Design Trade-Offs for Quality of Life

INTERACTIONS JANUARY– FEBRUARY 2018 VOLUME XXV.1 Association for Computing Machinery

SPECIAL TOPIC: Taking Action in a Changing World Attending to the Things of Design at CHI A Manifesto for Prosperity Radar Sensing in HCI 26 INTERACTIONS JANUARY–FEBRUARY 2018 INTERACTIONS.ACM.ORG COVER STORY Exploring Design Trade-Offs for Quality of Life in Human-Centered Design Gerhard Fischer, University of Colorado, Boulder

Insights To understand, foster, nurture, and of disciplines, this article explores → A QoL framework must explore support Quality of Life (QoL) is one of requirements and components to innovative sociotechnical the most challenging design problems create a framework for QoL with an environments contributing to of the digital age. QoL is a broad emphasis on specifc design trade- creativity and enjoyment. concept without a precise, generally ofs. Its insights and arguments are → Efficiency and productivity of accepted defnition. It is not out in summarized in requirements for the digital technologies does not the world to be discovered, but rather design of sociotechnical environments necessarily increase QoL. is an objective achieved by design. to address future challenges for human- → Design problems have no Design is choice: It is an argumentative centered design (HCD) grounded in a “correct” solutions; the success process with no optimal solutions. In QoL perspective. of a solution is a question of design, trade-ofs are universal and The frst phase of research and stakeholder value and interests. unavoidable because there are no best development in HCD was focused → Design trade-offs widen solutions or decontextualized sweet on concerns about usability and rather than narrow design spots independent of specifc goals, usefulness. As hardware and software spaces and show there objectives, and values. for many applications became readily are no decontextualized sweet Grounded in ideas and research available, new concerns emerged,

IMAGE BY JUERGEN_WALLSTABE BY SHUTTERSTOCK.COMIMAGE / spots for wicked problems. activities from a broad spectrum including design methodologies such

INTERACTIONS.ACM.ORG JANUARY–FEBRUARY 2018 INTERACTIONS 27 COVER STORY as participatory design, giving all The arguments here are grounded nature of wishing? Human beings value stakeholders a voice, incorporating in the problem domains our research things and relationships they must requirements and insights from diferent has addressed over the years, including make an efort to obtain and in which disciplines, and taking advantage of human-computer interaction, end- they fnd purpose, enjoyment, and fow a deeper diferentiation of human user development, lifelong learning, states through personally meaningful thinking. Frameworks grounded in creativity, urban planning, cognitive tasks [9]. QoL perspectives should incorporate disabilities, and energy sustainability. In order to gain empirical evidence the fndings and practices of these These are also the problem domains for about how people think about earlier foundations, but also need to which we have designed, developed, and QoL, we collected initial data with take into account new requirements analyzed sociotechnical environments [7]. questionnaires and interviews from derived from additional disciplines, Based on our emphasis on design, we diferent constituencies by asking including behavioral economics and have focused particularly on identifying them how specifc systems and gadgets social psychology [1], creative practices design trade-ofs as the most basic (including e-mail, smartphones, and end-user development [2], social characteristic of design; this article Facebook, Twitter, Wikipedia, Uber, production [3], and new agendas for attempts to create frames of reference and Airbnb) have impacted their lives. theory and practice in computing [4]. (illustrated with examples drawn Their answers clearly indicated the QoL as a concept transcends from diferent areas) for exploring and numerous trade-ofs, for example: the domain of