On this episode, we talk with Mark Shuttleworth, Founder of Canonical and the driving force behind Ubuntu. Mark covers why code isn’t the only contribution to open source projects that matters, the importance of staying focused on you project’s “main thing” and Ubuntu’s future with different categories of devices. In addition, Mark touches upon what he admires of OS X and Windows, his thoughts on wearables and virtual reality, the most entertaining mispronunciation of Ubuntu and much more.
Alexis: So thank you, Mark, for joining us on the show and carving time out of your schedule to be here.
Mark: Not at all, it’s a pleasure to join you, and I’m excited to see what you can do with Binpress.
Alexis: As are we. So before we get to Canonical and Ubuntu, tell us a bit about your background.
Mark: I am an open-source user and supporter and funder. I studied technology many, many years ago and was really interested in things that were only possible to do with open-source software, which is how I kind of got into it. Then I became interested in broadening the appeal of open source with the hope that we really could make it the standard way people innovate and share their innovations.
Alexis: So I guess going a little bit back further in time, when did programming come into the picture?
Mark: For me, I was hacking around with Python and some low-level languages 20 years ago. Most of my focus now is on thinking about how people use the systems we create. So I do a lot of work, both on the Cloud and thinking about Cloud developers, dev ops, and on the client, thinking about convergence of all the different kinds of personal computing. Then there’s a certain amount of thinking about how people manage systems if you’ve got millions and millions of systems out there, which we do in Ubuntu. It becomes interesting to think about making it really easy for people to manage those whether they are sort of invisible servers or servers that are sort of mission critical for large institutions or whether they’re personal systems.
Alexis: Nowadays, when you look into a server, you’d probably have an 80% chance that somebody’s server is running Ubuntu.
Mark: In some environments, yes, I think that’s true, particularly, in cloud environments where people are creating and destroying them very quickly. We did a lot of work five, six years ago in Ubuntu to make it really easy to use Ubuntu in that kind of very dynamic environment. A lot of my work now is kind of the next step from there to say, “Okay. How can we make it really magical for people who can consume resources at a large scale to do so.”
Alexis: All right. So again, continuing with this kind of progression kind of thing, when did you first dip your toes in open source?
Mark: When I was a student, I was very interested in the workings of the Internet. At that stage, there wasn’t any substantial kind of plumbing to do that with Windows or OS/2, which were the commercial operating systems of the day. Linux was there, rough and ready and complicated, but you could get to what you wanted to get to. I think for folks of that generation, that’s a common story. Open source was the easiest way to get to the point where you could start playing with the ideas that you wanted to play with.
I think it’s really important we keep that true value. Right? It should be the easiest way to get to get to hacking on mobile, if that’s what you want to do. Or the easiest way to get to hacking on the web if that’s what you want to do too.
Alexis: After you founded Thawte and sold it to VeriSign, is that the impetus for jumping into Ubuntu? Out of all the projects that you could’ve tackled, that’s where you started, an OS push?
Mark: Well the story of Thawte is interesting. As a kid in a garage in Cape Town, miles away from the epicenter of the Internet revolution, downtown Silicon Valley, and yet, I was able to build a service that could compete and for many years be a better service than the stuff that was coming out of Silicon Valley. The reason I could do that was because of a bunch of open source. Linux itself, OpenSSL, and a whole range of other things. So I thought that was an important experience to share: that open source is an incredible leveler.
We need innovation to be able to happen anywhere. If it has to happen inside of a Google or inside of a Microsoft, that’s problematic for all of us, so underwriting open source at scale, making it more accessible to a much larger audience. My goal in doing so is to broaden the pool of innovators that they’re the best stuff just at their fingertips, no matter where they happen to be. Whether it’s in a university, whether it’s in a dorm room, whether it’s at home, whether it’s inside an institution, and I hope Ubuntu has done that for a lot of people.
Alexis: So what was the Linux landscape just when Ubuntu was starting to come out?
Mark: Well, at that stage, rail has just essentially shifted to being proprietary, which was a major shift until then it had been possible to consume pretty much every Linux distribution kind of instantly, whoever you were. I thought that created an opportunity to have a platform, which was of commercial grade and still kept that promise. Then there was also a greater complexity in the market. Most distributions were extremely difficult to use. Again, I thought there was an opportunity to really focus on ease of use.
