INTERVIEW by TIMOTHY BROWNE | PHOTOGRAPHY courtesy of EBEN UPTON
When you think about a credit card-sized computer with 1080p high-definition 3D video that’s wowed testers at Google, Xbox, and other developers, you’d imagine said computer would be a pretty pricey system. Thanks to the efforts of Eben Upton, project director at the Raspberry Pi foundation, and his team, that computer actually costs a mere $25. Envisioned as a launch pad for young people to learn programming, the Raspberry Pi truly is a small wonder. I talked to Eben (because as awesome as the computer is, it just can’t give interviews yet) on the eve of the device’s launch about what it is, how he sees it fitting into educational curriculums, and just why it’s called Raspberry Pi.
I know you’ve been very busy with the upcoming launch, so to start, what is Raspberry Pi?
There’s Raspberry Pi the foundation and Raspberry Pi the device. Raspberry Pi the foundation is an organization we set up in Cambridge a few years ago to try to promote computer science education. A number of us working at the University realized that every year the number of people applying to read computer science at the University was declining, and that every year they knew less. We went from an environment where everyone knew some assembly language to an environment where some people had written web pages. The aim of the foundation — it’s a UK charity — is to try and fix this problem by making this device which is also called the Raspberry Pi, which is a $25 dollar computer. The intention is for this to be cheap enough that every child can have one. It’s a reasonably capable general purpose machine, it’s bundled with a lot of programming languages and programming tools, and it also has one or two standout features, things it can do very nicely: it can play 1080p high definition video and it has reasonable 3D graphics, so you can play and write interesting computer games on it.
Why did you decide to call it Raspberry Pi?
Not that kind of raspberry pie…
There’s a history of fruit-named computer companies! [laughs] There aren’t many that aren’t taken, and Raspberry is one of those. The obvious ones, I think, are British companies called Apricot and Tangerine, and I’m sure there are other fruit-named computer companies as well. So that’s the fruit. When we first looked at this, what we ended up building was an ARM-based PC that runs Linux, and that wasn’t what we set out to build. We originally set out to build something much more limited. What we wanted to build was something that could run the Python programming language, and so “Pi” was kind of a pun. So the name stuck and it outlived the justification for the name.
…that kind of Raspberry Pi
That’s great! A lot of articles about the device really sell the high concept of a $25 computer. Can you talk about how you see it being used educationally and socially?
I guess my view about the educational aspect is that the way it’s likely to work, at least in the UK, is that putting this kind of thing into the formal, government-sponsored curriculum is difficult. We have actually had some success with this. We’ve had some meetings with the government; there’s been some suggestion that the UK government is starting to de-emphasize the learning about Microsoft Office aspects of school computing education, and emphasizing the stuff that might actually make you money when you grow up. From our point of view, the most likely vector for this is a child buying it themselves, or your enthusiastic physics teacher, which was often how it worked in the 1980s: you’d have an enthusiasm for technology and be prepared to do stuff outside the curriculum, maybe outside the schoolhouse: the after-school computer club. So I think that’s probably the way it’s going to be used most.
I know the device has yet to ship, but where do you see it evolving from this point?
That’s right. This week has been Chinese New Year; the device is being manufactured in China, and we’re really hoping that they’re very close now. All of this year is going to be about scale. Scale is hard, because we’re a charity. Building these devices is profitable. Generally when you have a thing that you can build that’s profitable, you can go get capital to pay for scale. Now, obviously, as a not-for-profit, we don’t have the ability to sell shares in ourselves, so it’s hard to get risk capital. So what we have is money that’s been contributed by the loans from the trustees of the foundation and some loans and donations from high-net worth individuals in the Cambridge area. But that’s really only enough to build ten or twenty thousand units at a time, which isn’t anything remotely like the demand that we have, so a lot of this year is going to be about finding clever, hacky ways to get scale.
I think part of that is going to be about enabling other companies to manufacture these devices under license and part of it’s going to be about open source, about getting the device out into the world where anyone can make an unlicensed version. Those are kind of important. Then within the foundation, once we’ve found solutions to the scale issue, it’s enabling the things we actually care about. We’ve got this device out, but obviously it doesn’t have a very polished educational package associated with it, so a lot of it is going to be about arranging for that kind of polish.
Once the device is out there, do you see the consumer really taking over from there and developing its potential down the line?
Yeah. We’re six guys, we all have day jobs, and I’m very lucky that my wife has agreed, if you’ve seen the website, to spend a lot of her time making the community and PR aspect of this a success. But there’s a limit to what we can do. We don’t want to circumscribe what the community is going to do with it, and of course the open source thing is a big part of that. Obviously, the community can do whatever it wants on the software side, though until we’ve released the hardware as open source, they’ll be dependent on us for license and manufacture. So, absolutely. We have no idea what this is going to do.
Have you released beta versions to testers?
Yeah. We’ve had these alpha versions out in the wild for four or five months. We’ve had beta versions of the final device that we built at the end of December, and those have been in the hands of a mixture of people who bought them on eBay, a few strategic partners, and a couple have gone to the Xbox media center community, who’ve been enormously helpful to us. So these things are out in the wild, and they’ve been given enough of a kicking that we are confident that the design is good. That was the aim of the beta devices, to get to that point.
Is there anything in that testing that really surprised you, that you hadn’t conceived of?
I think these devices have only gone out to people who have been able to describe what they’ve wanted to do with it lined up well. They’re rare enough that we’ve haven’t sent too many out speculatively. We sent quite a few out to Google in the hope that they would do some interesting, speculative stuff with it. But apart from that, they’ve mostly gone to distros, to game developers, to middleware developers, the Xbox guys, people like that. Up to now there hasn’t been much scope.
What can Daily BR!NK readers do to help Raspberry Pi as both a foundation and a device?
Buy them! [laughs] Buy them when they come out and do interesting stuff with them and share whatever it is you’ve done. Don’t buy one and put it in the cupboard; buy one and do an interesting thing with it and tell the world what you did.