware vendor to take a chance on Linux PCs (internationally) sent a shot heard across the software business.
In this wide-ranging interview, Shuttleworth discusses the Ubuntu initiatives for mobile and mini-PCs, and his goal of making the Ubuntu user interface on par with the Mac. He also addresses the Novell-Microsoft deal, the controversy about proprietary drivers in Ubuntu, the Debian-Ubuntu relationship, and more.
In conversation, Shuttleworth speaks with an elegant English accent, likely due to his South African upbringing. He now lives in London, which is where he called from:
What about Ubuntu Mobile? What’s on the way in this area?
By way of background, we see tremendous opportunity for Linux in the consumer electronics and mobile space. There’s room for great diversity of offerings, so we imagine in due course you’ll have Linux-based products from very large organizations like Nokia and Motorola, who will have their in-house proprietary stacks built on top of Linux, much of which may well be open source, through to highly customized platforms for very specific devices like navigation and so on.
But we do think there’s room for an open, royalty-free, standardized platform like Ubuntu. In the same way that we deliver the desktop, we think that there’s value in bringing the same values to the mobile space. So that’s what Ubuntu Mobile is all about. We’ve been working largely with Intel, which has an initiative called Moblin. It’s an initiative from Intel to develop a series of frameworks and applications for touch-oriented consumer electronics devices. They are the primary [promoter] of it, and we’re integrating that into a distribution in the same way that we integrate GNOME or KDE into our desktop-based distributions. So it’s exciting work.
The version that shipped recently is kind of a 1.0 for their first cut of Moblin. And we expect that it will have a 2.0 sometime in 2009, and we will evolve the Ubuntu mobile offering in close coordination with them.
And what about the recent news regarding Ubuntu designed for small, ultra portable laptops?
Yes, that’s what we call our Netbook Remix. That’s really the desktop set of applications, but with a different launcher and application switcher. So you’re optimizing for content creation and you’re making some tweaks for a slightly smaller screen; content creation because you have a keyboard.
You have a smaller screen, you’re trying to use the screen real estate more efficiently, more compactly, than you would in a normal desktop. But you’re not radically changing the application, you’re just compressing, slightly, the user experience. And then in the way people select and fire up the applications, you’re optimizing for the case where people have a [ultra-portable] laptop where they’re only running 4 or 5 applications regularly. So it’s a slightly different emphasis but it’s still a desktop.
Can you say any of the companies you’ll be working with on this initiative?
I can say we’ve concluded agreements with several of the global brands but those will be announced in due course.
This question comes from a member of Free Culture at Virginia Tech: Ubuntu is certainly a major step in the right direction in terms of universal usability of Linux for the average person’s desktop/notebook. But the process isn’t complete yet. Are there any fundamental changes being considered to accelerate the release of a perfectly intuitive, out-of-the-box Ubuntu that will single-handedly dominate the world in the name of Free Software?
I think his premise is correct, that Linux has made leaps and bounds. And it’s not just Ubuntu – and Ubuntu certainly can’t take credit for all the progress that’s been made on the usability front. But we certainly champion the idea.
We genuinely believe that Linux can deliver what he’s asking for, and we want to be right in the front of the effort to help make that happen. But there are lots of other groups, upstream communities like GNOME and KDE, and other distributions, and all of us are playing our part.
I do think this [time period] is a unique opportunity for Linux to step up and appeal to, and deliver value for, the ordinary desktop user, or the non-specialist user. I think these Netbooks that we were talking about earlier are a very significant factor in making that possible.
We face an opportunity now, and partly that’s because Vista has not delivered to people’s expectations. Partly that’s because the Web is increasingly how people define the PC experience. It used to be, the PC was what you used to run Microsoft PowerPoint and today it’s what you use to surf the Web. And we can deliver a fantastic Web experience on Linux; I would argue a better Web experience than you get on Windows from a safety-security perspective. So for all of those reasons this is a really important time in the history of Linux.
Canonical, the company behind Ubuntu, has historically been very, very deferential to what we call our upstream communities – GNOME, KDE, and so on – in the definition of the desktop experience. Our view, very strongly, is that they hold the real expertise in defining that. And that, as a distribution, our primary job is to be a very efficient conductor of their good work into the hands of users. Which sometimes sounds like mooching, but in fact it’s a very significant and serious responsibility and implies a tremendous amount of work when it’s done well. And we sort of specialized in doing that, playing that role: we’re a conductor and our highest goal is to not get in the way.
Because we’ve increasingly been engaged in the definition of the desktop experience for some of these consumer electronics products, however, we’re now in a position to actually start engaging with those upstreams and investing in that desktop experience.
And so we started to build out a team that will focus on the specific user experiences that your reader is probably referring to, and our goal, very simply, is to make sure the Free software ecosystem can deliver a Mac OS-like experience, or an experience that will compete with the Mac OS.
We see Apple as the gold standard of the user experience. We believe that, while it can be a challenge, the innovation inherent in the Free software process can deliver an experience that is comparable and in many ways superior.
