How can interactions with embedded systems engage users in Broaden and Build responses?

The concept of interaction aesthetics (Lim et al. 2007) (Djajadiningrat et al. 2000) (Gaver & Dunne 1999)  has played a pivotal role in redefining the paradigmatic concerns in HCI. Historically, HCI began with the goal of human-machine coupling in high performance environments such as fighter jet cockpits (Sengers et al. 2007). With the invention of the computer, HCI shifted towards optimizing efficiency in task completion (Sengers et al. 2007) (Stivers 2004). Over the past 15 years, this paradigm has shifted towards understanding how to produce computationally supported experiences that engage users through embodied interactions, defined as interactions that situate users and computation to construct meaning that leads to new understanding of one’s self and the world (Sengers et al. 2007).

The role of interaction aesthetics within this paradigm is to critically evaluate the various forms/shapes interactions can take and how those forms/shapes effect user experience (Lim et al. 2007). In 2007, Lim et al. introduced the notion of an interaction as a gestalt. The notion of a gestalt implies a collective of elements used to construct the whole. Previous interaction aesthetic studies have explored multimodal engagement (Nguyen & Masthoff 2009) (Brotman et al. 2008) (Sinha & Landay 2002) and clarity of interaction as such elements (Brotman et al. 2008) (Gaver et al. 2003) for designing user engagement. Design of multimodal engagement determines what senses (visual, audio, haptic, etc.) a system generates feedback loops with. Clarity of the interaction refers to the degree by which the system explicitly reveals the method and the purpose of the interaction. This study will continue to use these two areas of interest, treating them as interaction variables capable of changing the interaction gestalt.

The purpose of deconstructing these interaction elements into testable variables is to understand their potential for generating Broaden and Build (Fredrickson & Losada 2005) (Fredrickson 2001) cognitive responses. Fredrickson proposed and developed evidence of the Broaden and Build theory as an explanation of the evolutionary relevance of positive affect (2001). The theory argues that while experiencing positive affect, a person’s ability to generate novel possibilities for future actions increases (the broadening) and over time, as a person makes choices based on these possibilities, a person increases the diversity of their skill sets (the building). Since the theory’s introduction, researchers have conducted qualitative (Fitzpatrick & Stalikas 2008) and quantitative (Fredrickson & Losada 2005) studies that suggest it enables human flourishing (Fitzpatrick & Stalikas 2008) (Little et al. 2007) (Fredrickson & Losada 2005). Positive psychologists argue that human flourishing determines psychological well-being through measurement of a peron’s abilitity to seek out and solve problems (Fitzpatrick & Stalikas 2008) (Little et al. 2007) (Fredrickson & Losada 2005). Hence, the significance of understanding how to design interaction aesthetics is two fold:

• Designers can make embedded system interaction more compelling by eliciting positive affective responses and;
• Designers can use interaction as a tool to improve the well-being of end users through the design of interaction aesthetics that engage users in Broaden and Build experiences.

This study will contribute to the domains of:

• HCI design by adding to the body of literature on interaction aesthetics through introduction of guidelines for engaging users in interactions that foster human flourishing;
• Positive psychology by revealing if interaction can be treated as an independent variable capable of eliciting Broaden and Build responses and;
• Embedded systems through the use of the Game as Life – Life as Game (GaLLaG) (Burleson et al. 2009) embedded system as both a technology platform capable of delivering compelling user experiences and a research tool for furthering ubicomp knowledge. Figure 1 presents a conceptual framework of this study.

This study tests several hypotheses implicit within the conceptual framework. These are:

• Interaction, much like other artistic mediums, can be deconstructed into base forms or primitives to form guidelines for designing embedded system interactions.
• Interaction, not only facilitates delivery of information, but is information in itself with the capability to sway the affective responses of users.
• Designers can sculpt embedded systems interactions that promote Broaden and Build experiences, increasing human flourishing and improving the well-being of users.