information and understanding the implications of “These systems connect me with large communications technology (ICT). design trade-ofs associated with QoL. numbers of people but isolate me from For example, the QoL objectives my surroundings” and “I enjoy the articulated in the Europe 2020 strategy QUALITY OF LIFE: opportunity to participate, but too (http://ec.europa.eu/europe2020/) A FUNDAMENTAL OBJECTIVE many demands create a participation include: democratizing societies, FOR HUMAN-CENTERED overload.” supporting employment and social DESIGN Figure 1 attempts to create an initial inclusion, improving healthcare, and QoL represents a fundamental objective framework for QoL, illustrating its supporting energy and environmental for societies in the 21st century. Does diferent aspects. In order to identify sustainability. Here, I analyze the it mean being happier? Having more more specifc frames of reference and driving forces for articulating and leisure time? Enjoying good health? design requirements, we will 1) argue exploring challenges for QoL derived Having a high standard of living? that QoL is best explored by analyzing from a focus on ICT, complementing Accepting QoL as an important design trade-ofs, 2) mention briefy other approaches such as value- concept does not imply that people some current global developments, sensitive design [5] and positive will agree which objectives will be and 3) describe in some detail some computing [6]. desirable or should be avoided. We have illustrative trade-ofs (see sidebars). ICT has made tremendous progress questioned a vision [8] that many people over the past few decades, but its regard as highly desirable: an “efortless DESIGN AND development and adoption have not world” representing the old dream of DESIGN TRADE-OFFS necessarily led to an improvement humankind to return to the Garden of Every positive value has its price in in QoL. Given the fundamental, Eden or Paradise (places where peace, negative terms ... the genius of Einstein ubiquitous, and global impact of digital prosperity, and happiness can be found) leads to Hiroshima. —Pablo Picasso technologies, it is time to move beyond and live a life of abundance free of all In contrast to the natural sciences analytical post-analyses of risks and work and pain and in which all desires that study how things are, design is benefts to explore the realization that would be satisfed immediately without concerned with how things ought more efciency and productivity do not any efort. However, when all wishes to be [10]. Design problems can be necessarily increase QoL for all of us. get fulflled, how would that change the diferentiated into tame problems, whose solutions are straightforward, and wicked (or ill-defned) problems that have beyond technology related concepts no defnitive formulation, no stopping from “can it be done” to happiness, flow, well-being, “should it be done” human potential, rule, and no boundaries. The most (social and ethical design) personally meaningful problems interesting challenges for ICT and HCD are wicked problems with no correct domains of interest specific QoL factors solutions or right answers (as is the case education, health, control, choice, privacy, in the natural sciences), implying that entertainment, Quality of Life autonomy, access, environment, energy, overload, attention the aim of design in these contexts is mobility (QoL) not to fnd truth, but rather to identify design trade-offs satisfying solutions. design methodologies avoid oversimplified solutions user-centered design, to complex problems, The huge variety of objectives, participatory design, move beyong binary choices, meta design, consider ambiguity as problems, value systems, and people’s value-sensitive design an opening for new insights needs and preferences makes design New Perspectives for Human-Centered Design trade-ofs the most basic characteristic of design. A trade-of is a situation that involves losing one quality or aspect Figure 1. Components of a framework for QoL. of something in return for gaining

