Interview: Linus Torvalds on Linux and Raspberry Pi
This year’s Millennium Technology Prize, the “Nobel” of the technology world, has for the first time been shared between the two nominees. Stem-cell researcher Shinya Yamanaka and Linus Torvalds, creator of the Linux operating system, will each receive €600,000. Linux underlies Google, Facebook and Amazon, but unlike commercial rivals Windows and Mac OS, Linux is open-source, free, and anyone can improve it. Jacob Aron, of our sister site New Scientist, talked to Torvalds in advance of the prize announcement.
You’re known for creating the world’s most widely used piece of open-source software, Linux. How is it different?
There are so many differences. There’s the fact that especially over the last 10 years we’ve tried hard to make Linux portable so it runs on anything from cellphones to supercomputers, and most people are probably familiar with the Android operating system, which uses Linux. People are also using Linux when they look at Google, Facebook or Amazon.
When I started, it was for my enjoyment. At some point I realised that I wanted to make it available on the internet. I was not that aware of the open-source movement, and I used a non-open-source licence that didn’t allow people to freely copy Linux. I had a friend who introduced me to the discussion.
So how does “open source” work in practice?
Instead of trying to control a project, you allow everybody to be part of the community and encourage them to make changes. The only real rule is you have to make those changes available to others. It turns out to be a wonderful model for development because, especially for Linux, there are many different people with different things they want to do. Having thousands of people involved means you get a very well-rounded system.
Did you think Linux would become so popular?
Hell, no. I think it was very healthy for the project that even today my time frame is a couple of weeks, or months, instead of this vision of where I want things to be in 10 years and how I’m going to take over the world.
What about the future of Linux?
I don’t tend to think in market terms. When I think about the changes we’re likely to hit, one of the big things for Linux is the hardware. When I see somebody making a suggestion for improvement, I think “how does this work in a world 10 years from now?” If you make the wrong choices and it turns out it really doesn’t scale up when you use 200 central processing units, then 10 years from now when everybody has a lot of CPUs, you might be up shit creek.
What about Raspberry Pi, a Linux-based computer costing $25? Will that change things?
What’s interesting about Raspberry Pi is that it’s so cheap almost anybody can buy it as a throwaway – throwaway in the very good sense that it could get people involved in computers who otherwise wouldn’t be. For a lot of people, it will be a toy gathering dust, but if 1 per cent of the people who buy it are introduced to computers and embedded programs, that’s huge.
It can get people into the mindset of using a computer to do everyday jobs that even five years ago it would have been ridiculous to use a computer for because they were big and expensive. With Pi, you can say, I wouldn’t use a real computer for this, but maybe it can control my water heater.
Is a future where homes are run by computer only possible with open source? If Raspberry Pi had to run Windows, would it be too expensive?
Yes. Open source is a very powerful way to try something new. The thing about trying something new is that 999 out of 1000 cases will fail. Having this easy entry into trying something new means having one case where it works is very good.
Raspberry Pi is a way to allow experimentation on an even smaller scale because you have the hardware, too. When you aim for that price you can’t afford not to use a free, open operating system.
Jacob Aron, New Scientist