So those two characteristics I think defined Ubuntu. We continue to stand by those values. For example, with OpenStack today, which is really the cutting edge of Linux at scale, a key focus for us is to be the easiest to deploy OpenStack if you want to setup a cluster and experiment with cluster-oriented Linux. Right? OpenStack on Ubuntu is an amazingly easy experience. We’ll get it so that it’s essentially a no-brainer for anybody who’s got more than one machine. Just spin it up as an open, fat cluster and off you go.
Alexis: How was traction at first when Ubuntu first popped onto the scene? Did it happen fairly easily? Was there any pavement pounding that you had to do?
Mark: Like all things, it seemed to happen both faster and slower than one realizes. So very quickly there was a lot of buzz about Ubuntu, but it took five or six years before real adoption rates really started to climb in interesting segments. So I guess the lesson for me is that people adopt new platforms when they start doing new things. So you either want to be a part of a fundamental change in the way people do technology or do stuff, or you have to wait for a series of those changes to kind of carry you forward. In our case, I think the really big ones have been Linux on the desktop, which went through a phase change of being hard to use to easy to use.
We really rode that wave. Put a lot into it, but lots of other people did too. And the Cloud, which is a fairly profound shift in the way people think about the backend of computing. And we really worked hard to make Ubuntu great in that environment.
Alexis: So I think one of the biggest — correct me if I’m wrong or if you have any figures on this — one of the biggest open-source communities has gathered around Ubuntu. How do you interact with the community and how has that changed over the years?
Mark: It’s a really interesting question. The first thing that we did as a ground rule was to say that we’d respect and value and show essentially the same appreciation for contributions of all sorts. I think there’s a problematic tendency in open source to think that coding skill is the only currency that matters. In fact, a well-rounded project attracts all kinds of contributions. So we just flat out put those things on the same footing, which was very unusual at the time.
We framed that in a code of conduct, which is now sort of standard practice, but we kind of pioneered the approach. And a membership system that you could gain access to, you could get a share of votes essentially that encourage people to pursue whatever kind of contributions they felt best able to make. It could be anything from advocacy to documentation or translation. That I think gives the feel of a place where you have a bunch of different kinds of people working together.
It adds a lot of depth to the project because we don’t just have people essentially vying to control one set of ideas, the architecture of the product. So I think a healthy project does that now as a matter of course. The other thing that we did is we very strongly said we should empower people to do work that’s slightly off the beaten track. It’s important for us to have a clear vision of what the default is.
So the Ubuntu desktop today represents a lot of choices that we make around the default. There’s a default browser. There’s a default desktop environment. There’s a default file system layout. There’s a default all sorts of things.
That means that every one of the end-million users out there don’t have to make those choices for themselves if they don’t want to. They can all get along and share stuff and it just sort of works. But we also provide a lot of common infrastructure and underwrite the operations of quite a few alternatives to that vision, different desktop environments in particular, but also just different ways of doing things. I think that adds a lot of richness to the project as well. People who have an idea that they want to pursue don’t have to bootstrap all of the infrastructure to support a distro, they can essentially just pursue that under the Ubuntu brand and a banner with a remix, and we’ll do a bunch of stuff for them.
Alexis: I just realized that this is Canonical’s 10th year in existence, so happy anniversary.
Mark: Thank you. It’s been quite a ride. It’s amazing how much change we’ve seen in the landscape over that 10 years. The things that people really cared about 10 years ago are much less a source of stress for people today. There are new things that people really want to get right.
Alexis: Speaking of change in the landscape and of all the good things that have accumulated in Ubuntu, what are some of the things in the landscapes that have changed in terms of competition – in whatever sense you consider that – that’s made you reevaluate your direction and or priorities for Canonical and Ubuntu?
Mark: I think the biggest shift has been the emergence of mobile as the centerpiece of personal computing. I wish that I had predicted that because I think we would’ve approached the problem slightly differently. In the era where Ubuntu was conceived, there was this very strong lock on personal computing that was platform driven. So I imagine that in order to accelerate things, you needed to find a way to build an alternative platform. So the platform became very much the focus.