So Canonical will in fact launch an effort to try and spearhead that. And over a period of 2 years, really move the dial forward on the desktop experience.
Interesting that you would reference Apple. Because several weeks back I wrote an article – I’ve been an Apple user most of my life – called “An Apple User Tries Ubuntu.” A lot of readers were upset when I said that, because I’ve been a longtime Apple user I’m going to stick with Apple, but I saw the virtues of Ubuntu.
And I don’t in any way react negatively when you put it the way you’ve put it – ‘Hey, I’m willing to try this Ubuntu thing, but you know what, I think Apple still delivers a better experience.’ I bet from the experience of a content creator that is a reasonable position. If you were a developer, I would take a different view. If you were particularly interested in the things that Linux does better, than I would say, ‘Hey, have you thought about these things.’
But right now I see it is true – that for an average user at home, a Mac would give a better experience.
We can change that. And while Canonical has historically taken something of a narrow view of its role in the ecosystem, I very much now want us to help to be the lightening conductor, the lightening rod, for that effort.
Will Linux will ever become a major player on the desktop, and if so, when? How many years away are we?
Shuttleworth in space
I do believe that Linux will become a significant force, and I think that Ubuntu will continue to champion our goals of delivering the best combination of both software freedom and ease of use. And so, putting those together, I do think, yes, we can become a force to be reckoned with. And offer a lot of value to the man on the street. We articulate what we’re about: Our passion is about Linux for human beings, it’s not Linux for Linux specialists, or Linux for anything other than the people who we care about.
How is the deal going with Dell? Have the numbers worked out as you hoped? Do you have future plans for the Dell-Canonical alliance?
Sure. The best answer I can give there is that we continue to expand the areas in which we work, some of that is visible. Dell has made Ubuntu available in markets well beyond the original coverage area and, in fact, if I look at the global picture, across all the countries in which they ship Ubuntu, there’s a very healthy volume in that.
And yes, I do think there are additional steps to be taken but this is not the right forum to [report them]. But there are concrete agreements in place and projects underway.
Is Ubuntu close to turning a profit?
The way I think of it, parts of it are. We clearly have no intention of making a profit on the operating system itself – we give that away. And so we build around that core platform a series of service businesses. And we have a number of those; some of them are closer to turning a profit than others.
And yes, I believe very strongly that we have to strive to deliver both the very best of Free software, and to do so on a commercially sustainable basis, and I think we can achieve that.
I think it’s positive in that it suggests that Microsoft is increasingly conscious of the need to engage with Linux. I think the terms under which they concluded that specific deal are negative for Linux as a whole, in that they tried to lock down, to entrench, a certain view of the economics of Free software. If you look at the deal, it very much assumes that software is being sold, so it tries to impose the economics of the ‘80s on the 21st century, and I don’t think that’s going to fly.
It’s a little bit like DRM, which tries to impose the economics of vinyl on a digital music industry. I think a lot of people are now saying, ‘Gosh, it’s not actually the music industry that’s suffering, it’s the record industry that’s suffering’ – the music industry is thriving.
And similarly, I think, any attempt to slow down the pace of innovation in the economics, as much as in the technology, is doomed to fail. And at heart I think that’s what’s wrong with the deal that was struck there.
It was very interesting to see, after the announcement of the deal, how much disagreement there was between the parties as to what the deal actually meant. And I think that’s a clear sign that it was something that was hastily concluded.
Anyhow, we don’t begrudge anybody their partnerships, we’re just very clear about the specific values that we hold dear. And we would not conclude a deal with Microsoft on the same terms – we don’t think that would be constructive for our users or for Free software as a whole.
Do you foresee a Canonical-Microsoft partnership? You’ve expressed interested in working with Microsoft.
I don’t believe it’s constructive to say that there are no circumstances under which you would ever work with a company. I have colleagues at Microsoft with whom I, and other people at Canonical, have very constructive conversations.
And there are areas where we have perfectly aligned interests. An example – a real example which has come to pass – is telecommunications policy. We’re both interested in there being a vibrant, well-regulated Internet market in every country in the world. Because that underpins the economics of the future of software. And so I’ve sat on panels with representatives of Microsoft, advising government folks on telecommunications policy, and we were absolutely saying the same things. We want deregulated, efficient markets where competition is firmly protected, rather than incumbents being firmly protected, which is usually what you see. And I have no problem with Microsoft on matters where we have a strong alignment.
Any company changes over time as its people move on. We’ve just seen a significant changing of the guard at Microsoft. There will be more of those. And so to say that, somehow the company is persona non grata and that we would never deal with them, would just be strange.
Including proprietary drivers in a distro has been a controversy within open source. I know Ubuntu includes a “Restricted Drivers Manager” to help you install proprietary drivers. Where do you personally stand on the issue of including proprietary drivers in a Linux distribution?
