Dan Klyn: Hello, my name is Dan Klyn and I'm a master debater. I'll just admit that to you now. In high school, I was in competitive debate, and not only did I meet my then girlfriend, now wife in debate class. I started out dating her partner and then moved on to much better things. This is me, and the caption, if you can't read it, reacting to a rumor that their team has staged an upset win. That wasn't what happened. It was a loss as usual. But we thought maybe we won. In high school debate...anybody here a high school debater? Were you a high school debater? Thank you. There are two ways that you can win at high school debate. The preferred route, since this is an educational activity, is through evidence, through facts, and through having a good case. That was not the way that I would win. The other way that you can win in debate is by having better arguments, and especially arguments over definitions. I went to a small, private Christian school. We didn't have a budget for the debate team to have researchers working for us like some of our competitors, so in the absence of being able to outperform with evidence, I had to learn to outperform with arguments. Arguing about definitions is something that I love. When I saw Jared Spool bravely offer a new definition of design, my sense was that I would enjoy taking this definition apart because that's what I do. But the more you left this work on you, this is really important. Design is the rendering of intent. I'm from Grand Rapids, Michigan. Our company has offices in Grand Rapids and Ann Arbor. Anybody here been to Grand Rapids. Awesome, wow. Who here is from Grand Rapids? Just a few of us. OK. If you are in Grand Rapids or if you are near the airport especially, you may have noticed this. Think about what Jared said, "Design is the rendering of intent." What was the intent of Steelcase in building a giant pyramid out by airport? It wouldn't surprise you to hear...I wouldn't think that this structure is a challenge to work inside of, and such a challenge that Steelcase has vacated the premises. In retrospect, how do you intent and design work? Some future archeologists might look at this and that the intent of that building was to enshrine the idea that back in the day, our employment was so stable. The furniture business was so good. We could make preposterous buildings and have the intent be for them to be cool buildings, not the intent for them to be functional spaces. Intention, if design is the rendering of intent, the stuff that we make has structures. Instead of going straightly from intent to structure, like sometimes we do -- this is one of my favorite images of all time -- I bet the tax preparers get people walking in, asking for pizza. [laughter] Dan: Intention and the resulting structure here, this was expedient. The tax business is only really hot for a couple of months during the year. They knew that they might be mistaken for a Pizza Hut, the structure of what they were doing. This is another thing that's really important about how Jared said that. Design is the rendering of intent. Is this the work of designers? All of the business deciders involved in this decision were doing design. They were rendering their intent. The structure that came out of that process, how would you know that was good or not? We might say that's really crazy, but how do you know? How do you know what a good structure is? In the parlance of how the kids talk today, how do you know if you're doing it right? As some of you know, I was grateful to Ivo for breaking the Richard Saul Wurman bubble in advance for me. The Understanding Group is founded on principles that were created by this wonderful 79-year-old man, Richard Saul Wurman. How do you know? I really want to know that I'm doing it right. His advice is always hard to take, but usually really important to listen to. There is no right way. What's the right way to offer a tax preparation service on a busy street when your budget is small and your needs for how the space is configured aren't that fancy? That Pizza Hut could have been the right way. Letting go of saying, "What is doing it right?" I'd like to talk about doing it good. There are any number of good ways. The Pizza Hut way might be a good way. There aren't any right ways. What I'm going to tell you about is a way to talk about how to design structure, how to get to good structure for the things that you're working on. If the things are apps that have a really narrow purpose that aren't all that complex, I think these principles still work in that context. The context that we work in looks like this. Which is vast amounts of all sorts of really complex information and hundreds of requirements. Not the requirement to say, "Yo." I'm going to tell you some things. I think these things apply equally well in both a small app instance, as well as a big information system. My credibility in saying that, I have much more experience with this than with the prior. Four ways to get at good structure for what you're working on. The first thing is back to Jared's definition: If design is the rendering of intent, then you have to get systematic with your intention. Who here was at UX Thursday last year? Awesome, many of us. I was, too. I got to tell you about a