Bibliography
Brotman, Ryan, Kelliher, Aisling, and Spicer, Ryan. 2008. Well how would you do it? – Facilitating the transfer of knowledge in collaborative design environments. Paper presented at the annual national conference for the Industrial Design Society of American, September 10-13 in Phoenix, Arizona.
Burleson, Winslow, Ruffenach, Collin, Jensen, Camilla, Bandaru, Uday, and Muldner, Kasia. 2009. Game as life – life as game. Paper presented at the 8th annual international conference of Interaction Design and Children for the Association of Computer Machinists, June 3-5 in Como, Italy
Djajadiningrat, J. P., Gaver, William., and Fres, J. W. 2000. Interaction relabelling and extreme characters: methods for exploring aesthetic interactions. Paper Presented at the 3rd annual international conference of Designing Interactive Systems for the Association of Computing Machinists, August 17-19 in Brooklyn New York.
Fitzpatrick, Marilyn, and Stalikas, Anastassios. 2008. Positive emotions as generators of therapeutic change. Psychological Integration. 18: 137-54.
Fredrickson, Barbara, and Losada, Marcial. 2005. Positive affect and the complex dynamics of human flourishing. American Psychology. 60: 678-86.
Fredrickson, Barbara. 2001. The role of positive emotions in positive psychology: The broaden and-build theory of positive emotion. American Psychology. 56: 218-26.
Gaver, William, Beaver, Jacob, and Benford, Steve. 2003. Ambiguity as a resource for design. Paper presented at the annual international conference of the SIGCHI Human Factors and Computing Systems for the Association of Computer Machinists, April 5-10 in Fort Lauderdale, Florida
Gaver, William, Dunne, Anthony., AND Pacenti, Elena. 1999. Design: Cultural probes. Interactions. 6: 21-29
Lim, Youn-kyung, Stolterman, Erik, Jung, Heekyoung, and Donaldson, Justin. 2007. Interaction gestalt and the design of aesthetic interactions. Paper presented at the annual international conference of Designing Pleasurable Products and Interfaces for the Association of Computer Machinists, August 22-25 in Helsinki, Finland
Little, Brian. 2007. Personal Project Pursuit: Goals, Actions, and Human Flourishing. Philadelphia: Lawrence Erlbaum Associates.
Nguyen, Hien, and Masthoff, Judith. 2009. Designing empathic computers: The effect of multimodal empathic feedback using animated agents. Paper presented at the 4th annual international conference of Persuasive Computing for the Association of Computer Machinists, April 26-29 in Claremont, California.
Sengers, Phoebe, Harrison, Steve, and Tatar, Deborah. 2007. The three paradigms of HCI. Paper presented at the 25th annual international conference of Computer Human Interactions for the Association of Computer Machinists, April 28-3 in San Jose, California.
Sinha, Anoop, and Landay, James. 2002. Embarking on multimodal interface design. Paper presented at the 4th annual international conference of Multimodal Interfaces for the Institute of Electrical and Electronics Engineers, October 14-16 in Pittsburgh, Pennsylvania.
Stivers, Richard. 2004. Shades of loneliness: Pathologies of a Technological Society. Oxford: Rowman & Littlefield Publishers, Inc.

By cool, I do not necessarily mean the latest Phillipe Starck table lamp, or the 3DO video game.  I’m more concerned with things like the wheel, or McDonald’s restaurants, of Federal Express of Good Grips kitchen tools – things that are so right that they become nearly invisible, part of our daily lives in a way that a product like the Print Vac never could.  It has turn signals, an odometer, and most impressively, a tiny printer that prints out everything you’ve sucked up.  The Print Vac is just one extreme example, (admittedly mythical) of a kind of design where something gets made just because it is possible.

It is what we call “technology centered design.”  Technology centered design has several distinguishing characteristics:  it is engineering-led, the capabilities are developed before the use is defined, and the products which result can only use the ever expanding range of features as a means of differentiation in the market.

There have been things as downright weird as the Print Vac over the years, such as a robotic shaving machine.  But more centrally, it results in things that we are all expected to use, to understand, like incredibly complex TV/VCR remotes.  They, and thousands of consumer products like them, rely on features to differentiate themselves, and quite obviously, in a case like this, the features begin to overwhelm the use of the things themselves.

At a near diametric opposition to technology centered design are things like a lovely console television finished in genuine simulated wood grain, an example of what might be called, “market driven design.”

Market driven design is what happens when the development process too slavishly responds to the opinions of everyone and their aunt as to what they like, what they want.  Its evils are many, but the results are knockoffs and “me toos” and things that just are not as innovative, as interesting, or as good as they could be.

The alternative to market or technology-centered design is an approach with the slightly misleading label “user (or human) centered design.”  The misleading part comes from the fact that, in the best examples of the work, the focus is no only on people, but on how people and things are connected, how people come to use and to understand what things to.

