15 / 06 / 2011

Bill Buxton: “The best interaction design is transparent, almost invisible”

  • Linkedin
Bill Buxton: “The best interaction design is transparent, almost invisible”

This interview is part of a series of articles dedicated to designBill Buxton, principal researcher at Microsoft Research (disclaimer : RSLN is edited by Microsoft France) and specialist of interactive design, tells us about his lifelong quest for natural interactions between humans and computers. A look at our technological future. 

Note to our french readers : if you rather like the french version of this piece, follow this link 🙂

RSLN: Over the past 10 years, we’ve seen a tremendous amount of new technologies coming out: new game consoles, tablets, smart phones, etc. Is it the affirmation of the digital designer? 
Bill Buxton: The truth is that I’m not sure the innovation cycles are any faster. I think there are just more of them, and by having more people playing the game visibly, it gives the impression that things are moving faster.
What’s certain though is that design plays an increasing role in the creation of new products. It’s starting to be integrated into many companies as a primary focus… but not enough! If a product if beautiful but doesn’t work, that isn’t design. At most, it is styling.
But then you see products that stand out in all regards, then design has most likely been at the core of their production process. Design has to be at the core of the enterprise, it’s a spirit of creativity and innovation that must be shared at all levels of the company.
RSLN: You’re a specialist of interaction design. It’s quite easy to imagine how an industrial designer would design a product like a chair or a car, but it’s harder to imagine the approach you would use to design an interaction between humans and computers…
Bill Buxton: In the past, when we started working on a new interaction, we probably started with what the actual task is, and we’d start designing interactions to fit in, assuming that people were working at a desktop computer. That was almost always the case from 1985 to 2005.
The thing that’s changed in the last ten years is not the technology per se but how it’s being used, and most importantly where it’s being used: inside, outside, in a car, on the street, in a restaurant, etc.
The key thing right now is recognizing the larger context in which your product will be used. If I’m making a mobile phone application, I need to know if you have a laptop with you, where is your phone, if you have an MP3 player, how it is connected to your phone, etc. I need to understand the digital, but also economical, social and geographical context in which it’s going to fit in.
That’s when you realize that interactive design is closer to social sciences than computer sciences. It resorts to ethnography, anthropology, or psychology. When you create a new product, whether heardware, software, or a service, you have to think of it as part of a larger ecosystem. You also need to anticipate how this ecosystem is going to change over the next 5 years. 
RSLN: How can you anticipate what tomorrow is going to be made of? 
Bill Buxton: First, you have to make a few interactive sketches of the concept you’re working on. Just like an industrial designer would use a pen and a sheet of paper to draw sketches of a product, designers of interactive products and services have to render tomorrow’s interactions in a way that they can be experienced before they are built.
This may sound like a paradox, but I believe that the only way you can engineer the future is to have lived in it yesterday. I’ll give you an example. Back in the 1970s, John Gould, a researcher at IBM, found a very ingenious way to thus render a computer system capable of perfect of speech recognition. He wanted to test if this was a technology was worth investing in before investing in it.  
That is, he wanted people to experience it before it had actually been built. So, what he did was ask people to sit in front of a computer and to speak into a microphone. Everything they said appeared in real time, perfectly written on the screen. The speaker experienced a computer that recognized and transcribed everything they said, but in reality, what was really happening was that an experienced stenographer was hidden behind a curtain, transcribing the text being heard in their headphones.
That experience cost essentially nothing. Yet, it enabled Gould and his team to understand the interactions at work in such software before it ever existed.   It was a preliminary “sketch” – an early concept rendering, serving the same function that sketches have throughout the history of design. 
RSLN: What is a successful interaction design? 
Bill Buxton: In some sense, a successful interaction design would be transparent, almost invisible, to the point that the user would be almost unconscious of the experience until after it’s over. It is just like magic. A good interaction design also needs to fit well within the society of appliances that surrounds it. I own a car equipped with an integrated communication system.
With this system, I can call my son without taking my hands off of the wheel or my eyes off of the road.  I can speak with him as if he was sitting in the seat beside me. I hear him through the car stereo speakers, and my voice is captured by a microphone built into the car.  