tool for making intention explicit, for making intention systematic. Something we've listed from the work of Richard Saul Wurman called performance continuums. In most businesses, businesses often offer services to people. As a result, people are complex and contradictory. You don't get to remove the contradictions, at least we don't. You have to work through the contradictions. Rendering our intent about all of these hard things to decide, are we going to service the customers that we have now or are we going to acquire more? Most businesses don't get to make that an either-or. In the middle of this continuum has the word "yet" and so, there's tool of a performance continuum to say what are these things that are held intention, and then where are we today. Mostly, we service the customers that we have. Where do we need to go? What would good mean? When we design it and we'd look at it, what does good mean? For the intenders, you can model their intention with this tool of performance continuums and you can stack up these continuums. Typically, we will have 10, 15, 20 of these that we're intention, and then gathering all those dots of intention to figure out what it wants to be. What would be good structure? It's tempting, some of the dots have a way that they want to go together, so it's tempting to just jump to what does the UI look like. But back to the pyramid, in between intent and structure, there's an opportunity with structure to have the structure mean something and to have the meaning of the structure line up with what the meaning of what you're supposed to do is, and for that to multiply and make you really effective. What if we didn't jump right to UI after we figure out what the intent is? The next step would be to get systematic with meaning. If you're Fidelity, what you mean across every channel imaginable, they still support the Blackberry. You mean different things in different places. Back to my distinction of being a master debater, one way to get systematic with meaning is by building arguments. Some of the arguments are already made. The argument for what meaning and structure are going to do for this business made itself, and it wasn't mainly thought of as a design decision. Here's another instance of something closer to what we work on though where we have to get systematic with our meaning, and do that like an argument is what I'm saying. Sometimes the arguments are already made. If you've ever had a loved one in prison and they're relying on you for basics, for survival, the UI that Access Securepak presents to you...guess what? You can't really look at the picture of the Snickers bar and examine whether or not this would be a tasty snack. The argument for how to do this UI for this structure is made by the fact that there's no competition, and most of the people who are doing the interactions here don't want to be. This is a reminder of suffering, so make this as fast as possible to order stuff, and that's what they do. What is good structure here, while some of the arguments for what would be good are already made just by the context of no competition, and we're trying to help prisoners? Vanguard has a bunch of really nuanced arguments for how its business is presented online. They are old school with a splash page to help you navigate. If you are an investment adviser, what they present to you is very different than what they present to you, if you are an institutional investor. That is also different than what you do for these other ancillary audiences. Sometimes, the arguments are made and you just run with them. Other times, the company is already making these really articulate, interesting arguments that you can connect with their intent to figure out what a good structure would be, based on the arguments for why these business segments even exist, why is institutional investing different than the other ones. Systematic meaning, getting systematic with it. This is hard across channels. Really great clients have really great arguments for why anything would be the way that it is. The World Wildlife Fund has a fascinating strategy for mobile here. On the desktop and on the tablet, the experience will unfold in terms of things that you are probably interested in. If you're here, you probably care about animals and you probably care about places where you live or where people you love live. The experience unfold left to right, species, places, et cetera. Then on the mobile, let's make sure that the two most important things we want you to do, that's what you get. The rest of it is hidden in a hamburger, of course. We didn't even have to look at the UIs for this. We could talk about, "What is our argument for what we want the mobile user to do? If there are two things that they could do before we lose them, let's have that be an argument, let's get systematic with what we mean." Then once you're systematic with what you mean, you need a language to talk about that amongst yourselves and with clients also. This is Denise Scott Brown. She's a hero of mine. She's a South African urban planner. Think of her as the Dana Chisnell of urban planning. She has a powerful partner that she does really important work with, Robert Venturi. Together, they came up with a language to talk about