28 INTERACTIONS JANUARY–FEBRUARY 2018 INTERACTIONS.ACM.ORG Domain Choice 1 Choice 2 Balance / Synthesis Analysis and Details

uncover unknown problem framing problem solving iterative design; integration [10] and brief analysis here alternatives and of problem framing and limiting factors solving

airplanes and cars self-driving cars advanced driver-assistance mobility for all [11] and Figure 2 systems

exposure to personalization privacy and serendipity context awareness; beyond [12] and Sidebar 1 information “more is more”

control, involvement, prescriptiveness and permissiveness and meta-design; libertarian [1] and Sidebar 2 and participation curriculum; consumer interest-driven learning; paternalism; different cultures cultures of participation levels of engagement

Table 1. Examples of design trade-offs. another quality or aspect. The concept of a Faustian bargain—selling your soul to the devil to gain unlimited SIDEBAR 1. DESIGN TRADE-OFF: knowledge—is used as a metaphor for getting something you want in PERSONALIZATION VS. SERENDIPITY exchange for sacrifcing something else The growth of technology has provided the foundations for more information being available (e.g., people accept new technologies at people’s fingertips and has offered more opportunities for participation and collaboration. and are excited about them without In contrast to technological developments, the growth in human capabilities is limited: Our being aware of their negative impacts). neurons do not fire faster, our memory does not increase in capacity, and we do not learn or think faster as time progresses. The mismatch between these developments has caused Identifying, articulating, and and contributed to information overload, participation overload, collaboration overload, and assessing design trade-ofs represents choice overload. Information overload exists for people in the following contexts: 1) keeping a unique challenge in designing for track of what is going on in the world at large or in the circle of one’s friends (by paying QoL. The value and the contribution of attention to e-mails, blogs, Facebook postings, and tweets), 2) needing infinite resources analyzing design trade-ofs is grounded for learning something (including the Web, Wikipedia, TED lectures, MOOCs), and 3) coping in the following objectives: with high-functionality environments (including digital cameras, word-processing systems, • Avoiding oversimplifed solutions drawing applications, and programming environments). Information consumes the attention of its recipients. Hence, a wealth of information that result from ignoring important creates a poverty of attention and requires strategies and support environments to allocate facets of complex problems efficiently among the overabundance of information sources to which we pay attention. • Uncovering unknown alternatives No person can afford to pay attention to more than a fraction of new things produced. and identifying the truly limiting Because of the scarcity of attention, people must be selective. The challenge is not to create factors that underlie problems sociotechnical environments based on anyone having access to information “anytime and • Transcending one-sided views and anywhere” but rather to create context-aware systems that focus on “the ‘right’ information, groupthink by overcoming the hype or at the ‘right’ time, in the ‘right’ place, in the ‘right’ way, to the ‘right’ person” [12]. System developments to reduce the information overload problem include developing underestimation associated with many techniques for personalization and making information relevant to the task at hand, for technological developments example: search engines, context-aware applications, adaptive and adaptable components, • Considering ambiguity as an personalized learning environments (based on learning analytics), recommender systems, opening for insight and refections, and personalized news streams—all help people to find the needle in the haystack. rather than a bug to be fxed The design trade-off associated with a strong emphasis on personalization is that • Appreciating the complexity and serendipitous encounters (e.g., giving people something that they want without asking; encountering interesting ideas, things, events, people by chance) and opposing opinions and richness of human experience worldviews are suppressed. • Providing evidence that there are The promises associated with a strong emphasis on personalization are accompanied no decontextualized sweet spots by by pitfalls. Systems tailoring their services (including news and search results) to people’s identifying interesting syntheses and inferred personal preferences and tastes may cause recipients to get trapped in filter meaningful compromises between bubbles (representing a unique universe of information computed by algorithms) [16], which opposing objectives. suppress serendipitous encounters, opposing opinions, and different worldviews. Filter Table 1 enumerates a few of the bubbles may lead to groupthink with a loss of individual creativity and independent thinking, as well as a tendency to minimize conflict and reach a consensus decision without critical major design trade-ofs related to the evaluation of alternative ideas or viewpoints. To transcend these contradictory objectives contents of this article. requires finding an adequate balance between personalization and serendipity by designing There is widespread agreement interaction mechanisms that allow users to select their own personal, situation- and among scientists, designers, and time-dependent best mixes of these design trade-offs—providing another example that decision makers that the formulation there are no decontextualized sweet spots. of a problem already contains half its In our research we have developed context-aware information delivery systems [12] to solution. If we misstate the problem, explore a mix between personalization and serendipitous information delivery. Supporting users to incrementally learn more useful operations in high-functionality environments we may preclude a satisfying solution. (including software reuse libraries, MS Office, apps on smartphones, MOOCs, etc.), The overemphasis on problem solving our systems (based on user and task models) differentiated between personally meaningful compared with problem framing and personally irrelevant information. is nicely illustrated by an example