When actually the fact that new form factors were going to emerge, meant that there’d be room suddenly for multiple personal computing platforms and lots of innovation stimulus across those form factors. So I mean that was a very interesting observation, is that change doesn’t have to come in the place where you expect it to come. It comes from something completely different and often that’s the easiest place it can make a change happen.
Alexis: So I guess this kind of segues into the whole convergence computing idea and an Ubuntu Touch. So I have to briefly mention the Ubuntu Edge. You were shooting for – what was it? $32,000,000 on an Indiegogo, made a quite a good chunk of that. It was one of the crowdfunding campaigns that’s raised the most money even though it didn’t fully fund. You mentioned in the video that if it works well — or if it worked well — you would have done this kind of thing annually.
Now with the first Ubuntu touch devices, the smartphones, heading out sometime this year, is that something that you’d ever reconsider? Launching another kind of Ubuntu Edge thing?
Mark: Well, you know, the goal of the Edge was essentially to see if we could create a niche market for the next-generation hardware. Right? If you work inside a phone manufacturer or a PC manufacturer, the real challenge is that you don’t really know what new things are going to be super appealing. There’s a million options on the table and it’s hard to know what the market wants. So we thought crowdfunding might be a way to kind of get a signal from that advanced set of users as to what they were looking for. I don’t think we did too badly in the sort of array things we pulled together into the edge.
Alexis: Not at all.
Mark: So for example, the Sapphire screen. I think we’re all expecting that to be short of the next round of Apple products.
Alexis: I was just about to mention that, yes.
Mark: The core convergence idea, I think you’re going to see as well. I don’t think it’s an accident that Apple started referring to its mobile processors as desktop class. I don’t think it’s an accident that they’ve started calling the Macbook Air and the iPad Air using the same terminology.
Alexis: The same branding.
Mark: I think it’s only a matter of time before they converge those. So in one way or another, I think it served a purpose in highlighting some key ideas that perhaps people won’t really chase it very hard at the time. I think it’s really important for us to have a reason for the silicon guys to really push at the limits of what you can do on a phone. We don’t need faster processes on your phone to drive bigger displays for phone purposes. We just don’t.
The displays are at such a high resolution now that your eye can’t tells the difference. 4K, 8K, it’s not going to make any difference on the phone, right?
Mark: If that mobile processor is actually going to be all of your personal computing and it’s going to be providing your movie viewing, your cinematic experience, and your personal computing experience across small screens and so on. Then we do have a rationalization for putting more cores, more threads, more GPU cores, and memory, and so on into mobile devices. So that’s what I was really excited about there.
Alexis: Alright. So speaking of Ubuntu Touch, the phones are scheduled to see a Q3 launch, right? So sometime this Fall?
Mark: Yes, indeed. So we are coming up to the first RTM drop. So what does that mean? That’s the point at which all the features that we thought we’d needed to get done are in the image. So people really wanted to test it.
They can do so knowing that roughly speaking, it’s a one-point-zero of the phone experience for Ubuntu. We know then that the different manufacturers will have some specific features that they want added. We have that list now from up to lead manufacturers. So there’d be some additional development time, which goes into the specific devices that they launch. But at that point, we have a working mobile product and the releases you can imagine will be falling into a much more normal Ubuntu-type release games where there’s a six-month lead bump. Every two years, there’s a LTS bump.
The predictability of all of that is familiar and friendly for people who are using stuff rather than people who are developing stuff.
Alexis: You mentioned the kind of missed opportunity in terms of mobile. Now you have Ubuntu Touch. It looks quite solid and now that you’re in a field where there’s Android, there’s Windows Phone, and there’s iOS. They have their foothold in the market. What’s your goal for Ubuntu Touch? Do you want to topple one of them? Do you just carve out your own niche and live comfortably in there?
Mark: Well the goal that we’ve articulated is very simply to get into market credibly, which means a credible volume. It’s only a couple of hundred thousand units with more importantly, a very credible return rate. The thing with personal computing is people are either… it either sticks or it doesn’t, right?