structure. They were studying in Las Vegas. Really radical at the time to say that Las Vegas could teach us about how to make structures be good and what they decided, and she invented this framework. Here's the, as promised, ducks in sheds. The structural language to describe a structure that will be so tightly bound to the purpose you have for it that it will be unmistakable for that, that idea of the structure could be helping you with what you mean. How about a duck-shaped building selling freshly killed fowl. What's a capon? A bird probably though, right? This structure tells me what to expect. God forbid, I want sell Frisbees out of there. When structure and meaning are so tightly linked, you can't uncouple them. That's a duck. The ducks companion, a more modest companion, is what they call a decorated shed, that that building could sell duck eggs or Frisbees or food. The meaning is tacked onto the outside. In working with the built environment, if you stack up the attributes of what is different between a duck and the decorated shed, what their recommendation is, because building...This building was really solidly built that they can move it, intact. Jerry told me that they didn't even take the chandeliers out. The purpose of this building as a theater, could we think of this as duck. Then an undifferentiated space, that would be maybe akin to a shed. One example that's written large here in Detroit that you can just look at out the window when you're driving around, and in the built environment, this is fun because buildings last longer than apps and websites. The Fisher Building started out as a duck. Today, 100 or so years since its building...That was supposed to just be the end cap. There was supposed to be another Fisher Building, exact same size, in between them, a huge thing as tall as the Empire State Building was the original design. They only built this one, which is really impressive. At the time, General Motors operated out of here. It was kind of like a duck. This list of things original, consistent, advanced technology for its time. Today, and they haven't since moved, if you're not from here, you don't follow GM. GM used to be headquartered here. Now, they're headquartered here. How is structure part of how they communicate? By moving from a decorated shed to a duck, they were saying some things. They may not have done the right things but they were saying some things. An example of a duck in our world, has anybody seen the new Virgin America experience for ordering tickets for airplanes? This is a magnificent duck. This is innovative, this is heroic, this is pretty from every angle that you look at it. If the people working on this had the structural language to know that what they were doing is making a duck, it may have been more comfortable and enjoyable for the whole team. That's what they ended up with. Another example from the kind of work that we do that I think of it as a duck is Square. Does anybody here remember the first time that you pay for something using Square? Goosebumps from paying for something? Just on the phone and you...Oh, my God. The product is a duck, and the website is a duck. The website, there's no ornament here other than the product. It's revolutionary, innovative, original, new words, Square. All the language around this, they invented. Looks expensive and it's interesting. If you follow these matters like I do, you may have noticed that PayPal copies them all the time. If you look visually at these two different UIs, they look about the same, but this is a decorated shed. One of the ways that it's a decorated shed, this one really telling, it's only pretty in front. Couple of clicks into this it's old PayPal stuff. The flexibility of a decorated shed, we can try to compete with Square by looking like them but we don't have to change our business. A decorated shed is more flexible. Fourth thing, last thing, you need simple models to work this out. To work out whether we're making a duck or a decorated shed while we're making it, expensive and painful. Simple models and in architecture, the beauty of architectural models, how a loofah and a toothpick are a tree, you can work through ducks. You can show the client ducks and decorated sheds as crude models like this, and that's a really cost effective and fun way to figure out. We intend some things, we mean some things. What should the structure be? Let's work it out with really simple models. I think this is akin to what we typically do with models. [start of video] Mugatu: Without much further ado, I give you the Derek Zoolander Center for Kids Who Can't Read Good. [footsteps] Derek Zoolander: What is this? [background sounds] Derek: A center for ants? Mugatu: What? Derek: How can we be expected to teach children to learn how to read if they can't even fit inside the building? [laughter] Derek: I don't want to hear your excuses. The center has to be at least three times bigger than this. [laughter] Dan: A high fidelity prototype, the client didn't understand. [laughter] Dan: This happens. This is real. Simple models. I promised a case study out of this, so I have the briefest case study ever. I'm working with an e-commerce company. They present a website today. The whole website is what they