Understanding that connection, defining the space in between people and technology is not as easy, actually, as being either market-or techno- centric, but in the long run; it’s a far more valuable effort.

Three ways of getting there
Be a genius.  If you could be – or hire – someone like Thomas Edison, who registered more than 1000 patents in his lifetime, you’d be pretty sure of making that connection from time to time.  This is not an entirely flip assertion.  Great, user-centered design inventions that fundamentally connect what things can do to what people need have been around for nearly as long as there has been some form of technology.  But for the longest time, an inspired connection was almost always the work of an individual like Edison.  Edison invented the incandescent light bulb in 1879.  He didn’t invent the parts, he didn’t discover electricity, but he looked at what was possible in the technologies he was fooling with and connected it to a need.

Be very intuitive. This connection can be, and often is, made by people who are decidedly not geniuses.  Deep intuitive insight is another time honored, but still rare, connector of people to things.  On the one hand, it is evident if fads, like the Rubik’s Cube or the Hula Hoop.  Rubik was no genius, but the cube certainly tapped into something out there.

The other hand in this case are more profound examples of intuitive approaches that more self-consciously sought to make the connection, to meet or understand a need.  High on that list would be something like Paul Rand’s 1961 UPS logo.  As the probably apocryphal story goes, this nearly timeless piece of work was the result of an half hour meeting with the UPS president and Rand’s daughter acting as test audience:  “It’s a package Daddy!”

Have a method.  Take something like the nuclear powered missile submarine, one of the more awesome technological achievements of our day – clearly, not developed by relining on either genius or intuition.  What made the first nuclear subs possible was, in large part, Admiral Rickover’s invention of the PERT chart, which coordinated and organized thousands of tasks, problems, and schedules over years of time.

Method has come to be a scary word.  To many, it sounds controlling, antithetical to design.  Methods do not have to come with manuals, diskettes, and years of training to offer powerful and reliable ways to make the connection between people and things.  Some of the best are exceptionally simple in practice, even though the might have taken years to develop.

For example, McKinsey & Co.’s justly famous “seven S” model of what makes an organization tick is as much of a method as it is a description.   It gives McKinsey consultants a way to look at the problem, a way to work through what they find out there at be sure they’ve examined, investigated, all the right pieces.  On a different but equally useful scale are things like notation languages and representational conventions.  These provided a common ground for groups of people working on complex issues.

Methods like these don’t automatically generate answers – which is too often the fairy tale claim of bad methods – but the y do something more important:  they structure the problem and provide ways of structuring the information that can lead to a solution.  Unlike management consulting or programming or nuclear engineering, design as a discipline is too often seen as without methods, and worse, without the need for them.  User-centered design is not just about paying attention to “needs” or “human factors,” but about having a method for making that connection between people and things.

A method must have focus and definition.  The methods we’ve mentioned above are not defined by their steps, but by what they take as their focus and by how they define the nature of that connection.  Rickover understood that the key to the problem of building a nuclear sub was convergence and coordination, not solving the individual problems – the focus.

In user-centered design, the focus has to be on the connection between users and things, not on one or the other.   The definition of the connection is probably the trickiest part of all this.  Most people think that need is the best way to describe it.  (I really need sneakers that light up when I walk.) I define this connection somewhat differently.  The connection that is most important for design is not need, but how people use things to make meaning in their lives.  People interpret the world and their place in it through the things that they use.  To ourselves and to one another, what we use to make sense of our lives grows out of the material conditions of our lives.

Making meaning

A five thousand year old guy was found frozen in a melting glacier in the Alps in late summer 1991.  Forensic anthropologists were able to reconstruct approximately what he looked like and build a reasonably lifelike model.  Fascinating, in a morbid sort of way, but what does that really tell us about how he lived, about what his life was like?  It wasn’t until archaeologists excavated the whole site and found his tools, his clothes, his equipment, that they were able to tell a more complete story about him.

Arrowheads and dagger points are one thing.  Beepers and other paraphernalia of modern culture do not tolerate the simple equation of meaning with function.  What they “are” depends on a whole lot more than what they do.  Beepers, for example, first emerged into the culture as an exceptionally handy gadget for physicians.  Not too long after that, in urban neighborhoods across America, there seemed to be a sudden explosion of 13 to 17 year old doctors hanging out at street corners and near public telephones.  And finally, we are used to them, are, as a culture, accepting and extending them.  The pager is being transformed again into a basic part of the personal communications infrastructure.

