We talk with Linux creator Linus Torvalds

Colonel of the Kernel

© Images Courtesy of Swapnil Bhartiya

© Images Courtesy of Swapnil Bhartiya

Article from Issue 192/2016

The 25th birthday of Linux is an important milestone for everyone in the Linux community. Who better to help us ring in the occasion than the man who started it all: Linux creator Linus Torvalds. 

The 25th birthday of Linux is an important milestone for everyone in the Linux community. Who better to help us ring in the occasion than the man who started it all: Linux creator Linus Torvalds. Linus was a 21-year-old Finnish university student when he sent a note to the Minux user group stating that he was working on a new operating system that was "… just a hobby."

Since then, Linux has found its way onto millions of systems around the world, and Linus has become an iconic figure for a vast community of Linux users and developers. Our news editor, Swapnil Bhartiya, sat down with Linus for an interesting exclusive interview on the kernel, the Linux desktop, and the quest for balance in work and life.

Linux Magazine: It's the 25th anniversary of Linux. You sent out that humble email some 25 years ago…

Linus Torvalds: It's not even clear what the real anniversary date is. There are really two of them. August 25th is when I made the public announcement, when I emailed that I would be announcing it. The actual release, 0.01, was two weeks later, but that one never got announced publicly at all.

LM: Can we consider Linux's beginning when you started working on it? When was the very first line of code of the Linux kernel, as we know today, written?

LT: I don't even remember exactly when the coding started, but it must have been roughly March or April of 1991. I got my machine in January; that was the first Linux machine. But at that point I had no idea that I needed to write my own operating system. Then it took a few months to get Minux set up and from there it took me some time to realize that I needed to do my own OS. I think I started coding maybe in April. It is clearly more than 25 years now. It's a long time.

LM: Linux has come a long way from that humble beginning. It's running on almost everything. Do you ever reflect back on 25 years to see what all it has achieved?

LT: Not really. Because of the whole 25 year thing, I get that question a lot from journalists over email. One thing that I react to is that so much else has changed in my life and in many ways Linux has stayed the same.

I mean the details of Linux have changed completely – the kind of machines it runs on, the number of people involved, the number of companies involved. All of that has changed dramatically. But at the same time what hasn't really changed for me is the fact that, even 25 years ago, the reason I did Linux was the interaction with the hardware, to explore it, and the technical challenges that come with it.

While the details have changed, I don't write any code anymore; it's more about directing people. The basics, the really core basics are still the same. The thing I'm interested in tends to be the close interaction with the hardware, the technical challenges.

In a very real sense, Linux has changed much less than everything else around it. The fact that, in the last 25 years, I now have three almost grown-up children, and I moved from Finland to the US. I went from being a university student to working at a startup to working at The Linux Foundation. Everything else has changed so much, and, at the same time, the reason I'm doing Linux has not changed at all.

LM: I understand that back in those days you had new hardware and you wanted to do something. Do you have the same itch to scratch now, as most hardware these days works with Linux or comes with some form of Linux preinstalled?

LT: No. That has changed. It used to be, but not for very long actually. I wrote Linux because it was something I needed, but that stopped very early on. At that point, it was doing what I expected it to do. Since I made it available, there were people coming in and saying "I need this" or "I think it should do this." If that was not the case, I would probably had been done by the end of 1991. I will give you an example. Early on my machine had 4MB of RAM. That's ridiculously little, but it was plenty for what I was doing, and GCC, at that point, didn't actually need that much. But somebody had a machine with 2MB of RAM, and he could not compile the kernel under Linux because that was not enough for GCC. That was around Christmas 1991. I started working on paging to disk so that this guy with two megs of RAM could actually build his system on it. So literally in late 1991, I was implementing major features that I actually didn't need, but because somebody else was asking for it.

From very early on, the motivation for doing Linux went from this is something I need, to other people are using it. It changed, but it changed so early that I think of this as, this is how I do Linux. That the things people ask for really come from outside. That's not new to me anymore.

LM: I recall my previous interview with you where you said that it's not spreading the code, but also the vision. Different people come to Linux with their own vision, instead of you creating a vision for everyone. That's quite remarkable. I have never seen that with any other project.

LT: And that's what made it [Linux] much more interesting. Some other projects start with a vision of where they want to go and Linux never really had that. Linux didn't even start as an operating system. That meant that from the very first day after I made the announcement, when people started making comments, I think I was more open than a lot of other projects to just taking input from other people.

