Remarks by Bill Buxton, Principal Researcher for Microsoft Research, about the challenge of designing a new Web experience
Las Vegas, Nev.
March 18, 2009
ANNOUNCER: Ladies and gentlemen, please welcome Bill Buxton. (Music, applause.)
BILL BUXTON: Good morning, here you are in Las Vegas, right? The city of luck. So are you feeling lucky? I mean, how could you not be feeling lucky, right? The economy is doing fantastic. Right? (Laughter.) And you're all in this profession called ‘experience design,’ which of course your clients have no idea what that means, and it leaves your mother thinking, "Couldn't he have been a dentist or an accountant?" Right?
And the whole thing is that you couldn't be actually in a better profession precisely because of this economy, by getting this experience right, you get this return on that investment that will let you not just survive, but drive some kind of economic turnaround.
The important thing to understand here: It's not about luck. At least luck is not about chance. And that's the key thing, it's different. We can control our luck. See, I have all this kind of control. So I went to a Chinese restaurant and I got this great thing, this was last week, it said, "Your hidden creative talents will be revealed." Thank God. My 60th birthday was last week and finally, finally they're going to be revealed. This is fantastic. (Laughter.) I spent my entire life trying to be a designer. So you can share this with me. So we'll bring this with us to get this return on experience.
So the key thing I want to say is that I'm not joking when I say this is a really good and a really important time to be focused on experience. And I want to tell you this because you need stories. Stories are the best form of viral marketing. If you're talking to clients, repeat what I'm going to tell you right now.
Think back to 1926 to 1929 and the introduction of industrial design in North America. That's when the first people started consultancies and hung up their shingle that said "industrial designer." What they were doing then as industrial designers is precisely what we're doing now as experience designers. That is, starting a new profession for which there are no degree programs, where we come from all different directions, and we make up stuff, but we really make it. And furthermore, by the kind of experience that comes from this, we help our clients make it.
Let me give you an example. In 1929, Henry Draper was a great industrial designer. He was a theater set designer on Broadway, started his design consultancy. And from that, one of his major clients was Bell and the telephone and the receiver that all of us, at least our parents, grew up with came from his design. He's fantastic – helped shape the whole field of industrial design.
Raymond Loewy came out of fashion design. He wasn't an industrial designer to begin with, but he brought us all kinds of graphics like the Coca-Cola bottle, also the logo, all these other logos, and the greatest design car ever made in North America in my humble opinion, the Studebaker Avanti. God rest its soul in peace. It really was fantastic. Huge, streamlining, the full refrigerators and stoves, the redesign of those that brought those companies to success came from him, return on experience.
Walter Dorwin Teague is another one of the three heroes who started the field of industrial design. He started his practice in 1926. Kodak was his principle customer. They had this appliance that everybody wanted, it was this thing, this is a vest pocket camera. It came in every color you wanted as long as it was black. It was really successful, but they wanted to grow the market.
So what did Teague do? He had this great idea. He said, "We're going to repackage it, we're going to release it in five different colors simultaneously in beautiful silk-lined boxes, and we're going to charge a premium of double the black ones. And he did that and he drove the design of Kodak throughout the '30s and '40s and '50s. And the lesson from this and these successes at that time from '29, think about it, what idiots would do a startup of a design consultancy on the eve of the Great Depression? But every one of these companies is still in business today. They didn't just survive the Depression, they thrived in it as did their clients because they got the return on experience. And the lessons they learned are still applicable today as this example shows you.
Guess what? Apple had a product. It was white instead of black. It was really successful. They wanted to grow the market. What did they do? Gee, what a concept, five different versions, five different colors, guess what? Except for one color the same one, same number, same colors. Did they copy? No, they learned from the past and exploited it.
And what I'm trying to say here is all of us, if we understand what's going on before, we can learn and apply and grow that knowledge, and that's called creativity, that's called design, that's how you get the return on experience.
Now, what is this thing called experience? We're talking about experience. Is it just the new black, the same old thing, same old stuff in new clothes? No, it's not. It's a re-focus of attention. It's what in many ways successful products always had, but we never called it that.
So the principle thing is: Industrial designers generally talk about designing objects, the thing, whereas it's actually the experience that's engendered by the thing in the box which is the true experience. It's the experience from your interactive media that is the true product of your endeavors. It's not what's on the screen. It's not the graphics, it's what the graphics trigger, what they prompt.