Mark: So it’s very important that we get into market with something that is tight and coherent and elegant enough. That a relatively low portion of the people go out and buy the device, take it back, and say it’s not the device they want. So we’re doing a whole bunch of things to try to ensure that. If we were able to get past that initial round, then I think we’ll grow. There’s all sorts of, I think, helpful things that we can do in terms of the App ecosystem.
I think the Android and Ubuntu App ecosystems are pretty close. We can build on that. I’m quite excited by what we can do if we are able to have a credible alternative in the market. We know it’s incredibly hard to do that. The Samsung guys have repeatedly bounced off their launch of Tizen devices for understandable reasons.
Various other projects have sort of struggled along. I think we have something pretty tight and beautiful. I don’t know when last you played with it, but I track the sort of daily developer proposed channel on Nexus 4, 7, and 10. It’s feeling really quite good. I think there’s a bunch of stuff that came out of our last round of user testing that we’ll tweak before we actually go to manufacture.
If you haven’t looked at it in the last six months, go ahead and throw an image at one of your devices. It’s really quite cool.
Alexis: All right. I recently upgraded to a One Plus One. So I think I’m out of luck in terms of official support. I’ll see if I can load it up there somehow.
So this is a question that came from a fan. I asked on Reddit, and Facebook, and Twitter, and Google Plus, and everywhere to see what folks were interested in, and this kind of fits right in here. After the phone and desktop and TV, what’s the next category of devices you want Ubuntu to come to?
Mark: That’s a really interesting question. We did actually provide the core platform for a prototype wearable device for a very large brand name manufacturer.
Mark: They put their own UI on it, but it was exciting to be part of that sort of next wave. I think the areas that are particularly exciting are wearable and VR, but we’re seeing almost every category of device becoming smart. So I think there are lots of opportunities, not just in the sort of one-to-one personal devices. Your one watch, your one set of virtual reality glasses, your one phone, and so on and so forth, but rather in the plethora of devices that we all accumulate. Right? How do we get out of firmware hell where all of those devices are running blobs of Linux with an undefined app capability, and undefined security update capability, and so on?
So yes, we could chase form factors ourself, but I think it’s also interesting to think about how we underpin an ecosystem of people that are all innovating and making all kinds of devices. All of which share a common problem, which is they need a reliable, trustworthy platform that updates itself cleanly and simply. The work that we did in Ubuntu Mobile to create that very clean update mechanism is pretty fantastic. I love the fact that I sit down everyday, and I press a button, and bang. You know those devices are instantly crisply up to date.
I can roll them back. It’s a much cleaner, simpler, more structured world than, say, laptops or servers.
Alexis: Or having to ask AT&T or Verizon to allow you to update your phone.
Mark: It’s not even an issue. It will update and that’s that.
Alexis: Right, right. Exactly.
Mark: There’s a mechanism for them to make sure that the kernel is one they have certified between them and the manufacturer. Don’t ask for the platform. That’s one we provide and everybody gets the same updates on the same day. It’s pretty fantastic.
Alexis: Actually, before I dive further into the mailbag or dive off into my own questions, I find it interesting that you mentioned VR. I’d like to get your take on where you see it going if it’s having any – where it’s moving in terms of applications that aren’t gaming because, as you know, Facebook snapped up Oculus with that kind of vision.
Mark: Yes. I think obviously, true VR it’s sort of self-limiting in the sense that you’d walk into things if you were walking around in the world wearing a set of goggles. So augmented realities are obviously interesting, and I’m not sure that goggles are going to be the way that shapes up. VR itself, I think, has huge potential. Not just for gaming, but for all kinds of training simulation, and fun, social activities, which I guess is a kind of gaming. I’d love to set up a little VR studio just to sort of kick the tires on what’s possible.
I’m sure that this is going to be increasingly mainstreamed. The video arcade of yesteryear–
Mark: — will be replaced by the sort of the holodeck of the future. How much fun would that be? I mean it’d be great.
Alexis: That’d be fantastic. I was actually thinking not too long ago, “Man. Wouldn’t it be nice to see arcades make a come back in that sort of fashion?”
Mark: Well you could have a lot of fun with distributed gaming, right?
Mark: And arcades.
Alexis: Speaking of gaming on Ubuntu and Linux in general, this is also another question from the mailbag. There’s been some progress with graphics drivers on Linux, but what are you doing, if anything frankly, to make Ubuntu more attractive to game developers and gamers?