sell, except there are a couple of things that you can get a custom quote from them on, but it's all blended inside of shopping. In order to help them to take their intent, which is to grow this business into rendering it somehow, we spent a lot of time in that middle step of meaning, what should it mean by working through really crude models? Because they can't see anything else. When your whole website is selling things, the idea that you might need to do something differently than that is hard to see. Rather than me having to wireframe the whole world and show them Old World New World, how about just the difference between these two models? What used to be two little things and a small process, what if we bifurcate the experience? This will be all that now. This will make more sense, it's all blue. This new world will be yellow, where we'll add two new big things. Then, in the middle yellow and blue make green. Really simple model, and at the end of the process of getting systematic with what we intend, with what we mean, having some language to talk about what we might do and what might be good before actually having to build everything, and then proving it out with simple models. That's how we do it. Thank you. [applause] Jared: Excellent. I have a question. What struck me about these ducks versus decorated shed things is that you're asking your stakeholders and your clients to get beyond their current emergent need that is driving the project. You're actually going to say, "Look. We can do this thing you're asking for right now, and it will be fine the way you're asking for it, but there's a future that you want to be concerned about." I'm curious what you think some of the challenges are with talking to stakeholders and clients about imagining their future world versus today's world and add implication. Dan: That's an important question because it's hard. The reason they're already paying you to be there is because something's wrong. Another one of Mr. Wurman's uncomfortable observations that I've found to be increasingly true is that without a language like ducks and sheds, without some other language, the language that people default to is based on what they already know. When you're working with stakeholders, something's wrong, they want a future that's better and more successful. Unless you give them something else to talk about i in terms of, they will be talking about it in terms of "how." They'll be talking about it in terms of the app, the website, the AJAX, the hover, and the responsive. Those are all how. That's because that's what they understand. What they're buying ultimately is, "How will we solve these problems?" Our challenge with this stuff is we're not saying you don't get a how. We're saying, "Could we please set that aside because if we are talking too early..." Once I've got their intent, I could jump right to design, but I don't want to. I want to wait and do some work with that, making what they mean systematic. We're just building this case. You probably remember this as well as anybody in the room. When information architecture first was a thing being sold to business customers, the metaphor of how this is like buildings... Fast-forward to today. Where did the Arab Spring happen? Where did an event in human culture that will be referenced for millennia happened? It happened on Facebook. When you go to an event with your kids and all the parents are on their phones while their kids are supposedly performing, where are they? They're in a place. Another way that we do this, Jared, is by asking them and, as our meetings get fancier and fancier, the stakeholders have more and more first-hand experience with architecture and the built environment. Maybe they've had the chance to build their dream home or at least customize a home. We just say, "Hey, before you spend a couple of million dollars on a building, because the building is going to be there for a while and it's supposed to be really valuable, how about if we work through some simple models first before starting to build the building?" They're like, "Oh, yes. I get it." Jared: The Arab Spring is an interesting thing because, in order for the folks who were part of the insurgent or revolutionary forces, as part of that, to disguise what they were doing, they were finding all sorts of interesting ways to communicate. I'd read in "The New York Times" that they were using dating sites because nobody was monitoring the conversations on dating sites. Dan: They were going to places where they could make things happen. If you were to say, "This app I'm working on..." Did anybody here leave some stuff on the bench at home that you have to go work on tomorrow, that you're going to have two days' worth of work to catch up on? That thing that is on your bench that you'll go back to tomorrow, if you think of that as a place, it's a place made of information that people go. Just like this building is wonderful for what we're doing today and it would suck for some other for some other purposes, just thinking that framing of, "This is a place made of information," does it work well as a place? Audience Member: Hi, Dan. Dan: This is just like old times. Audience Member: I was Dan's student. My question about your forceps, knowing you have a