I explain this because I spent some time working with the designers at Trek, the guys who designed all of Lance Armstrong's bikes, plus the mountain bike that Roland Green won the world championship on. And they gave me a bike, you know, choose your clients well and you get good perks.
And this is my mountain bike. And I love saying this, it truly is a carbon copy of the bike that Roland Green won the world championship on. Now, here's a rendering of it. And it's objective, it tells you everything about the bike, you would recognize my bike beside this picture, that it's the same bike and so on and so forth.
And the question is: So what? There's millions of other bikes, why would you care? That's not what I bought – actually, they gave it to me, but that's what I would have bought had I bought it.
Here's another rendering of the same bike. Now, have you ever seen this little animation from Pixar called Red's Dream? It's got some characters, it's got some energy in it, right? It's starting to give some character. It's not a neutral rendering like the previous one. It's down this path that I'm talking about, but it's still not about experience. That is not what you buy. That is not what you aspire to have is a thing. No, no, no, no, no, no. This is what experience design is about. It's like scaring yourself so you almost need Pampers, going down a hill, and you're just going nuts, and you know you're not going to die because your equipment is really good and, man, is it great. And you're just living to do it again and again and again and keep pushing the ante.
What kind of bike is this? I don't know, you can't see it. Who made it? I don't know, you can't tell. It's not about the bike. It's about screaming into that river with a huge smile on your face and your adrenaline pumping. But it's not specific to mountain bikes, for any product, any service, any online thing, any object you make, you have to know, what is the nature of the grins and the adrenaline or the smile or the laughs or whatever that you're trying to provoke from your users?
Same with me, same with Microsoft, all of us. That's our job. This is what we're aspiring to. Now, the problem is if we win, if the high return on the products comes or the service comes by shaping the experience, how do we achieve that? How do we design it? How do we as designers, over half of you come from a design background, another half –probably less than half come from a development background. How, collectively, can we actually really tailor what we're making to generate that type of experience so we get the appropriate and desired return?
Well, here we can do it by design, but what does that mean? Where are the problems? Where area the pitfalls compared to graphic design or architecture or industrial design?
I want to use an example to try and point this out. So this is my phone. It's a Dash phone made by HTC. Here's a little exercise. So I know that half of you are designers, but even you who are computer scientists, and even if you're pencil- challenged, like me, I bet you can do the following exercise. In 15 seconds, bang, draw my phone.
Now I want you to draw that phone in your mind's eye, if you have a piece of paper, do it. And I guarantee –- I guarantee no matter, as I said, how pencil challenged you are, every single one of you is capable in 15 seconds of drawing that phone in such a way as it's distinct from this other HTC phone, my second phone. It's called the Touch.
Now, the point I'm making here is you can all do that. You can draw the object. You can sketch it. That means you can work through ideas quickly. All right. Draw my phone's interface. Now, clearly, the interface is different from the object. But you just noticed something, you can't sketch the interface. You can't work through concepts as quickly about the interface as you can about the object. That should worry you because design is about rapid iterations and working through different concepts so that you not only get the design right, but get the right design.
All right. We still haven't gotten the end of the test. The real test is draw the experience of using that phone. Because if you believe what I said with the mountain bike example, that's what our job is. How do we do that? And if we can't have the same facility, quickness and versatility to work through different concepts when it comes to experience as we could when it comes to the object, we'd better start rethinking our way of working. We'd better start rethinking the structure of our teams, the composition of our teams, and the nature of the tools that we use because if we want to be successful designers, we have to be able to get that same fluency.
And it's not about the phone. Everything I've just said is true whether you're delivering Web-based stuff to desktops, to placed ads in terms of kiosks, or in terms of interactions on the mobile phones or whatever. Everything I've said is true about the need to be able to do design.
So how do you do this? And for me, if I'm going to really distill it to one basic thing, it's about finding a balance between two different components because I can hear you already, and more importantly, I can hear your clients and your managers already saying: "Well that's fine, but Mr. Buxton they have told you you have to be able to do five different versions of every single question you're being asked, each one equally valid. So we can't afford it because we're already behind schedule. We don't even have time to do one solution and you're telling me we've got to do five?"
What are you going to say to them? If you don't – if you can't do five equally valid solutions for everything, everything question and come up with and not have a preference, then A) we're probably not doing things right because our job is not to answer questions, it's to ask the right questions to get us to the right question that'll get us to the right answer.