The same artifact, but with different meanings, based in different uses; conversely, as the uses expanded, it meant different things to the people who saw it, who thought about using it.

Where this all comes down is in the choices people make about what they buy.  Two artifacts as simple as two flashlights offer profoundly different options for what someone wants to say about what is important to them.  One (red and white plastic) costs less than two dollars at the corner drugstore, the other (blue anodized aluminum) costs about seven and can be found in upscale catalogs like Sharper Image and Hammacher Schlemmer.  A purely functional analysis might say that the Maglite is “better” because it has more features, is more durable, and comes in a wider range of colors and sizes.  Yet for some people, the cheap one tells a better story, a story more amenable to their own values, their own sense of what is important about a flashlight.  There is not, obviously, a single continuum along which one of these is better, one worse.

Making it work

How do you figure out how to make things that people will want to use, that people will value as tools for constructing their lives?

This is the basic framework we use for doing it.  Basically, it is a series of models and concepts that, in a progression from the concrete to the abstract and back again, are tools for understanding the connection between people and things.  It has two major components: the first is about moving from what’s out there up to structures that make senses of it.  The second, which is moving from structures to new things, we’ll go over further on in this paper.  We rely on our clients to tell us what they can do, what the nature of the things we are dealing with is.  The real challenge, especially once you move away from needs, is understanding the people, because what you are looking for is not what people say, but what they do, what they use to understand the world.

Anthropology is the discipline most concerned with doing that, with connecting culture and people and artifacts into a system.  Margaret Mead, and the other pioneering ethnographers had two main tools – a notebook and an “informant.”  Classical anthropology was built on these tools.

Just as beepers are more complex than arrowheads, our tools have gotten a bit more complex.  Our fieldwork comes through the rather new discipline of “video ethnography.”  Once we decide where the best place to watch people using things is, we take a bunch of cameras, a bunch of computer controlled video decks, miles of cables, microphones, and small gray computers with little colored logos out into the real world.  For hours at a time, for days and days, we videotape people doing the things they do, without interference, where and when the usually do them.

When we get all these hours of videotape back to the office, we use a set of computer programs we’ve developed to help us log, analyze, and structure the material.  What we are looking for are what we call ”particulars.”  This is a very “thick” description of what is happening in the places where our client’s products are being used.  In the same way that McKinsey “7S” model helps them, we’ve developed a model with the catchy acronym AEIOU, which stands for Activities, Environments, Interactions, Objects and Users.

If you break your description of just about anything down into activities, environments, interactions, objects, and users, you can be fairly sure that you’ve hit all the particulars.

Activities are goal-directed sets of actions, things that people have to or want to accomplish, like filling up their car with gasoline.

Environments are where activities take place.  These need to be considered in fairly broad terms: we also need to keep the activities in mind as part of the definition of the environment.  A convenience store is self-contained, but an airport control room, where all of the radio connections between pilots and planes and FAA centers effectively extends the environment beyond the room where the controllers sit.

Interactions are, unlike activities, always dyadic, always between a person and someone or something else.  Interactions are the building blocks of activities.

Objects are the building blocks of environments, and key players in the activities and interactions.  What is interesting is that you have to see things in use to really describe them.

Finally, there are the Users themselves:  Who comes there?   Who does what ? How do they act?

Add all of this detail up, and what do you have?  Nothing, until you begin to look for patterns, for the regular connections between users, activities, or interaction and the environment and objects.

Take the example of a gift shop.  A person takes a look into the aisle, but doesn’t really move into it.  We came to call this behavior the “end of aisle lean.”  I’m sure you’ve seen it, and probably engaged in it without too much thought about it.  But it is interesting where it led us.  Our client had asked us to think about re-organizing their stores, to give some thought to how they should word their signs, where they should place them.  What this pattern revealed to us was that thinking about the words on the signs was the completely wrong idea for how people find stuff in this environment.  Our clients had assumed that people came in with a specific item in mind and needed help finding it, and that they read the signs to do that.  Instead, what we realized they do is come in with an idea of the kind of thing they want, and that that is defined, for customers, in terms of the visual language of greeting cards.  So what these people are doing looking into the aisle, scanning not the signs, but the look of the displays.  You look, you see a bunch of pastel colors and gig gold foil letters and you don’t need to read the categories to know that this just isn’t for your 15-year-old nephew.  What patterns eventually reveal, once you begin to figure out where they come from, are the most important, most abstract level of thinking about how people interact with things.  We the things that lead to patterns structures.  Now, there are lots of ways to characterize structures – anything from rational choice economics to unresolved Oedipal conflicts – but within our approach, the most valuable and generally applicable one is the notion of frameworks.