If you looked at in the time frame 25 years ago, the other big operating system project was obviously the BSDs. They had literally 20 years of history behind them. They had people who knew how things were supposed to be done. Linux didn't have that, because I didn't have that. So Linux was a project that was much more open to just saying: Okay, we'll do that.

It was much easier.

Initially individuals, and later companies, just wanted to push Linux in their own direction. Because I didn't have any particular goals in mind, I was perfectly happy with that. The only goal I had, and that's still the case, [was] that you can do whatever you want with Linux, but let's just make sure that the technicals are good.

LM: By design, you created a project that allowed people to take it any direction they wanted, so what are the compromises that you are willing to make, and, at the same time, what are the compromises you will never make with Linux?

LT: The thing I tend to care about is if you're getting a driver, let's just get it working. A driver does not impact the big picture development; it doesn't impact maintainability in the long run. One driver may be really ugly, but I'm very happy to get code like that merged into the kernel. Usually it comes through Greg's [Kroah-Hartman] staging tree, and some of those drivers are really ugly; they are doing a lot of bad things. But that's fine. If they don't impact another driver, we can fix them up.

But when it comes to really core, very core kernel stuff, I do want to make sure that there's a clean design. That there's a point that it's maintainable in the long run. Partly because the code is well written, but partly also because you have a notion of where hardware is going and where usage is going. So you have to pick something where the interfaces work for different hardware, where the interfaces work well for different loads, because different people will use Linux for so many different things.

So that's where I don't have strong opinions on what you do with Linux, but I do have very strong opinions on the technical side of making sure that really core stuff is well set up so that we don't paint ourselves into a corner.

LM: Let's change gears for a while. Let's talk about Microsoft. They have changed so much in recent times. They have a Linux-based OS for Azure; they are bringing PowerShell for Linux. What do you think about it all? People often quote this, did you ever say that if Microsoft ever wrote applications for Linux, you won?

LT: I do think I did say something like that. But you have to realize that back in the late '90s the situation was very, very different. We, in the Unix camp, used to make jokes about Windows crashing. It's gotten much better frankly, not that I've used Windows for a long, long time.

I think the situation just has changed. It used to be that it was a very antagonistic situation, partly [because] Microsoft used to be so dominant. They were the target for a lot of antagonism. I used to make jokes about Microsoft, and I actually stopped doing that sometime maybe 15 years ago. But it used to be very common. It used to be very much Microsoft versus Linux and Microsoft versus Unix. One of the reasons I stopped making Microsoft jokes was that it always generated all this unnecessary attention.

There were stories written about Linux versus Microsoft but that was never actually the case with me. It was more like making a cheap joke about Microsoft, but at the same time I didn't care. That was not the reason why I did Linux. Microsoft was mostly, fairly neutral on Linux except for those few outbursts from Microsoft that were bad. They do seem to have changed in the last couple of years.

LM: What do you think about this transformation of Microsoft?

LT: I really don't follow it that much. I don't think it's even so much about Microsoft. I think the computing market, and the operating system market in general, has matured to some degree. It's such a different situation than it was 20 or 25 years ago.

What I am trying to say is that it's actually across the whole board. Tech companies used to be much more hostile to each other. You had the Sun versus Microsoft wars. I don't think it was a very nice marketplace. There's still hostilities, and you still have the Google/Oracle thing going on and things like that, but I think the market has kind of calmed down, and people are not quite so antagonistic anymore.

LM: That's true. But I have come across desktop users who are still antagonistic and don't like the idea of running Microsoft software on Linux. So from your perspective, is it okay to use Microsoft apps on a Linux desktop?

LT: I remember back in those days before we had good applications for doing presentations, very early on I was doing presentations about Linux and I would actually use PowerPoint on Wabi. Before Wine, Wabi was Sun Microsystem's Windows emulator. So I would actually run PowerPoint on Linux back in 1995 or something like that.

LM: (Chuckles) Is that the reason you don't do presentations anymore?

LT: (Laughs) No. I'm happy that it was only for a short while I actually used PowerPoint. If you look at the whole "me versus Richard Stallman," I always felt that to me the whole open source is not about how commercial software is evil. To me open source [has] always been about how I think it's a much better way of doing development.

I think it's more productive when you cut through all the bullshit, through the borders between companies, and you can just work together with other people. That's why, obviously, it's so much more fun, too, because you don't have to worry about who you're talking to. You can just take input from anybody who has a good idea.