INTERACTIONS.ACM.ORG JANUARY–FEBRUARY 2018 INTERACTIONS 29 COVER STORY described by Herbert Simon of the people but rather as a problem of driving, no more distracted driving), importance “to uncover unknown accommodating cars everywhere— 3) fewer cars (based on the reduced alternatives and identify the truly even to the point of rebuilding cities for need for owning a car), 4) better use of limiting factors that underlie cars, thereby diminishing alternative existing cars (more than 90 percent of problems”: transportation choices. all cars are not driven at a particular A few years ago, the State Department While there is no simple recipe moment), and 5) spending more time was troubled by the congestion that for how to identify the truly limiting on personally preferred activities (e.g., afected its incoming communication lines resources or the “real” design problem, reading a book or sleeping instead whenever there was a crisis abroad. The an emphasis on 1) the identifcation of driving). The arguments of the teletypes, unable to output messages as of relevant design trade-ofs, 2) the opponents are centered on: 1) loss of rapidly as they were received, would fall involvement of all stakeholders to control (many people fnd driving fun many hours behind. Important messages bring as many voices to the table as and engaging), 2) loss of jobs behind to Washington were seriously delayed in possible, 3) the analysis of errors and the wheel (trucks, taxis) and in the car transmission. breakdowns, and 4) the iterative design industry (as substantially fewer cars are Since printing capacity was identifed by building inspiring prototypes needed), and 3) unclear responsibilities as the limiting factor, it was proposed (creating objects to think with) will if something goes wrong. Beyond the to remedy the situation by substituting all contribute to emphasize problem two endpoints of “no technological line printers for the teletypes, thereby framing in order to avoid developing support” and “complete automation” increasing output by several orders of solutions to the wrong problem. Figure (these challenges have been pursued magnitude. No one asked about the next 2 illustrates an important contribution with aircraft for some time [14]), link in the chain: the capacity of ofcers at of a design trade-of analysis: It widens numerous driver-assistant systems have the country desks to process the messages the design space. The truly limiting been developed and more are becoming that would come of the line printers. A factor (as described in the printing available in today’s cars, including deeper analysis would have shown that example above) is not self-driving cars, adaptive cruise control, collision- the real bottleneck in the process was but rather explorations for creating a avoidance systems, parking assistants, the time and attention of the human framework for “mobility for all” in the and car-to-car communication. decision makers who had to use the 21st century [11]. Exploring the impact of self- incoming information. Identifcation of the driving cars on QoL, the question bottleneck would have generated in turn a SELECTED MAJOR CURRENT remains whether the truly limiting more sophisticated design problem: How DEVELOPMENTS THAT IMPACT factors have been identifed. Exploring can incoming messages during a crisis QUALITY OF LIFE the future of mobility provides an be fltered in such a way that important To illustrate the necessity for a opportunity for a paradigm change, information will have priority and will comprehensive QoL framework, from a focus on the automobile to the come to the attention of the decision two current examples of major consideration of many more options makers, while unimportant information developments, including the design and design trade-ofs [11]. There are will be shunted aside until the crisis is trade-ofs associated with them, will be major stakeholders who have a vital passed? [10] briefy discussed and analyzed. interest in shaping and infuencing this This is not an isolated example. The Self-driving cars and mobility future: car manufacturers, information lessons learned here can be applied for all. Ten years ago, self-driving technology companies, environmental to the design trade-ofs between cars seemed to be more a topic for groups, and politicians, to name a personalization and serendipity science fction than a near-term few. Figure 2 provides an overview of (Sidebar 1) and to the relationship of reality. As rapid progress is made the multifaceted dimensions of the self-driving cars and mobility for all at the technological level (“it can be transformative changes that should be (Figure 2). Like any useful tool, the done soon”), the design trade-ofs considered in exploring “mobility for car has purposes to which it is ideally associated with the social and ethical all” rather than “self-driving cars” as suited. It is particularly well suited to levels (“should we do it”) (Figure 1) take the core problem [11]. low-density areas; it can also be very center stage. The proponents of self- The sharing economy. The sharing useful in dense cities as a taxi. But those driving cars [13] argue: 1) independence economy [3] has emerged as an with an interest in promoting cars for people who cannot drive (based on alternative model in several domains succeeded in casting transportation impairments such as age or blindness), (e.g., Uber and Lyft in transportation, problems not as a problem of moving 2) fewer accidents (no more drunk Airbnb in accommodation) by providing consumers with alternative, convenient, and cost-efcient access to resources and services. Many stakeholders stress the positive Those with an interest in promoting cars impact of the sharing economy on succeeded in casting transportation QoL: Consumers see it as a way to (sometimes) get cheaper and more problems not as a problem of moving readily available services; drivers people but rather as a problem of and renters earn some additional money; and the companies supporting accommodating cars everywhere. these services have become some

30 INTERACTIONS JANUARY–FEBRUARY 2018 INTERACTIONS.ACM.ORG of the most economically valued ones. Others experience the negative Mobility for All impact of the sharing economy: Taxi companies and hotels are driven out global problem framing of business; professional drivers and hotel employees lose their jobs; taxes facilitate and support it reduce the need for it are not paid; and safety may be at risk.

The sharing economy is facilitated major alternatives telework Skype/Hangout and supported by ICT developments; researchers and practitioners in HCD and related felds will be responsible mass transportation automobiles for shaping these developments (airplanes, trains, buses, bikes, walking) that ultimately impact the QoL of many people. Figure 3 illustrates the multifaceted developments associated self-driving cars human-driven cars environmental impact costs with a shared economy (many of which are related to the concepts associated independence safety enjoyment assistant electric fewer owning renting with mobility for all, documented in for people systems cars cars a car a car unable to drive car sharing Figure 2).