The idea of a framework is one that has roots in both literary theory and cognitive psychology.  But the idea can be illustrated fairly simply.  Take a billion or so stars visible in the night sky as the raw material of a perception.  A framework is the set of biases, assumptions, and knowledge that influences what you see when you look at them.  A framework is what explains, or at least describes, the difference between how Galileo looked at those stars, and how Vincent Van Gogh saw them.  Objectively, the same sky, but seen through different frames, not the same sky at all.

Frameworks are not simple, nor monolithic, nor even stable.  They are quite complex, permeable, and dynamic.  What is most important to know about the concept though is that there are three distinct kinds of influence on them, three ways in which the world (and its artifacts) affect how people understand things: individual, social and cultural.  Take a simple (albeit fairly cool) object like a pair of shoes.

My friend Julie bought a certain pair of shoes because she likes them.  They are comfortable and they fit her style and image of herself.  They make a nice alternative to her green Doc Marten boots.  Those kinds of reasons, those kinds of choices, result from an individual level of a framework.

But the choice of a shoe is also constrained, influenced by social factors, but what particular kinds, brands, styles and colors mean to the people with whom the wearer associates (works, hangs out, wants to impress, and so on).  In New York, in Chicago, in LA, shoes have an important role in gang life.  They aren’t just personal choices, they are emblems, currency in a language of identity and membership.  Wearing Nikes (or certain kinds of Nikes) means you are part of one group and decidedly not of another.

Finally, that shoes can mean anything at all is a cultural matter.  “Clothes make the man” makes sense to us.  It doesn’t to a lot of people in the world.  And certainly, that there is a meaningful difference between Thom McCanns and Doc Martens would be ludicrous notion to the rather large portion of the world for whom shoes are either not an issue or purely utilitarian.

Making it real

The second (downslope) side of the “user centered” process is the move from describing the world to building, or perhaps re-building, the world.  You’ve gone out, described the particulars in mind-numbing detail, noticed, identified, and cataloged the patterns, and made sense of those patterns by figuring out how people’s frameworks make the patterns happen.

This is where the fun starts.  When you know the structures, you can manipulate them.  You can use them to offer alternatives to people.  The world you’ve described is built on what is already out there, but the world you can build can be abased on the new technologies, new capabilities, new things that have been sitting waiting on the sidelines while the people of the equation has been being filled in.

Here’s the difference in a quaint midwestern homily:  My mother was a classic late 1950s, early 1960s cook.  Betty Crocker was her goddess.  She cooked step by step, with precise measurements, times, and temperatures.  But every once in a while she would call up my great grandmother and say, “Gram, how do you make those biscuits?”  And my great grandmother would respond with something on the order of, “well, you take some flour and soda and salt, add enough water, mix it until it is the right consistency and then bake them until they are done.”  This drove my mom nuts.  But my great grandmother understood the structure of baking.  And she could make anything – even things that had never been made before – because she was working with what happened to be in the kitchen that day.  This is what great design can do once it has a great structure to work with, once it knows how things affect each other.  Maybe not as fattening as Grandma Ethel’s spice cake with penuche icing, but just as tasty.

The move from these structures out to real things isn’t my forte.  That’s why I work designers, architects and programmers.  But let me walk, very quickly, through the steps that come out of structures:  conceptual models, designs for things, and prototypes of those things.

Going back to the gas station, one of the structures we developed was about the way in which people oriented themselves, how they found the places they needed to find.  We also found that there was a consistent mismatch between what the attendants used to locate people and cars and the framework that customers used to locate themselves and the other important things on the lot.  This conceptual model was one of the directions that came out of that.  It developed the notion theta there needed to be a common point of reference and easy visual access between the attendant and the customer.

One of our product designers developed this as a design direction in a document that we gave to the architects we were working with on the project.  It is not a spec drawing in any senses, but it did clearly turn the model we had sketched out into something the architects could work with.  Which they did.  Not all of the design concepts are as robust as this one was, but they all, because they are so well grounded, are as clear and as useful to the people who make things as this one was to the architects on the project.