And so doing multiples is critically important and the challenge is how do you balance doing multiples, which is the essence of design, as opposed to just following down one path and gradually refining things, with a budget in terms of dollars, time, and personnel? And, again, it comes down to technique.
And the principle thing there, and this is around sketching. When I talk about sketching, the whole point is they're cheap tasks, it's disposable ways to encapsulate ideas, embody ideas in a minimum amount of time, really quickly, low-fidelity so that we can actually explore a much broader design space. And to do that, you can't just do drawings like we used to do, but there is something called sketching, there is something where we can sketch experience, and if we have the right tools and the right team and the right technique, we can do it as effectively as our colleagues in graphic or industrial design do.
Now, what's the essence of experience? To me, it's about timing. The principle thing is it's even harder in cinematic form because it's not taking place over time, it's interactive, and it can take various paths. How do you do the pacing, the whole – there is a plot development to the things we do. We know this.
And so I'm just going to give you a summary of some of the techniques. So here's just a way of sketching the most trivial interface that I can think of. It basically is, you turn on the interactive screen and it says "touch here, not here" but it gives you a place to touch in both places. Because that's what we see in the top left.
The second one is, because I'm an idiot, I push the "don't touch here" because I'm a contrarian, right? I'm a designer, that's my job to do what it says not to do. So I do that. Well, the next screen pops up and says, "Can't you read?" And then there's a “continue” to go back, and that's the top-right one.
Then I come back and sort of click on the “continue.” I click the right thing, and bingo, I come through and it says "thanks." OK, really stupid. The point I'm trying to say is I don't care how good you are in Flash or Blend or HTML, you can do this on Post-it notes much faster to sketch this out, get the flow, than you could in any other technology. If you are a trained programmer, the last thing you should do is start programming because you worked four years or more really hard to get those skills, why would you waste them on trivial things that are unworthy when you can actually go through these other techniques, get it right, and then apply your programming chops to really nail it?
Now, this kind of sequence is like a cartoon. If you just draw the arrows, it starts to become something what would we call a “state transition diagram.” So we've got the states, which are the different states of the interaction, and the arrows show the transition one to the other.
The state transition diagrams are really interesting. And I want to talk about this because there's a really important point, and here I'll have to make an interlude and make the worst joke of the time. So, I have to say this, I'm Canadian, so I'm allowed to say this. What do Canada and transitions have in common? They're both dominated by the states. (Laughter.) Right?
Now, the deal is this: how many of you – I know this is trivial, but this graphic, how many of you have been in design sessions where something like this is passing, pretending it's a story board – is up on the wall to describe what you're about to build, right? Both hands and the foot, right?
Yell when you see that. Here's why: If you don't have as much detail in the transition as you do in the state, you're going to get it wrong. It's not about here – watch, right, I'm here in this post. I'm here in this post. Now I can get there like this, I can get there like this. OK, I'm going to stop being an idiot, but the point I'm trying to get at, how you characterize that is not how I got there, it's not where I went, started to or where I went to. That's true with everything we do in interactive. So if you can't characterize the movement, the timing, you've got a problem.
Now one of my buddies, a guy named Ron Burge in the U.K. used some of this and extended a little bit where – I'll show you in a larger screen – where he starts to have the screen shots, but notice below the screen shots, he actually graphs out the state transition diagram of the sub-network of the screens in that sequence. And it's actually kind of neat because it turns out in the human factors literature, we know that the complexity of the graph of that sequence is a very strong indicator of the complexity of using learnability and so on of that part of the user interface. And therefore, it's going to have some effect on the experience. Well, that part's really good. But the problem is, it still doesn't talk about the timing, it still doesn't talk about the nature of the transition.
Now, what I envision is that some day we're going to have tools where we can sketch stuff out in the appropriate kind of renderings, we can have graphs underneath in terms of the state transitions. But when we have the graph, we can click on the states to be able to get the graphs that are there, but we can click on the arrows to actually then start to sketch and draw the movements and the transitions and characterize that so that we can in fact have equal facility moving back and forth.
And when I talk about you need better tools, this is the kind of thing I'm envisioning that's going to happen. And the reason this is important is that I think – you know this whole story, if the only tool you have is a hammer, you see the world like a bunch of nails. Well, the thing about design is that if you – we talk about it as being the most creative business in the world, it's not at all. It's the most negative because you start with a million ideas, the sky is the limit. And you finish up with one single design that ships.