So I think open source is just a, technically, much better way of doing things. I think the licensing is important but it's important as a way to keep everybody honest, to keep open source open. And that's great. It's never been this religious war for me that it is for some people.

So when Oracle started porting their database, when a lot of big companies came in and started making proprietary programs for Linux, I was like: that's great!

LM: It's just a platform like any other platform. You should be able to run whatever it is, right?

LT: Right. But at the same time, I do think that in the long run open source tends to take over because it's a better way of doing things, especially the infrastructure part. If somebody is doing a database that isn't open source, I would be worried about that in the long run because it's such an infrastructure play; everybody needs it. MySQL obviously already took a lot of it, and there are other open source databases.

In the core infrastructure play, I think open source will take over in the long run. But I don't think it's because of ideological reasons; it's because of technological reasons.

LM: It's already happening. AI and machine learning is touted as the fourth industrial revolution, and a lot of machine learning technologies are being open sourced by Google, Amazon, Microsoft, … . Everyone seems to be open sourcing their stuff … .

LT: If you actually look back historically, basic applications used to be, maybe not exactly open source because people weren't conscious about it, but if you look at how software programming was done in the very early days, people would just ship tapes to each other.

So almost always when you're doing something completely new and the market doesn't really exist, it's very natural to share anyway. And then when the market becomes bigger and economics take over, the sharing often stops. That's where having a good license that ensures the openness can't go away is actually important. It means that when a project grows up from being experimental, from being fun to work with, and becomes a big commercial thing, if you pick the right license it will stay open and actually it will grow. I do believe in that.

LM: Licences are important. If you were to start a new project today what license would you choose? [See the "Free Software Licenses" box.]

LT: I think it is like a personal choice; different people have different opinions on what matters. I don't like the language of GPL v2; it's much too long and much too complicated, but I think the basic ideas are really the right ones. Which is basically saying: "Hey, I give you source code, you have to give me source code back, and this is irrevocable, and nobody has any extra rights, and it stays like that forever." I think that's just wonderful.

I picked the GPL v2 originally because that was what I agreed with. The legalese and all the language is less important to me, I wanted something that was well known, and I still absolutely agree with it. I would still likely choose GPL v2.

I don't like the GPL v3, and the reasons I don't like it are pretty much out there. But I understand people who want to use it. I also understand people who want to use BSD. There are a lot of licenses out there.

Free Software Licenses

Several different licenses meet the Free Software Foundation's definition of "Free." Free software licenses fit roughly into two categories:

  • Copyleft licenses, like the GNU Public License (GPL) – include a clause that says, if you make changes to the code, you have to share your changes with the community
  • BSD-style licenses – do not include a copyleft clause, which means you can adapt the code without sharing your changes.

Parts of Apple's proprietary OS X operating system are based on BSD Unix, which had a BSD license and, therefore, could be integrated into closed-source code. The same thing could not happen with Linux, because it uses the GPL v2 copyleft license.

The GNU project and the Free Software Foundation now recommend GPL v3. Linus and other kernel developers objected to some of the changes with the GPL v3, especially regarding the treatment of Digital Rights Management (DRM) technology, so they stayed with the GPL v2 as a license for the Linux kernel.

LM: I have noticed a pattern that community-driven projects tend to choose GPL-like licenses, whereas big company-backed projects tend to choose BSD or MIT type licenses.

LT: It is true that if you're mostly a commercial entity you tend to go with something like BSD because then it leaves you more options going forward. But that's where I actually think it is a mistake. It does give you options to do other things going forward, but it also means that the community around it does not feel protected because they know that everybody has these options, too, to just take it and go with it. As a developer, if I start a project and I give it away, I really want people to use it, but at the same time I want people to give their improvements back, and that I think is fair. I call it tit for tat.

If somebody else does a project and I want to join that community, I also feel much more protected if that other person chose the GPL because then I feel that any improvements I gave back will be part of this continual improvement. While if you use a BSD license, I think the community is less protected. But, this is my personal opinion; there have been lots of very successful BSD license projects, too. So I don't know.

To some degree the license proliferation has been a huge source of confusion, so I'd rather stick to a couple of well-known licenses and just say: You may not love the GPL v2, but at least everybody knows how to deal with it. You may not love the version 3, but it's one of a handful of licenses. There is a new, shorter version of the BSD license, and everybody knows how that works. It's nice.