To reiterate the process:  start with describing the world that is out there, use methods to get to useful abstractions about it, and then use those abstractions to put things back into the world.  Then start all over again, evaluating and watching the results.

What cool things do

Because we use things to make sense of the world, when you put something out there that is different, you change the way that people think about other things.  What you change, actually, are the frameworks.  And if you do this well, and do it in directions that are strategic, you begin to control the rules of the game and leave your competitors in the dust, not because you’ve got more features than they do, but because you have redefined what is valuable about the thing and they are still playing by the old rules.

Frameworks change as the material of which they are made changes.  And those changes can be powerful and long lasting.  For example:  helicopters were one of the key symbols of the postwar Red Cross, symbolizing the marriage of technology and humanism.  But the development of the Cobra gunship and its widely televised use in Vietnam changed all that, changed what they could mean to people, changed what their primary meaning was for almost everyone.  And helicopters – no matter how often they are used to deliver aid or rescue children – will never really be a good symbol for the Red Cross again.

What does it take to do this work?  Four things:  data, patience, teams and iterations.  And some cameras.  But these are the more important elements.

Data You need lots of data to look for patterns.  Margaret Mead spent more than three years in Samoa.  You’ve got to be able to sample widely, got to have situations to look at when you’ve got a hunch about a pattern that you are chasing down.  And although we think it works best, data doesn’t have to be videotape; it can be interviews, photos, notes, whatever.

Patience One of the things that Freud strove to teach the early psychoanalysts was what he called “evenly hovering attention.”  What he meant by this was that it is important not to discount anything.  That you can’t really listen if you’ve got too many hypotheses going into it.  This is probably more important than having lots of videotape.  You’ve got to let the patterns emerge.  You can’t force, can’t hurry it.  We sometimes spent hours going over 30 seconds of tape, and weeks thinking that there was nothing there to discover.  But there always is.

Interdisciplinary teams We could not do this in an office full of anthropologists.  It would be even worse in an office full of psychoanalysts.  Designers and architects and clients all participate in the fieldwork, in the analysis.  Different people – with different frameworks – notice different things.  Different disciplines bring different sets of skills to the party.  We also use teams to supplement the basic fieldwork by bouncing those results off of those generated by other methods and other kinds of data.

Iterations As much as I’ve tried to make this all sound like it moves smoothly from one stage to the next, it doesn’t.  The first set of patterns always turns out to incomplete;  the first set of structures never can explain all the patterns.  You’ve got to be willing to throw out those efforts when they are not coherent or far reaching enough to be useful.  Moreover, the concepts, designs and prototypes themselves have to be part of an iterative process, working out bugs and kinks.

Finally, I want to be clear that this approach is not a panacea.  It is not an approach that will solve any and every problem.  It is particularly well suited for issues in product design, interface/interaction design, and environments – in short, whenever you can watch someone doing or using something that you are interested in rethinking.  We’ve also used this approach successfully in less obviously suitable situations, such as understanding and redefining processes that work groups use, or figuring out how an as yet uninvented product might work.

Conclusion

Design, broadly defined, is fundamentally a part not only of everyday life, but also of the dynamics that make our world constantly changing, constantly evolving.  What I’ve been outlining here is not a method for designing things, but a methodical approach to understanding that world, those dynamics, so that design can play a more important, more profound role in it.  WE can talk all we want about “profound change” and “understanding users,” but it requires something more, some really good tools to thing with, to develop a design –centered approach to doing that.  We believe that these ideas, these methods, are a big part of getting there.

It doesn’t take a lot of research to understand that design is a powerful tool in contemporary business thinking. Nor is it difficult to notice that design is increasingly central to an escalating number of parts of everyday life: more kinds of companies in more and more varied sorts of endeavors are using design, design processes, design partners, and design thinking in their work. And because design work has, in most of those applications, taken deep and considered understanding of the people who are going to use, inhabit, or experience what design makes, we tend to think that a broadening portfolio for design is a good thing. To most, it is nearly self-evident that something designed is better than something that is not. Saying that something is ‘well-designed’ is a compliment for anything from a kitchen tool to a strategic plan. We use the term “good design” almost reflexively as evaluation and often, as explanation. When something is tabbed as “good design” (by a magazine or an award from a professional association) we agree or not in the instance; but we accept, usually tacitly, that good design is different from indifferent design or bad design.