What that says to me is you're throwing away 99.9 percent of every idea you have, and most of them are yours. And because yours are all thrown away, you're the most important designer in the whole project because you triggered him to have the idea that ended up. Who should get the reward? You, because you're the one who triggered it, he just had the idea because you set him up. It's a team, right?
What I'm trying to get at, though, is that here we're doing ideation. And we need one set of tools, a different management style. You cannot be really anal, right? These things are far too important to take seriously. We need to be able to play. But as we come around, we've invested more and more and more and we change and we move from sketching to prototyping, from ideation to evaluation because we need different tools along the whole path, and we need to know when to use which and who should be using them.
All right. What about Microsoft? So Microsoft has – what I want to talk about now isn't that Microsoft gets design, I want to say Microsoft when we're talking about this return on experience, we're not just talking the talk, we're walking the walk because it is as important for us as it is for you.
So let's take a couple quick looks. When I joined Microsoft about three years ago, there was only one person in the leadership who came from a design background, it's now 10. There's been serious growth from people we've brought in to get fresh blood, but there's a lot of people that have been promoted from within.
But it's not just at the senior level. In less than two years, the UX head count has grown almost 150 percent. And we're growing UX people at about twice the rate that we're growing technologists and engineers. And there's about 800 designers and UX researchers now in the company.
But I want to make it personal, I'm wrapping up. Just give you two people here just as an example. This could be you, but Monique and Audrey – Monique has been about five years with the company, Audrey's been about seven months. But they're young, junior designers and they've designed one of my favorite products that has come out of the company recently, and it's this thing called the Arc Mouse.
And it's a simple thing, but it's this piece of elegance. You know it folds, it's just gorgeous, and it works really well. The point I'm trying to say is you create a culture where you have the management from young designers who are incredibly talented to have their way, get their ideas out there, and get them shipping and do it. And, yes, they are designers. They use the standard type of classic things you would do.
So the next thing I want to do here is just sort of quickly talk about the Zune. So OK you can sort of say the Zune, you know, iPod's got that whole space driven. But that's not the point. Here's the point: The Zune went from the mark one to the mark two in nine months. The versions in this photograph came out nine months after the first one. So from a standing start to being able to have something which absolutely is world class in consumer electronics from a company that doesn't do consumer electronics, was from my perspective, a remarkable feat.
What I want to say about this is that it's not about the MP3 space, it's about a measured mile where you've got something direct to compare it again. And the most important thing is it's not about the device, it's about the software and the whole ecosystem.
And so I want to just sort of end up with the quickest demo you've ever seen at a MIX conference to make this point about ecosystem. So I'm on Win 7 on an HP Touch Smart, and so – and that's how short that demo was. Scott's going to show you that stuff later, I'm not going to waste time. (Laughter.) I told you it would be the shortest demo in the world. (Applause.)
Now, quick, so here's this. We've got the Zune – you're going to see that there's no damage here. I work in research. Research is where we get people, we study people, we study design, we do all these types of experiments to figure out how to do this experience design. And from that, we come through to this notion that we've got Surface and then from Surface we go to Win 7. From Win 7 we go to Silverlight. And then we go back and we come down to Live labs and the deep Zune, back to Silverlight.
And then there's this thing called the Internet, it's going to be big some day, to the PC, and yes, through six degrees of separation, we arrive at Kevin Bacon. (Laughter.) And the point is this: That almost every part of this humongous company called Microsoft is involved here, from research to Live Labs to the Windows team and they're all singing the same song, they're all working in a coherent way to figure out how can we – and then through you as well, enable people to get the ultimate return on experience? How can we let – and that's the demo you will see later – is how can I make it so that we can deliver on thin clients over the Web the same type of rich experience that you get on an expensive piece of hardware like Surface or like in Win 7?
So it doesn't matter what the platform, where it's delivered, how it works. How can we have a unified way so we can develop these interaction techniques, this rich interaction, deliver the experience and get the return no matter what the platform is and where you only have to do it once, you don't have to do it separately for each platform. That's how we get the return.
Now Seneca, right? He's a philosopher from Rome 2,000 years ago, made this statement: "Luck is what happens when preparation meets opportunity." What I've been trying to do is try and scope out the nature of the opportunity and ensure that in my opinion it is real. It's not just real, it's critical to get the return experience if we want to get out of where things are right now, and our clients depend on us to be able to make this happen.
Now what Scott's going to do is then say what can we talk about, what can we show you, how can we help you in the preparation so that you too can get the optimal return on experience? Thank you very much. (Applause.)