By the time you get outside of those three, you have to say which one is it? Is it the Mozilla license, is it the so-and-so? So, I'd rather keep things simple and stay with GPL version 2 just to avoid complexity.

I made a mistake once. I started Sparse for doing static analysis of the kernel. It's not a very well-known project; I started it as a compiler front-end thing. I decided to pick a very modern, simpler license that wasn't well known. (chuckles) I don't even remember which license that was. I thought, it seems to be much simpler than the GPL v2, and it was. But it was not compatible with anything else, and it was different enough that you had to always explain it. We finally got it relicensed. That was many years ago, but it was actually a painful experience to use one of the smaller licenses just because it became such a problem working with people who wanted to use some of the code from Sparse for some other project. It was not compatible with anything else, and it was just not worth the pain to me. So I'd still pick GPL v2.

LM: Let's now change gears from professional life to personal life. You work from home. I also work from home, and it's quite a challenge when you have kids. So how do you maintain the work-life balance?

LT: I've talked to people who cannot work from home because they have kids and they say: "I can't concentrate. I need to go to the office so the kids aren't there and they don't disturb me and they don't distract me when I'm working."

I am not a very nice person. My kids never distracted me or disturbed me when I was working because they realized very quickly that daddy's not a very fun person if you come and scream in his ear when he's working. So they would never come into the office. It was not that the office door was closed and it was not like this is the border you can't cross. There was never that kind of rule. It was more like, that's not the fun room. That's daddy's work area.

Now we have a big house, and the kids are big enough where it's not an issue anyway. But I much prefer working from home because it means that I don't have to plan working. So if something happens, I just go to the computer.

A lot of the time, I don't have to be on my computer; I read mails on my phone or tablet, but not during the merge windows. Merge windows are special times where I have to be at the computer for eight to ten hours a day because that's what it takes.

LM: You always merge from home?

LT: I try. Sometimes I do merge windows when I'm traveling, and it's really painful. Outside of the merge window, a lot of the time I end up waiting for results. For example, there is a bug discussion that's going on, and it wasn't me who found the bug. Someone else found it, and it's being reported and it's really just email – and emails coming in at odd hours, especially if someone is from Europe or the rest of the world. So a lot of the time, I end up waiting around for people. I actually read email a lot on my tablet or my phone, and, when something comes in, I go up to my office; that's when I need to do actual real work in front of a real computer. But I can do the first order filtering of just seeing what's going on from the phone.

LM: Are you like Richard Stallman, where you bring your laptop to the dining table; in case there is pull request, you will move the plate aside and start working … .

LT: No. No. I won't do that. I want to be there. When I'm going away for a week and I can't react to pull requests the same day, I let the top maintainers (not everybody) know that I will be traveling and I will have so little time that I won't be answering that fast. I don't need to feel that if a pull request comes in I will pull it five minutes later. No. I will pull within the next 24 hours.

LM: So is the family time like strict family time, no work?

LT: My wife is stay-at-home mom. When the kids were small (and these were three different kids) in the US you have to drive them to anything after school. So we always had this situation where my wife can only be in so many places at the same time. So I would just know that after school usually somebody needs to be driven. That's much less of an issue now because two of the kids drive themselves and are off to college. But that's kind of the work/life situation I was in. It wasn't like "Okay, dinner time is when the family gets together and then we watch TV afterwards." No.

LM: Do you watch TV?

LT: No. I don't. My wife watches some TV.

LM: Have you heard of Mr. Robot?

LT: No but I have heard of Game of Thrones. I've never seen it.

LM: Now that your kids are big, they're going to school. They are moving out for college. You will be in a different phase of your life. You will have all the time to yourself. So will you still be doing Linux or will you start something else?

LT: That's odd because when I got my first child, at that point people were worried: Oh, now he has children and he's moving away from university. They (kids) happened about the same time. Some people were saying, "Oh, that's bad. Now he will stop maintaining Linux." And now the children are going away, and you're asking me whether I will stop maintaining Linux.

LM: I am not saying you will stop maintaining Linux, I am saying you have more time now, so will something change?

LT: I don't think it will change a lot. We've had this whole fairly traditional family style where the husband brings in the money, even if he does it in a bathrobe, and the wife takes care of the children. So, it's actually my wife who's getting impacted a lot more by the fact that the kids are going away. But we still have one; she's in high school. She will get her driver's license in a couple of months, so she will be driving herself at that point.