Despite all that, we don’t usually think of ‘good’ in its moral or ethical sense, as in “good vs. evil,” as the sum of choices made against choices not made. Positive, thoughtful, or responsible design has included in it not just a final form, but a large number of distinctions understood and choices about those distinctions made in a particular light, with particular reasons for them. A child’s toy, perhaps, can be safe or unsafe, educational or mindless, responsibly made or using the worst but cheapest materials available. It isn’t a toy made to be bad, but in each of the moments where choice matters, the choice made has been, in the current context at least, toward something less than as good as it could have been.

In this paper we’ll argue that this particular sense of ‘good’ in good design matters not only in the end product, but in the way design works in general, in design processes, in design thinking. We’re not arguing for any one moral or ethical position, but simply that we, collectively as a discipline, as a field, have a responsibility to acknowledge the role that values play in the work that we do. By knowing that that articulation needs to be done, and by using it to anchor the forward end of the design process, we do better work; we do good work. And with that, we can speak more clearly to other interests (customers, users, management, investors) about what ‘good design’ means.

Good Design and Design Research
When we take hold of a powerful tool and use it to shape the daily lives of real people, we are laid under an obligation, a responsibility, to understand not only how that shaping could affect those daily lives, but how it should do so. The “good” in “good design” has, in the last twenty years or so, migrated from the relatively simple appreciation of an end-product’s formal properties to include the ways in which a product becomes what it is: the process of designing. In the course of that migration, “users” and “experience” have become central to the way design works, to how the things which it produces are evaluated. Under any number of labels (“user-centered design research”, “ethnographics,” “anthrojournalism” and so on) the (largely) social sciences-derived research which informs the work of design has grown into a small industry of its own. Taken as a whole, design research has resulted in a collective paying of more attention to people rather than less. That’s a ‘good’ in pretty much anyone’s book. But it is also, in practice, a bit like supposing that because an M.D. is doing rounds, looking into patients’ rooms and signing the charts, good medical care is being practiced. If designers have been less than explicit about the values that inform the choices they make, it seems that design research as a whole has been even less so. The most widely accepted ‘point’ of design research is to inform the work of design. To provide a basis from which the work of design, development, and strategy can proceed. It is a bit circular: we do research to inform the process of design, which requires that we understand the users. Circular or not, it would be just fine if what was required to “inform” design were no more than a scan of current conditions. A pH strip dipped in the pool. A thumb licked and held up in the breeze. But the best design work doesn’t need the thumb in the air; good designers or teams or practices are usually plugged in and working at the ragged front end anyway. What we need from research is more than description, and especially, more than a list of “needs,” explicit or implicit, met or unmet. Although description is an essential moment in the work of ethnography, relying solely on description, and the simplified-empirical position that we are only reporting “needs” is a kind of responsibility dodge which lets research work go forward without taking up the side of the implicit bargain that requires us to say why we are interested in knowing what we want to know. The work of design research too easily takes as justification that we are describing only how people behave, or the perennially-popular-ethno-marketing refrain “what they actually do instead of what they say.” We cling to the position that it is not our job to say what anyone ‘should’ do.

But research and design, especially when woven together, are loaded with hundreds of value-laden choices, and with motives (from our embeddings in business) that are impossible to avoid. When we look at the entire arc of design process, including the research work which informs it, we believe that we need to be clear about what we hope the should will be even while we are depicting the could of design possibility. Understanding the role that our own values play, how they intertwine with the understanding and aims of our subjects, is not a hobble on design work, it enriches it.

I’ve argued in earlier papers that in the developmental side of design research, one has to be comfortable with the idea that research does not provide definite answers to particular questions. And that thinking that the work of ethnographic research should ‘answer’ a question has led to the intellectual shortcut of using ‘unmet needs’ as a catch-all substitute for the much harder interpretive work that great design takes off from. (ref. for “Uncertain Answers” paper). In that work, I argued that instead of identifying needs, research should take as its object (following Clifford Geertz) the ‘thick description’ of how experience is organized for the user. And that the work of design can use that description (aka a “framework”) to understand what could be changed through a new product or interaction or experience – whatever kind of design work was on the other end of the research. I still think that’s true, but incomplete. The idea that the rich, descriptive summary of a “framework” is what design practice needs from research is a good one, but in effect, it has only shoved the issue of choice and direction a bit further off, onto the design itself.