Mark: I think that the two key things are first to invest in the very lowest layer of the graphic stack. Make sure that it can be super fast, super lightweight, and that it can give developers access to the full-horse power of the metal. So for us, that’s a project called Mir, which is a low-level graphics display server designed to work across every kind of form factor, every kind of architecture, and do so really, really well. If I just by my phone or tablet, how much faster and smoother that is with each successive release. It’s really quite fantastic the work that that group is doing.
Then the other half of it is app-oriented. That’s working with games developers, system designers, and so on to make sure that there’s an ecosystem of developers that are stretching that. I think that the work that the Valve guys have done with Steam is really good. They’re keen to make sure that Ubuntu’s a first class platform for all of their games, just because it’s so easy for developers to consume. As long as that stays true, I think we’ll be making a very credible contribution to that layer of the stack.
Alexis: Now when I asked for questions that the community would like to ask, there was a lot — and I mean a lot — of passion surrounding Mir versus Wayland. I’m not intimately familiar with the intricacies of that, but you’ve probably explained this a billion times by now. For their sake, what went into making that decision? Why did you go that way instead of Wayland or something else?
Mark: Well, there’s a team of developers that had very credible low-level graphics experience that looked at the existing options and the options in development, and felt that the shaping set of values for those options wasn’t necessarily the same set of values that we had. I think it’s really important in open source that we recognize that it’s good when you have that sort of diversity. If we said we only need one way to cache content on the web, well then you wouldn’t have Squid, and you wouldn’t have Varnish, and you wouldn’t have nginx, and you wouldn’t have HAProxy, and any other number ofvery useful tools that people have created. In this case, there’s quite a lot of industrial energy tied up in that stack. There are silicon manufacturers who have an agenda.
There are operating systems and manufacturers that have an agenda. It’s easy to put out quite emotional language, emotive language saying, “Oh. Don’t trust the other guys.” But really the question that has to be asked, what’s your agenda, and trying to stop people from investing in open source. Right? This comes back to, I think, what it’s like to be an open-source developer.
There’s a lot of sledging that goes on. Amongst perhaps the less thoughtful members of the open-source community, it’s very easy and fun, sometimes fun just sort of pile on to that sledging. Well, hold on a sec. As an open source user, you benefit tremendously from that kind of innovating, and competition, and diversity. To trash that is very self-defeating as a user.
So I’m quite happy to underwrite the efforts of the team that wants to build Mir. By every metric, by every measure that I can see, they’ve made something that’s pretty incredibly efficient and fast. Certainly, it’s proven to work across the range of devices that we care about. It’s also proven to work across a range of devices that lots of other manufacturers care about. When you look at the fact that some of the cruises and the slap downs, all the work clearly came from executives that have a corporate agenda and a vested interest in the position.
Alexis: A vested interest…
Mark: Perhaps it suggests that the Mir team should be given some free reign to see what they can achieve. Anyway, that controversy seems to have died away. There’s support for Mir and GTK. I believe there’s support for Mir in a bunch of other places. I have no doubt that all the apps people care about will work well.
We’ll see some good healthy competition at that layer, which is perfectly useful. I do think that it would’ve been a good idea for those guys to go and recruit some other people to provide an independent perspective on why it was important to have that diversity. I guess that’s a social lesson learned.
Alexis: I guess I’d like a bit of insight into the structure of Ubuntu. You mentioned the Mir team and that kind of thing. How do you lay things out within the company? How do you organize things? I know you mentioned Valve earlier.
They’ve got a flat structure. What makes Ubuntu tick? Or Canonical, I should say?
Mark: From an engineering point of view, we have converged the teams that are responsible both for the platform and for various projects like Mir that use the platform or are part of the platform. So with pretty close working relationships between the guys who make, for example, Ubuntu images that get published onto the Cloud, and various people who’d make parts that go into those images. I think you have to keep some separation there, otherwise your platform becomes perpetually in development. Somebody has to have the ability to say that’s not ready.
Mark: And to do so with this certain amount of authority. It’s usually best not to have that person who’s writing the code.
Alexis: It’ll never be ready.
Mark: Right. Or it’s always ready.