LM: I have two kids and I miss them when I come to these conferences. How do you feel, emotionally, about your kids growing up and moving away?

LT: I'm so happy about it. I thought that babies are not even very interesting. Children really start getting interesting around the age of nine. I'm not saying exactly when they turn nine but at the age of nine, they change. Before nine they're very immediate. They don't think really; they react.

At the age of nine, roughly, they start to have abstract thinking. They start thinking much more interesting things. They turn into humans. I saw that as a huge step forward. When a child turns from something that isn't very smart, maybe slightly smarter than a dog (laughs), don't get me wrong, and turns into a real human and starts thinking at a completely different level. I was very impressed with that. I thought it was very noticeable, too. Now, teenage years, not always pleasant. You get arguments; you get boyfriends coming home. Ah! Not always the happiest moments. But I'm actually really enjoying seeing them turn into adults. Again, I think when you [grow] into a real independent person, that's another big step and that's happening right now for them. I'm actually very happy; I think it's nice to see them grow up.

LM: Do you have any influence on them as a software person, as a celebrity who is changing the world?

LT: I don't think so.

LM: You're very humble, that we know. But is there any discussion at home with your daughters about what they want to do or what you're doing, any influence like that?

LT: No. Not really. They obviously very much know about the Linux thing. They have computers that run Linux, and that's not an option (chuckles); this is not something they got a choice in.

My oldest daughter is going to a technical college. She will presumably do computer science and engineering. We'll see. But it was not something we pushed them into. We did push them into getting a real degree. We said we'll pay for college, but it has to be a real job. You're not getting some degree where you don't see yourself with a real career afterwards. It was like you don't have to be a tech person.

The second one right now is going into pre-med. She's looking at neuroscience. The third one says she's interested in being a school teacher, which my wife used to be.

LM: School teacher in the US?

LT: In the US, I am not completely convinced it's a good career. But coming from Finland, I know being a teacher is a really very well respected job. I'm not complaining; I'm slightly sad that in the US they're not very well respected.

LM: Finland has an excellent education system; did you ever consider going back for the kids? We lived in Germany, and we were tempted to go back after having kids.

LT: That's where we thought we would go, too. We thought we'd move back to Finland when the kids started school. By the time the kids started school, we had learned how the US system worked. And then we thought by the time they have go to college maybe we'll move back to Finland, because college is free. And even that never happened.

LM: A last question while we are still on the topic of family. What will upset you more, if your daughter brings a boyfriend home or brings an iPad or Windows machine home?

LT: Oh, no. The iPad or Windows machine is much worse. The boyfriends have already happened. It's like you know it's going to happen. It's all fine. It's like whatever. It's just slightly awkward still.

LM: Going back to Linux. These days we hear a lot of security-related stories popping up. What's your perspective: Are there more bugs being discovered or now are more bugs slipping in?

LT: Quite frankly, part of it is that there's this whole scare culture about security. We have bugs, don't get me wrong. To just give an example, fairly recently there were tons and tons of articles being written about how 1.4 billion Android devices were open to vulnerability to this networking bug.

Yes, it was a bug. We actually followed the spec a bit too closely, but from a security angle the vulnerability was that you can do it in the lab, but in the real world it really doesn't matter. Yes it was a bug. In theory 1.4 billion devices were involved, but anybody who actually knew what was going on would ever really care. But then you have these scary stories written about it. That's why I dislike the security community so much because they try to drum up these stories.

We had bad security bugs. We had really nasty ones where I said: "That's just stupid; that was not good." They happen, and we try to fix it. But, more importantly, we are working fairly hard on hardening infrastructure.

LM: What kind of work are you doing in that area?

LT: Bugs will happen, but when they happen, hopefully, all these other safety mechanisms mean that, in practice, they don't end up being security issues, or they are so hard to exploit that it's not usually a problem. So, we do have a very real project going on. It's part of CII [Core Infrastructure Initiative].

Kees Cook has been the main person leading it on the kernel side now. He is instrumental in integrating code to kind of create these fallback safety nets. We're working on that, and we've worked a lot on security issues in general – for example, the supervisor-mode access protection, supervisor-mode execute protection. We are doing things where when bugs happen and a user program fools the kernel into jumping or accessing something in an insecure manner, the hardware will actually catch it because we talked to Intel to make sure that they have the hardware capabilities to do it. So, we've done a lot of things like that. Will we ever get it perfect? No.