For more years than I would like to admit, I used some version of this argument in almost every introductory talk or workshop presentation I gave. And in the not insubstantial portion of those where my friend and colleague Hugh Dubberly was present, he would patiently raise his hand and ask me some form of the question, “I think you mean what it ‘should’ be, not ‘could’ be, don’t you?” After all this time, I’ve finally realized just how right he is: “Could” vs. “should” is a simpler way of framing potential vs. commitment, chance vs. responsibility. And going back to our opening, when you recognize the power of a tool, values, commitment, and responsibility all become what we should do in the work we do. In this paper, we’re going to look at how we need to rethink the descriptive part of design research in light of values, in order to, put simply, do good.

Design & Research, simplified

Let’s start with a simple proposition: the work of design is getting from a set of conditions which exist at the beginning of the process (now), to a future set of conditions which include ‘now’ to ‘next.’ we are researching then, are the conditions under which those choices will lead to a consumer, user, making a choice themselves.

Here’s a better way to think about the point of design research: understanding frameworks, at a dynamic level, not just a snapshot

Slide Show Notes, story flow:
1. Let’s start with a simple proposition: In other words, taking the present set of ‘what is’ conditions and developing what ‘could be.’
2. There is an idealized, pure innovation view of design that more or less ignores the now in order to shout real loud about ‘next’!
3. Set against that, is an (also idealized) research-driven view embedded in market research which pretty simply extrapolates the now to the next.
4. The truth of good design work is somewhere in between,
5. The rise of human-centered design, design research, and ethnography gave us a more nuanced, more complex, understanding of the ‘now’
6. The main point of those approaches the idea that by understanding the individual, social and cultural influences on why people think the way that they do, why they use what they use, why they need what they need (or think that they do), we’ll be better able to design well.
7. Prior to this, the language of ‘good design’ had been largely confined to aspects of the ‘product’ (even if that product was a logo or typeface). Occasionally, we looked at the process. But with the advent of “user-centeredness” emerged the notion of a “framework” as something that could not only be described, but – especially through the work of interaction design – as something that could be designed.
8. “Experience” as an object of design further established the role of something that was personal to the users, but could also be described and altered through the work of design.
9. Design as imagining the ‘what if’ of a users framework.
10. Design research has lacked an object – a thing to be about. Theory is nice, but that’s not what we’re talking about. And “needs” is neither a useful design construct nor reliably ascertainable.
11. There is a parallel between the work of design and the notion of successive approximation in theory building.
12. Summarized in the abstract-concrete by now to next model
13. A nested set of ideas:
Frames (are how people think)
Models are descriptions of the way that the elements interact
Scenarios are articulations of possible changes to the conditions of the model.
14. Definition of ethnography in the design world: description and interpretation, toward an end, within constraints.
15. The implication that if you can make anything, you need to choose what to make, where to intervene, what to target.
16. Choice, and the ability to affect other folks’ choices, means values play an important role.
17. Two lines from here: the dynamic change story and the values story.
18. The dynamic change story: the difference between designing the object, and designing an object, its behavior, its environments, and where it will end up (the ball and the hole slides). Why it is a model, not a snapshot.
19. Borrow from physics the idea of components to understand dynamics, like force, direction, resistance, spin… control (add or remove)
20. How designing those control the ‘metaphorical’ trajectory
21. The values story — what sustainable, environmental issues, and design for social impact have done to change the landscape – what Knorr-Cetina calls the ‘fictional symbolic’ the set of things we all come to use, to agree to.
22. Lakoff on republicans. IHI ‘preventable deaths’
23. Redraw the original picture of frameworks with values as the ‘forward anchor’ of the design process – not form.
24. The bigger goal of design work is to clearly understand the values (as both initial condition and as endpoint), the ‘signposts’ (from the World Wildlife Fund paper) and the ‘choice architecture’ involved in getting from the current set of forces to a new frame, experience, and stuff.
25. We do this using a profound and empathic understanding of experience, design is like judo, working with forces to redirect them towards something.
26. Solving for both ends: A context in the future. Has four parts
the initial conditions
the endpoint
the signposts
and the frame architecture. Expand and example for each of these terms. What they provide is a new laundry list of what you can understand and what you can change
27. Good Work comes out of trying to do good within various frames of reference: cultural, social, domain, field…. Not an absolute. But you need to understand the levels.
28. The positive notion of subversion from Marcuse
29. What’s the right word for this values-based, systemically-scaled, experience-frame design? Not sure.
30. Examples (pos + neg): Segway, Hybrid Escalade, IHI