If I get to my destination before the conversation is done, I can park the car, pick up my phone, close the door, and continue chatting with him without thinking about what just happened. Within a few minutes, the device serving as my phone changed from my car to my actual phone, and I didn’t even notice.
Almost everything changed technologically but nothing interrupted my conversation. That’s a wonderful example of a transparent and integrated interactive design. 
RSLN: What you’re describing underlines your quest for natural user interactions… How can a human-computer interaction be natural? 
Bill Buxton: It’s actually a very fun question because if you think about it, what’s natural is what we’re born with. When you’re born, you can hear, you can see but not very well, you can scream, poop, pee and eat, and that’s about it.
But there is one other thing that we’re born with, and that is the capacity to learn and to acquire skills that are appropriate for the world we live in. If you live in the Arctic, you’re going to have different skills than if you live in the jungle. Our job then is to design for the skills that are appropriate for you. I think of technology as a mirror. For it to be natural, it has to be a perfect reflection of its user.
It has to accurately reflect the motor/sensory skills, cognitive, social and emotional skills of the intended user. When the interaction is an accurate, rather than distorted, reflection of all these skills, then I believe is it worthy of being called natural.  
RSLN: Can you give us a concrete example?
Bill Buxton: Kinect (the game console without a joystick developed by Microsoft) and its “ancestor”. When you leave the check-out counter at the supermarket and you have a bag in each arm, the doors open automatically. Kinect’s great-grandfather is the sensor technology that opens the door for you at the supermarket.
It’s a very natural interaction that reduces complexity without you even noticing. Interaction design used this old experience in a new product, Kinect. This game console didn’t grow out of a vacuum.  Without noticing it, we’ve been living the future for decades already. 
RSLN: Another example of a natural interface is the touch and multi touch technology… touch screens are everywhere today… 
Bill Buxton: The multi-touch interfaces that you find on many of today’s modern devices grew directly out of a drum that my colleagues and I built in 1984. I say that not to boast or make undue claims, but because I want to make it clear that I’m not anti-touch.
I helped invent the technology. But just like any piece of technology, I believe that touch is appropriate for some uses, and not appropriate for others. As important as it is, I think that touch got over-adopted.
Today, you even find touch screens on the consoles of some automobiles. That means that you have to divert your eyes from the road to use the technology. Good old physical knobs on the contrary enable you to reach down and find the radio dial without taking your eyes off of the road.
In this case, touch screens are dangerous and should not be allowed. Touch screens offer a natural interaction in a certain context, but no technology is natural in and of itself. I think people got confused… 
RSLN: What are your sources of inspiration? 
Bill Buxton: Every designer that I know and respect is passionate about things outside of his or her work. Some of them do pottery, others sculpt, etc. It’s crucial to be widely experienced in order to be a good designer.
Personally, it’s music that brought me to design. I was playing electro-acoustic music, and I was frustrated with the instruments that I could purchase. If you asked me what I am, in some ways I would say I’m a luthier (stringed instrument maker, ndlr).
With the technologies that I make, I want to capture the same qualities as those from the best luthiers. A violin not only has to sound good, but it also has to feel good, to be made of the right material, to be adapted to my movement, etc. 
RSLN: What technological future can we expect? 
Bill Buxton: In the future, we can expect a greater integration of the different technologies. We are heading towards this “society of appliances”, an environment in which user experiences are much more connected with one another. Imagine that the type of technological integration that we saw with the mobile phone in my automobile existed in your home, your office, your school or your supermarket.
Then realize that your car is “just” a special kind of building.  It has doors, a roof, a heater, a stereo, seats, etc.  This analogy is not far fetched. It is by seeing such relationships that we can, today, know the technologies of tomorrow. But we have yet to imagine the experiences they will offer us. 
> Photos : 
Bill Buxton, by Kate Hutchinson
1949: Born in Edmonton, Canada.
1973: Bachelor’s degree in music from Queen’s University, in Ontario.
1978: Master’s Degree in computer science from the University of Toronto
1984: Contributes to the invention of a digital drum regarded as the first “multitouch” interface
1994: Becomes principal researcher at Alias Research in Toronto
1995: Receives the Canadian Human-Computer Communications Society Award for contributions to research in computer graphics and human-computer interaction
2002: Creates his own design company, called Buxton Design
2005: Joins the team of Microsoft Research
> A look further : 
– RSLN tag: "design numérique"
– Please, have a look to issue #10 of RSLN, dedicated to design :