Figure 2. The multifaceted dimensions of mobility for all (the nodes in black are not further A RESEARCH AGENDA elaborated). FOR THE FUTURE The previous sections of this article have explored QoL as a transformative examples: Uber, Lyft, Airbnb, Etsy, technological developments (“it can be done”): framework for HCD grounded in a TaskRabbit, PartatmyHouse, … ubiquity of mobile devices, networks, platforms number of fundamental requirements, including to avoid oversimplifed Sharing Economy solutions that result from ignoring important facets of complex problems (relying on one-sided approaches design trade-offs social and ethical issues (“should it be done”): instead of considering design trade- • quality of life (who in society ofs), and to explore a continuum of blessing (benefits) curses (pitfalls) benefits from the changes); • autonomy of citizens • job losses • from ownership to services richer sets of choices by identifying the • exploitation of • circumventing labor laws most adequate balanced approach (by excess resources • evading fair taxes exploiting the strengths and avoiding the weaknesses of the endpoints Figure 3. The multifaceted dimensions of the sharing economy. defning the trade-ofs). The following sections briefy An alternative for a QoL framework amounts of data are: 1) everything describe a small number of objectives includes developments that shield that can be measured should be for a research agenda exploring QoL as people from unwanted information measured, and data is a transparent a transformative framework for human- and help them to focus their lives on and reliable lens through which to centered design. their interests, passions, and dreams make informed decisions; 2) we should Questioning the “more is more” by exploring the design requirement analyze the behavior of learners (e.g., philosophy of life. Digital technologies “less is more.” This approach will with learning analytics); and 3) we have accelerated the production and emphasize the following requirements: should illuminate patterns of behavior consumption of information. This 1) innovations protecting people and functions that we are unable to information comes about from humans from the tidal wave of information observe and analyze. The positive acting as bloggers, producers of movies, (including “do not call” telephone impacts of big data on diferent QoL and participants in social networks; lists, e-mail flters, and cellphone-free dimensions are: 1) overcoming beliefs, and from sensors embedded in the zones); 2) regulations that shield people opinions, and mistaken assumptions cyber-physical systems surrounding us from being available at all times (e.g., by documenting a descriptive account [15]. “More is more” has its attractions limiting or barring e-mail after work of how things are (thereby exposing and rewards and has been embraced hours to achieve a desired work/life when our intuitive view of reality by many people. For example: more balance); and 3) opportunities and is wrong); 2) reducing information slides in a presentation; more Facebook incentives for humans to change their overload via personalization and friends, Twitter followers, and behavior (e.g., engaging consumers in making information relevant to the task LinkedIn connections; more students occasional “information celibacy” by at hand (see Sidebar 1); 3) collecting in a MOOC; more publications and abstaining from the constant flow of information and giving feedback on a higher H-Index; more apps on our information). activities in the physical world (e.g., smartphones; and more “new version” Beyond the exclusive reliance on activity trackers providing data messages from the companies whose big data. Some of the assumptions for a “quantifed self”); 4) enabling systems we use. behind our ability to gather huge new business models (as the sharing

INTERACTIONS.ACM.ORG JANUARY–FEBRUARY 2018 INTERACTIONS 31 COVER STORY SIDEBAR 2. DESIGN TRADE-OFF: PRESCRIPTIVENESS VS. PERMISSIVENESS Sociotechnical environments for autonomy and can do whatever they want promising mix. Numerous attempts to computation, learning, and decision to in a self-determined way. As a specific explore and support the middle ground by making can be characterized by opposing example, Table 2 contrasts “curriculum- exploiting the strengths and minimizing choices on a continuum: being prescriptive driven learning” (primarily prescriptive) the shortcomings of the prescriptiveness or being permissive. In prescriptive with “interest-driven, self-directed versus permissiveness spectrum include: environments, designers define context learning” (primarily permissive) learning guided discovery learning in education and rules by creating finished systems at environments. (representing a mix between curriculum- design time (by defining rules, checklists, The opposite choices described in driven and self-directed learning; see and workflow processes), and teachers the two columns of the table are design Table 2), meta-design, and libertarian define curricula and provide guidance trade-offs. The challenge to contribute paternalism (briefly described in the by acting as “sages on the stage.” In positively to QoL lies (as in the other section “A Research Agenda for the permissive environments, users have examples given) in identifying the most Future”).