LM: But it's not always the kernel, there are many different components.

LT: I think one of the big wake-up calls was all the OpenSSL issues. The kernel is quite special when it comes to security. But when you look at the real security issues, it's often in libraries that are used in all the applications that try to do something secure, and, when those libraries fail, they create issues. We are working on it. I will not say that we'll ever be perfect. We're working on being better.

LM: You are doing this at kernel level. It's true that the kernel community patches bugs quickly, but the problem seems to be with distributions or hardware vendors where these patches don't even reach the devices.

LT: That will be a bigger problem in IoT [Internet of Things], it is not so commercially viable yet. But look at mobile phones; some mobile phones never get updated. Is it annoying? Yes. That is part of why we're trying to do the hardening thing so that even when you don't update, hopefully, it won't be catastrophic. It's a very hard problem to solve.

The good news is a lot of people are doing security on many different levels, which is the only way to do it right in the first place. Yes, we do the best we can do in the kernel, but distributions are trying, too. People are moving on into using containers to limit, when security problems happen, to a smaller part of the system. People are doing a lot of such things. You add all these different layers that, hopefully, make it more and more inconvenient to punch through all of these layers to get to the really serious bugs and exploit them. But, I have to say, some of those attack people are pretty smart people, and clearly they're not all criminals; some of them work for the government.

Part of the issue is that as developers we are looking at giving access to new features, giving access to hardware, and things like that. When we do that, we have a completely different mindset from the people who are looking at attack surfaces. When you read about an exploit you say "Wow, I would have never thought about that," because you come from a different direction.

LM: Let's talk about the desktop. People always ask which distro you use. What's the reason behind that choice?

LT: I try not to care that much about my distro, but what I don't want is to have three distributions that look different because I put the same thing on my wife's machine, my desktop, my laptop, and my kids' machines. For the last 10 years, probably since I switched away from Power PC, I think I've used Fedora.

Right now most of the machines are Fedora 23, a couple of them have been updated to 24, but then I will install my own kernel.

One of the reasons I like Fedora is they tend to be fairly good about new kernels. Red Hat, in general, has been very good about kernel resources. They have been helpful in testing, and, obviously, they do have lot of kernel engineers. One of the reasons I ended up going with Fedora was that they did a good job on the side that I cared about.

LM: The Linux desktop that we wanted has not happened yet, but Chromebooks are doing wonders, especially with the arrival of Android apps. If I can paraphrase, is this the desktop you were looking for?

LT: It's not the desktop I'm looking for, for my needs.

LM: What I meant was Chrome OS as Linux on desktop in consumer space.

LT: If you were looking at things from an end-user standpoint, I think these days it actually makes sense to see mainly the browser. The browser is clearly a big portion of the system and Chrome takes that as the starting point and says "we want you to get the browser experience and then we have some other small things on top." There's a lot of people who seem to really like Chrome OS as a desktop just because they don't have to worry about kind of a traditional desktop.

I think right now it looks like Chrome OS is really taking Linux and making it happen. It's just that the thing I use is the old workstation environment, and it's really doing well. It's doing way much better than the older workstation and what it used to be.

I think that's a sign that the market has become more mature. If you look at why the PC made such a huge difference in the '80s and '90s, it was because the PC finally made a unified platform that could do a bit of everything. You didn't have that before. That's why the PC really changed the world.

But the fact is that back when the PC changed the world, you really needed something that could do a bit of everything because you didn't know what you would need.

What has happened in the last 5 to 10 years is that people have started knowing what they need. Now you can make specialized devices again. The cell phones took a bit longer, but they clearly matured, too. In the last two years not a lot has really changed in cell phones, now you know where the goal post is. Now you don't need the original kind of PC that could do a bit of everything anymore. I think the market has changed, and what I thought of as a desktop was the general purpose PC that doesn't necessarily make sense anymore, economically, where we are now, except as a workstation. There's a lot of people who don't really need that anymore and are perfectly happy doing lots on their phones and maybe on a tablet.

LM: Can we say that the year of the Linux desktop is here with Chrome OS or you are still waiting?

LT: I think, yes, maybe Chrome OS is basically the Linux desktop, but at the same time that whole thing has not played out completely yet. Give it maybe five more years and see where things are. But yes, I do think that we may be in a situation where Chromebook is the desktop for people who don't do development and that's fine.

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

comments powered by Disqus