Alexis: Or it’s always ready, yes.
Mark: It’s always ready for today. Whereas, for a platform you just take a slightly longer turn view of the maintenance cycle, and the update cycle and so on. So at Canonical, you’re typically either thinking a lot about Cloud, and rather than Cloud, I would say scale out, large-scale infrastructure. What it’s like for people who are managing lots and lots and lots of servers. It could be micro servers.
It could be Cloud servers. It could be big physical servers. It doesn’t really matter, but they’re dealing with large numbers of them fairly anonymously in the sense that no one of them is that important to them. What they care about is just the behavior of hundreds of thousands of them. Or all you’re working on the personal computing side of things where you are either working on particular apps or the shell or the whole platform.
There are some teams that crossover between layers, but broadly speaking, most developers are, I think, either thinking about a personal system or an impersonal system, a cloud system.
Alexis: Right. Okay. So you have 10 years of experience with open source in Canonical alone. If you could distill it, what advice would you give to developers interested in starting a business around their own open source projects?
Mark: I think the first thing is to really think about the commercial model that will sustain it. It might be that the answer to that is just that it’s ancillary. For example, you’re part of a team. You do stuff for your customers. Along the way, you’re going to write this piece of open source on the side, and it’s useful to get it out there as open source. I think that’s a perfectly reasonable position to take. Usually, those kinds of projects will go over the course of their lifetime. They’ll go, you know, they’ll hop across various different institutions.
It’s a very different proposition if you’re saying you want your career and your income stream to be somehow specifically derived from that software, then what you’re looking to do is kind of make a commitment to that code. You’re going to be investing in it and it needs to return, invest back in you. I think it’s important for people to figure out where they stand on that spectrum. There are different tactics that are appropriate, different approaches that are appropriate for those different situations or scenarios.
Alexis: Alright. We’re now 10 years into Ubuntu. Where is it 20 years from now?
Mark: 20 years from now…
Alexis: Look into your crystal ball.
Mark: It’s impossible to speculate. Here’s my profound hope, though. So my profound hope is that people innovating will choose to do it not just on open source, but as open source. I think we pay quite a high price for the fact that a lot of innovation happens behind closed doors. I think it’s really important for us to find ways to encourage people to share their innovations faster.
Eventually, that stuff all leaks and gets out, but we waste a great deal along the way. I tend to think, step back from individual concerns and questions, and look at the balance, benefit to society as a whole. Where we are very wasteful as a society, we pay a high price for that waste. We may not pay it personally or individually, but we do pay that price eventually. So it’s reasonable to step back and say, how can we get companies and startups and individuals, and so on to feel confident that they can publish their core work as open source? Therefore, allowing others to benefit and share and ride on the innovation that they’ve made, while at the same time expecting that they can generate good returns from that as well.
Alexis: Now for a few more fun questions, what’s the most entertaining mispronunciation of Ubuntu that you’ve ever heard?
Mark: Ubu-nut, I think, is a regular and always a gem. There’s quite a lot of Un-buntu and Ubut-nu as well. That’s okay. In a future life, I’ll choose easier names.
Alexis: So what’s your personal hardware configuration?
Mark: I am using a Dell XPS 15, which I really enjoy. It has a touchscreen, which is useful to me for a lot of the convergence work that we’re doing. It’s a nice, beautiful display — the current version. I kind of oscillate between 13 and 15 inches. After two years of 15 inches, I’m ready for something smaller. After two years of something smaller, I’m ready for 15 inches.
I have a cluster of servers that I use for simulating a lot of the work that we do around cloud and cluster computing. It takes me just a couple of minutes to take 16 servers and turn them into an open stack, or a Hadoop cluster, or a set storage cluster. That’s the kind of ease of use that I want everybody to have access to. A lot of the thinking and the work that we do is to imagine the everyday experience of someone who has to do that. Right?
The first time I built open stack, it took two weeks. It was up and it worked, but it was going to be just as hard to change it, as it had been to set it up.
Mark: Now that’s essentially a trivial operation. I think we can achieve the same for other large-scale kinds of infrastructures as well.
Alexis: Do you use any other kinds of operating systems?