Curriculum-Driven Learning Interest-Driven, Self-Directed Learning

characteristics problem is given by the teacher or the system; problem is based on the learner’s needs and learning is driven from the supply side interests; learning is driven from the demand side

strengths organized body of knowledge; pedagogically real interests, personally meaningful tasks, and cognitively structured presentations high motivation

weaknesses limited relevancy to the interests coverage of important concepts may be missing; of the learner or the task at hand unstructured episodes; lack of coherence

primary role of the teacher sage on the stage—presents what guide on the side—answers questions he/she knows and is prepared for posed by learners

planning versus situated responses anticipation and planning of learning needs arise from the situational context the learning goals and content

distribution over lifetime decreasing in importance from school to increasing in importance from school to university university to lifelong learning to lifelong learning

assessment “standard” assessment instruments “innovative” assessment instruments are applicable are needed

unique research challenges presentation of an organized body of knowledge; task identification; context awareness responsiveness to individual differences

Table 2. Distinctions between curriculum-driven learning and interest-driven, self-directed learning.

economy mentioned above); and Meta-design and libertarian notion of system design beyond the 5) helping us be environmentally paternalism: Distributing control. original development of a system by responsible (e.g., with smart grids and Meta-design (focused on “design supporting users as co-designers. smart meters). for designers” [17]) is a theoretical It is grounded in the basic A trade-of analysis of big data framework to conceptualize and to cope assumption that future uses and uncovers some of the negative in unique ways with design problems. problems cannot be completely consequences for QoL dimensions: 1) In a world that is not predictable, anticipated at design time, when a the tendency to value what we measure improvisation, refection, evolution, system is developed. Users, at use time, (based on data that is easy to obtain) and innovation are more than luxuries: will discover mismatches between their rather than focusing on measuring They are necessities. The challenge needs and the support that an existing what we value, 2) the quantifcation of design, particularly in the context system can provide for them. These of fundamentally holistic human of wicked problems, is not a matter mismatches will lead to breakdowns experiences combined in many cases of getting rid of the emergent, but that serve as potential sources for to the reduction of a single number, 3) rather of including it and making it an new insights, new knowledge, and privacy violations, 4) the elimination opportunity for more creative and more new understanding. Meta-designers of the positive aspects of forgetting, adequate solutions to problems [2]. use their own creativity to produce and 5) the narrowing of our exposure Many approaches force all the design sociotechnical environments in which to diferent themes and value systems intelligence to the earliest part of the other people can be creative, defning determined by flter bubbles [16], design process, when everyone knows the technical and social conditions for leading to an increased polarization of the least about what is really needed. broad participation in design activities. (particularly online) discussions. Meta-design extends the traditional It is important to point out that the