Proprietary drivers are a horrible kludge, they’re a little bit like introducing a cast iron pot into a titanium machine; you have something that is inherently brittle and therefore reduces the value of the whole.
To us, Linux is titanium. It’s malleable and lightweight and elegant and tough. And then you go and introduce this clearly cast iron component that’s a significant setback. And what’s most restraining about it is that it ends up hurting the people who think they’re trying to get some sort of advantage by building that cast iron piece, the hardware vendors. So, in our engagements with hardware vendors, we’re very, very clear that proprietary drivers are a very ineffective way for them to try and get the benefits of access to the Linux market.
Having said that, it is possible to do it in a way that is legal. We don’t think there’s anything that violates the GPL in what we do. We certainly wouldn’t do that.
There are ways to build proprietary drivers that do violate the GPL, but there are also ways to build them that don’t. You carefully separate the various pieces of the system. Even if you do that, even if you’re within the letter of the law, it’s still a cast iron piece. And so we still say to people: this is a lump of stuff that’s fragile, and [when] we do a security update it’s going to break. When we need to move quickly, your piece, your hardware, or your virtualization platform, or your whatever, is the one that’s going to suffer – and your users are the ones who are going to suffer, and they won’t be thrilled at you for that.
So the way we feel about propriety drivers is very, very clear: they’re sub-optimal and somewhat harmful.
The flipside to that is, we feel very strongly that I’d like to be able to give Free software to my grandmother. And I’d like her to be able to put that CD in, and have it come up on her computer. And I’d like her to experience all the power of Free software and thereby make a stronger economic case to the vendors to take Linux seriously, and to actually gives us ‘titanium’ components rather than ‘cast iron’ components. So that’s kind of the basis upon which we’ve made this distinction.
It was controversial that Ubuntu was willing to ship drivers that included proprietary blobs. I dislike it, but I think it was the right decision.
As far as the “Restricted Drivers Manager” is concerned, when we did an analysis, we found that most users of most distributions – even those professed to be very sort of purists about Free software – most users ended up with [proprietary] software blobs on their system, because they would just follow what ever instructions they found through Google to install to make their hardware work.
For example, you run into a situation where you want to make something work and it pops up and says, ‘Do you want this’ and you say ‘Yes,’ and you end up with proprietary stuff on your system– but you don’t know it.
So we built the Restricted Drivers Manager specifically, not to help people install proprietary drivers, but to tell people what proprietary stuff was on their system. And now that’s been embraced by all the other major distributions.
I think all the other major distributions are effectively following Ubuntu’s lead when it comes to hardware management. It’s a core competence we’re set to really pioneer. So that was a really long, rambling answer.
I think your grandmother analogy was the operative analogy there.
Yeah, you can use that analogy to cover any number of sins, though. A lot of people would like us to have Adobe Flash built into Ubuntu by default. It would mean that when they hit a Web site that has Flash content, it works the first time, immediately, rather than saying ‘Do you want to install Flash?’
But we don’t do that. There are lots of areas where we specifically exclude proprietary software from the default user experience. We allow people to install that easily, but we don’t make that decision for them.
The hardware one, because it is so low level and hard to get right, we have compromised on.
Some developers say that Ubuntu has weakened Debian by enticing developers away from it, moving away from an emphasis on community to an emphasis on commercialism – how do you feel about this?
I disagree with that so very strongly. I’m very passionate about Debian – I’m a longstanding Debian developer myself; twelve odd years in the running, so I can speak with some confidence on this.
If you just look at the raw numbers, Debian has gone from strength to strength. It has more developers now that it ever has had in the past. Many of the Debian developers who joined in the last four year have come to Debian because of Ubuntu. The whole Debian ecosystem has really been dramatically changed – turned around – by Ubuntu.
And we can’t take all the credit for that. I mean, we chose Debian, we work with Debian because I and everyone else on the team here believe that it is, at heart, the right way to go about organizing the very best talent to produce a well-structured, well-organized, core of a platform.
But, we also feel very strongly that having Ubuntu adds a tremendous amount to that. And that the complementary relationship between Debian, as a core upstream community, and Ubuntu as an organization which is genuinely focused on actually delivering on a very predictable, high quality schedule – a release every six months – the combination of those two is unstoppable.
I think it’s very divisive. There are a number of folks in the Debian community who deeply resent Ubuntu. And they’re very vocal. But I think they may not be conscious of the damage they do in so strongly articulating their view.
And I, again, I have no hesitancy about the role we play. I think we’re a real force for good within the Debian ecosystem. I’ll be at DebConf in August, and have been for every year for the last five years.
In conclusion, so how was your trip into outer space? Did it inspire your approach to Ubuntu in any way?
Perhaps, yeah – I hadn’t thought about it those terms. But it’s certainly true when you have the great, great privilege of seeing the world as it really is, from that perspective, you want everything you do to have an impact that transcends boundaries. There may well be a connection there.
It was obviously a tremendous privilege, and I hope to fly again – but I don’t want to hog up the seats.