'Intel doesn't design actual products'
17th Sep 2010 | 15:06
Why Intel wants to create platforms and services - not just chips
It's all about the experience
Never mind the technical details of Sandy Bridge or the 50Gbps silicon photonics link; what Intel wants to create is not just chips but platforms and services - and it's depending on Genevieve Bell's Interaction and Experience Research Lab to make them right.
Bell is an anthropologist as well as by training; her father was an engineer, her mother another anthropologist and she spent much of her childhood on field studies among aboriginal tribes.
With that background, she looks at what makes new technologies successful in a very different way. Take electricity.
Consumers already had gas lights and oil lamps and candles but an electrical supplier in Niagara made electricity interesting by throwing glamorous parties with women – known as the electric fairies - in diaphanous frocks standing on electrical wires and clutching lightbulbs.
The way the latest iPhone adverts don't mention that it's a phone? That reminds her of the way Bell labs persuaded people to buy a second phone with the 1959 Princess phone, a telephone that came in five colours and was advertised with the slogan "it's little, it's lovely, it lights; at no point," points out Bell, "did they suggest that it makes a telephone call or talk about how many calls it makes or how long the cord is or that you might need a phone plan…"
Thinking about experience
It's not easy teaching the engineers at Intel to think as much about the experience as the product that their technology will create, says Bell but she thinks it's vital.
"Intel doesn't design actual products," she explains – and her lab isn't trying to create products that OEMs will just put their name on. "But unless we know what we need to deliver ultimately, it's hard to design things right."
Before the Experience Lab, she was working with the Digital Home team; a job she jokes that she got because of her criticism of Intel's ill-fated Viiv platform; while Intel engineers were promising to "unleash the PC in your TV" she was pointing out that people already had a screen in their living room and they didn't want it to behave anything like a PC.
"We put up with things in PCs that we would never put up with in a TV. Imagine the first time the TV told you it needed a new driver or the first time your Tivo said it needed to defragment before you could record a programme – or the first time your TV blue screened!"
Instead, she says, Intel should have been asking "What is the essence of TV that people love so much? What is it that's so compelling that we still organise our day, our time and our furniture around it?" The very un-PC answer is that "People love TV because it's not complicated. It's one button to a story they care about."
TVs "not bigger PCs"
TVs "not bigger PCs"
The same goes for many of the scenarios where Intel thinks the PC will end up. "TVs are not just bigger PCs," says Bell pointedly;"phones are not just smaller ones – and cars are not just PCs with wheels."
It's not just about shoe-horning an Atom chip into the dashboard or the fridge (the Experience Lab is working on projects to use facial recognition to identify drivers and passengers and on a kitchen system that would recognise ingredients and even accidents like spilling coffee).
"The Lab is about how we move from defining experience as a list of features to something more intuitive – and to do that we need to ask the right questions. When you know why people love something you can work out how to make it even more beloved and you can answer some questions about how you create something that will be loved in the future." Another useful question to ask is "What do people hate?"
You wouldn't know it from projects like Viiv, but Intel has 15 years of experience in social science research, cognitive psychology, human factors engineering and design.
For the first time though, the Experience Lab combines social scientists with engineers and both disciplines will work on research projects like facial recognition that works in real time by offloading work to a cloud service – or to spare cycles on another PC in the home.
"We need to move from things we know about people to things we can hand to engineers," Bell says; "and rather than having an anthropologist standing up and saying 'we can build this' I have engineers who can make that happen". Bell and her colleagues have spent over a decade working out what people love; over the next ten years she wants Intel to build more of them.
One idea; "Intel is always very interested in thinking of how you move stuff around; that include human IO, not just compute IO. Think of different sorts of interactions, new forms of input methods that go far beyond simple replacements for the mouse and keyboard."
Liked this? Then check out Hands on: Internet Explorer 9 review
Sign up for TechRadar's free Weird Week in Tech newsletter
Get the oddest tech stories of the week, plus the most popular news and reviews delivered straight to your inbox. Sign up at http://www.techradar.com/register