32 INTERACTIONS JANUARY–FEBRUARY 2018 INTERACTIONS.ACM.ORG goal of making systems modifable by be free to do what they like. The nudge 5. Friedman, B., Kahn Jr., P.H., Borning, users does not imply transferring the framework shares many objectives that A., and Kahn, P.H. Value sensitive design responsibility of good system design meta-design pursues in the context of and information systems. In Human- Computer Interaction and Management to the user. In general, end users will human-centered design. Information Systems: Foundations. P. not build tools of quality equal to that Zhang and D.F. Galletta, eds. Routledge, of a professional designer; they are not CONCLUSION 2006, 348–372. concerned with the tool, per se, but If information and computing 6. Calvo, R.A. and Peters, D. Positive rather with doing their work. However, technologies are developed to improve Computing—Technology for Wellbeing and if the tool does not satisfy the needs or the QoL of all humans, then it is Human Potential. MIT Press, Cambridge, tastes of the users (which they know necessary to analyze what those needs MA, 2014. 7. Arias, E.G., Eden, H., and Fischer, G. The best themselves), then users should be are and how technology is required to Envisionment and Discovery Collaboratory empowered to adapt the system without meet them. Beyond basic needs (food, (EDC): Explorations in Human-Centered being dependent on developers. water, and shelter), needs for necessity, Informatics. Morgan & Claypool An interesting design trade-of importance, and urgency are not Publishers, San Rafael, CA, 2016. discussed broadly in behavioral something imposed by nature upon 8. Fischer, G., Greenbaum, J., and Nake, F. economics and public policy is to fnd humanity, but rather are conceptual Return to the Garden of Eden? Learning, working, and living. The Journal of the a balance between the binary choices categories created by cultural choice. Learning Sciences 9, 4 (2000), 505–513. (see Sidebar 2) of paternalism (being Humans are creatures of both 9. Csikszentmihalyi, M. Flow: The Psychology prescriptive) and libertarianism needs and desires. Therefore, a QoL of Optimal Experience. HarperCollins (being permissive). The book Nudge [1] framework should be grounded in not Publishers, New York, 1990. introduces and advocates libertarian only understanding new media and 10. Simon, H.A. The Sciences of the Artifcial, paternalism and associated concepts technologies in terms of productivity, third ed. The MIT Press, Cambridge, such as choice architects and well- efciency, reliability, and from economic MA, 1996. 11. Mitchell, W.J., Borrini-Bird, C.E., and chosen defaults. The implications perspectives, but also in exploring Burns, L.S. Reinventing the Automobile: and the consequences of employing innovative sociotechnical environments Personal Urban Mobility for the 21st nudges are illustrated in the book with that contribute to human creativity, Century. MIT Press, 2010. a variety of examples contributing to gratifcation, and enjoyment. 12. Fischer, G. Context-aware systems: The QoL in domains such as health, wealth, Design trade-ofs are important ‘right’ information, at the ‘right’ rime, in and happiness. Nudges distribute because the future of the digital age is the ‘right’ place, in the ‘right’ way, to the control among choice architects (e.g., not out there to be discovered; it must be ‘right’ person. In Proc. of the Conference on Advanced Visual Interfaces. G. Tortora, S. policymakers in governments, meta- designed. Technology developments Levialdi, and M. Tucci, eds. ACM, New designers, teachers) and customers are not inevitable, and the HCD York, 2012, 287–294. (e.g., citizens, users, learners). Nudges community should demonstrate that 13. Norman, D.A. The human side of are less coercive than commands, design alternatives are possible. As automation. In Road Vehicle Automation 2. scripts, workfow processes, researchers, we need to explore and G. Meyer and S. Beiker, eds. Springer requirements, or prohibitions. The understand the implications of design International Publishing, 2015, 73–79. 14. Billings, C.E. Human-Centered Aircraft appeal of libertarian paternalism trade-ofs, and engage multiple voices Automation: A Concept and Guidelines. is rooted in the respect it has for in constructive controversies. As NASA Technical Memorandum, Report individual autonomy represented by designers, teachers, educators, and No. 103885. NASA Ames Research the libertarian component. members of scientifc communities, Center, 1991. The QoL design trade-ofs in we need to encourage and support 15. Bell, G. and Gemmell, J. Total Recall. relationship to libertarian paternalism learners of all ages in exploring Dutton, New York, 2009. center on the issue of whether (and QoL requirements, and provide 16. Pariser, E. Beware Online “Filter Bubbles” (TED Video), 2011; http://www. in what ways) individuals want opportunities for nurturing mindsets youtube.com/watch?v=B8ofWFx525s. government, teachers, or parents to for thinking, refecting, and acting in 17. Fischer, G. and Giaccardi, E. Meta- protect them from their own mistakes an informed way by considering design design: A framework for the future of or poor decisions. The critics arguing trade-ofs in all areas of human life. end user development. In End User against nudges believe that individuals Development. H. Lieberman, F. Paternò, may be imperfect decision makers, but and V. Wulf, eds. Kluwer Academic they still possess more information ENDNOTES Publishers, Dordrecht, the Netherlands, 1. Thaler, R.H. and Sunstein, C.R. Nudge— 2006, 427–457. about their lives than others and should Improving Decisions About Health, Wealth, an therefore have more control over their Happiness. Penguin Books, London, 2009. lives than others. By being nudged they 2. von Hippel, E. Democratizing Innovation. Gerhard Fischer is a professor adjunct are deprived of responsibility for their MIT Press, Cambridge, MA, 2005. and professor emeritus at the University of actions and decisions. Supporters of 3. Benkler, Y. The Wealth of Networks: Colorado at Boulder. He is a member of the nudges argue that whatever designers How Social Production Transforms Computer-Human Interaction Academy, an Markets and Freedom. Yale Univ. Press, ACM Fellow, and a recipient of the RIGO Award and decision makers do, they will New Haven, 2006. of ACM-SIGDOC. In 2015, he was awarded an inevitably be setting contexts and 4. Shneiderman, B. The New ABCs of honorary doctorate from the University of default positions anyway and that the Research. Oxford Univ. Press, Oxford, Gothenburg, Sweden. libertarian part allows individuals to U.K., 2016. → [email protected]

INTERACTIONS.ACM.ORG JANUARY–FEBRUARY 2018 INTERACTIONS 33