Mark: The vast majority of what I do is Linux. I keep some Windows and Mac OS around in the house. It’s fun and useful to see what the other guys are doing as well, but I don’t have any Android, personally. That’s about it.
Alexis: Now one question from the mailbag. I think this might be it for the mailbag. Which features of OS X or Windows are you jealous of, if any?
Mark: Interesting. It’s interesting. I think OS X really expressed very clearly an opinionated view of usability, I guess, two decades ago. It shouldn’t be, you know, we shouldn’t drop our appreciation for what was done there.
In the case of Windows, I spent a fair amount of time over the years talking to Microsoft engineers, and I think what they do at scale is pretty extraordinary. It’s one thing to sort of produce a product for a very specific audience. It’s another thing to produce it for the world. So far, if I look at Apple, I’d say, well in a sense, they have the luxury of a niche market. It’s like once they could squeeze down to just the people who really cared about a very specific set of things, and is therefore, forces the company to say we’ve got to be the kind of company that really cares about a very specific set of things.
That’s kind of beautiful and empowering. What I think is interesting about Windows is you’ve got a platform that has to work on the average PC. The average PC is a mess inside, right?
Mark: If you look at BIOSes and firmware and all of the nonsense that goes on in the manufacturing process and cycle, it’s a very different constraint for a company to have to operate under. I think there’s a ton of really good stuff that we don’t really appreciate in the Linux world because we get to essentially recompile the whole system every time, or often enough that we sort of can fairly easily obliterate the mistakes of the past. But in the Windows world, they don’t have that luxury. I certainly can appreciate some of the techniques and approaches and institutional discipline that’re required to produce platforms that can be maintained and extended for decades.
Alexis: Two last questions here before we let you go. This one’s going back to a bit more serious. What’s one mistake you’d rather not repeat in your journey with Canonical? Or on the flipside, you can answer, which one decision that you’re particularly proud of, especially if it was kind of murky at the time.
Mark: I think the key thing I have learned — there’s a great expression from the guy who was the CEO of Netscape, which is, “The main thing is to keep the main thing, the main thing.” The challenge of being a platform is that everybody who joins has a different idea of what the main thing is.
Alexis: “You’re losing sight of the desktop,” some users will say or that kind of thing. Right.
Mark: Right. Not just externally, but internally as well. Nobody joins a company like Canonical without having in their minds is a complete, fully flushed out, completely rigorous vision for what an operating system should be because everybody’s using them all the time. Everybody’s doing different things with them. I think what might have been useful would be to start with some very specific pieces of the puzzle and just focus on those for a while. At the time, I felt that the operating system was the hard thing to do, and so that needed to be my focus.
I think it’s true the operating system is the hard thing, and I’m glad that has been my focus. It did make it harder to get the right thing out of other specific pieces of the team because the operating system tends to crowd everything else out. An OS is a big beast that can easily be a distraction and allow you to forget that the main thing needs to be the main thing.
Alexis: So if you weren’t working on Ubuntu, what would you be doing for work or for fun?
Mark: You know I think in life… life is short for all of us whether we realize it or not. I think you have to have a pursuit that you feel very passionate about, you feel it’s important, that you feel it’s bigger than yourself. I’m in a very fortunate position that I can do that, but it always comes at a cost.
There’s always the sense that one could be doing something for oneself, rather than for other people. In my case, I would find something that I think I thought was always interesting and important and pursue it with everything that I had. Who knows what that might be, but for now, Ubuntu is wonderful and it fills the day and quite a lot og the evening as well.
Alexis: Personally, I would like to think that would be space travel of some sort and duke it out with Elon Musk, but a man can dream.
Mark: I think that’s clearly somebody’s passion. I tend to think that the really complex things are much closer to home. One of the strong impressions I had looking at earth from a distance was, in essence, how artificial and simplistic the rest of universe is compared to the complex mess that we have at home on the planet. So I’m kind of motivated to focus more on things on this spaceship that we’re all sharing, right? It’s the only one we’ve got.
Right, it was a pleasure to speak and I wish you all of the very best! And the people who give Binpress a whirl, I hope that it works out really well for them.
Alexis: Thank you very much, Mark. For our listeners, we’ll catch you next week.
Mark: Cheers, Alexis.Binpress Open-Source Podcast