good structure, is where does user research come into this? Oftentimes, I'm also told that continuous research and usability testing is a good idea to do as you create a project. Where does user research fit into this system? Dan: I drew a doodle the other day of a dream world. I will use this for a shameless plug. The Understanding Group is hiring a Design Research Lead. In our perfect world, we would do research before and after each step of doing of things. When you framed it as, "Where is user research," for us it's stakeholder research to get the systematic intentions. The systematic meaning part is a combination of contextual research and, maybe even as you said, usability also. We don't have product sheets. We're not testing them from a usability standpoint, unless maybe it's a taxonomy that we're testing the sensical or nonsensical nature of the labels. The short answer is, all along the way. The simple loofah model with the sticks and the things can be tested, just like you could test the Derek Zoolander School for Kids Who Can't Read Good. The short answer is, as much as possible, as much as you can sell in to the process. Jared: Would it be fair to say that user research is really just about informing the process at every step, and, therefore, the nature of the research is going to change depending on what the team needs to be informed about? Dan: Back to your marvelous definition of design, if it's the rendering of intent, the rendering is getting increasingly vivid, increasingly realistic. The checking's with intention. The other thing that happens in this process, intention is here, all along the way. Here, it's starting with the deciders and the people who're paying for it. As we move through it, then it's also the intention of the users. That design being the rendering of intent, it's not just the business owners' intent. Unless you're at Liberty Tax Service and you just move into a Pizza Hut, and you rely on the guy on the Statue of Liberty outfit to get the customers in. Some of you have seen this. [laughter] Dan: Objections? Come on! Jared: I'm not sure it's worthy of objection. It's good! It's good stuff! The list of attributes that you showed for ducks and sheds, are these good talking points with a client? Is this a place you can bring out in a kickoff meeting and ask what's going on? Does this have some bias to it that will make people always want to be a duck? Dan: Yes, and yes! I love your questions. It's good because yes, and not all clients. With some clients we're already doing this. We're trying to extract what the adjectives are for what would describe a good outcome from what we're working on. We're already doing this. If the clients can go there with you, if they would entertain some big words in some weird process, you could work with them through this, and then look at pictures of buildings that they all know, from the neighborhood, and then talk about it. In the built environment it's so expensive and buildings last so long, or at least they used to, that figuring this out is mostly a matter of economics. Increasingly, because we've been wrestling with this at TUG, how does this from the built world really translate to software and apps, and things? What's not true about this analogy in the built environment and where we work is that it's not necessarily more expensive to build a duck from a cost of the developers, from the development time. Experience debt from building a duck is akin to the kind of material lock-in that you have from pursuing a duck in the built environment. Every stakeholder, if they have the chance between an undifferentiated structure, where what we mean is tacked on...imagine buying a Magento eCommerce site, throwing your products on it, putting your logo there. That's a kind of a decorated shed. Of course, we want a bespoke thing, like skinny ties. Show that to any stakeholder, "This unique thing that you do, we could make a site that is really about this, down to the skinny tautness of it." Of course, they want a duck, and then you can talk about, "Well, once you have a duck, you're stuck with the duck, and a duck can't sell Frisbees." Jared: Your real estate analogy of ducks and sheds somewhat falls apart, in that no one wants to buy a duck or shed. The complicated fact is sometimes you get a client who's like...they're using the new words, "Put it in the cloud and make it responsive," and they have no clue what that means. Then you have an agile developer who's like MVP, which means essentially they're building a shed in iteration one. No matter what, you're getting a shed. Can you talk a little bit about when using this type of terminology, the management of perception and perspective for the client? Maybe you're trying to talk them back from the duck, but you know they're going to see the shed first, and they've been promised somewhere in the middle...that sounds really complicated. Dan: There is a lot of complexity in what you asked. I'll hone in one little tine part because I don't like hard things and other things were hard. [laughter] Dan: This is easier, which is...I was trying to be funny and I've lost it! [laughter] Dan: I'm eating into the next presenter's time. Afterwards, let's re-ask this. Thank you so much for your time. [applause]