phenotypic
research
design

New approaches to data have sparked debate over how we construct and apply knowledge in many domains. In governance, for example, the dramatically expanding capacity to collect and process data has brought conflicts between traditional notions of privacy and public safety to the fore. Analogous debates are occurring in realms such as business and medicine as we come to recognize that no thread of our social fabric is immune to data-driven transformation. Yet, curiously, the impact of data is rarely addressed as a matter of design — and this despite the fact that designers are actively and intimately engaged in shaping the ways that data are generated and used in other disciplines.

We often hear it said that design and “design thinking” are generative processes. Yet, when it comes to data, designers generally adopt the frameworks and procedures of other disciplines. Even in the field of data visualization, where the promise of data analysis drives virtually every aspect of the practice, data are treated as an exogenous and often exotic commodity. The design disciplines have yet to develop their own approaches to data. While there are many reasons for this deficiency, there are also reasons for concern: designers may find that they have been “deskilled” by those with a tighter grasp on the power of data to effect change.

This chapter outlines a way that designers might draw more value from data. Counterintuitively, it suggests that the key components of a design-oriented approach to data can be found in the natural sciences. Tenets of evolutionary design are examined and juxtaposed to the goal-oriented practices with which designers are generally more familiar. It is not the process of evolution itself that is emphasized, however. Rather, it is the mechanisms by which data are used to describe evolutionary dynamics, and the intrinsic limitations of such description. As is illustrated with examples drawn from urban and architectural design, such limitations can themselves spark the development of new uses of data and, thereby, new ways of engaging the built environment.

Ultimately, the chapter revisits Christopher Alexander’s A Pattern Language to highlight the value that contemporary designers can derive from data. Significant as Alexander’s approach to design has been, it is nonetheless hamstrung by the imposition of an extrinsic structure. Components can be combined in numerous ways, but the number and form of these combinations are limited and, most problematic, they do not facilitate the development of perspectives that have yet to emerge. Today’s designers can overcome such impediments. Given the granularity of data that we now collect from virtually every corner of our environment, and an understanding of how these data feedback into perceptions of progress, we can lay the foundation for a more powerful pattern language. The chapter concludes with a sketch of how this task might be achieved.

Recent advances in the capacity to collect and manage data have been addressed in a wide range of academic journals as well as the popular press. Often, the instrumental value of these processes is emphasized. Pundits and journalists, for example, frequently depict “big data” as a source of innovation, highlighting opportunities that have been derived from the detailed analysis of routine socio-technical interaction. Given the emphasis placed upon matters of application, there has been remarkably little discussion of ways to address the putative value of such analyses from within the institutional context of design education. This paper describes an initiative to prepare design and management students for the data-rich environments in which they will practice. It explains our motivation for introducing these students to basic analytical and computational methodology as well as the framework in which we do so. As exemplified, this approach fosters forms of exploration and experimentation that diverge from conventional approaches to both scientific research and design practice by decoupling the symbolic or referential value of data from their attributes as media. Ways that such training increases our students’ capacity to speculate on future conditions are examined and discussed in light of the larger objective of drawing attention to new ways that designers and managers can use data to steer, as well as to reflect upon, the course of innovation.

Big data is often portrayed as a source and driver of innovation, but design practitioners rarely figure into such descriptions. It is usually scientists or analysts who are depicted and the production of knowledge that is discussed. This process differs from that of design in significant ways; scientists provide new insight into extant reality, while designers boot-strap their way to new futures.

It is evident, however, that the ability to collect and process vast amounts of data can facilitate the process of design. And it is often scientists and analysts who are best positioned to undertake such activity because they are trained to work with data. These skills enable them to engage in speculative processes such as predictive analysis, for example, and to mine data in ways that bear more resemblance to bricolage than hypothesis testing. While such pursuits may not be informed by heuristics that are characteristic of design, they do suggest ways in which designers can generate meaningful value from data. They also presage a less desirable alternative. If designers do not develop the capacity to work with data, they may be deskilled as the socio-technical significance of this “medium” continues to grow.

Though it may be more common to associate data with the hard evidence of empirical science than with the speculation that is intrinsic to design, much of the data that is currently being generated in the course of people’s everyday interactions with their technology can be a tremendous resource for designers who have been trained to work with data. This is because these data are more than a window on the nuances of human behavior, they are also a means of leveraging such behavior. The empirical insight offered by these data is valuable to scientists and designers alike, but designers need not constrain their uses of data to the Scientific Method.

Designers can complement scientific practice and extend their own strategic capacity by using data to increase and diversify the range of actors engaged in design processes and, thereby, more fully engage the socio-technical systems that drive innovation. We consider this capacity to be an important component of future design practice, and we believe that design educators should teach designers to leverage the new opportunities that data can offer. This is not to suggest that designers should be trained as scientists or analysts; rather, we should provide our students with the conceptual frameworks that they need in order to use basic quantitative skills as a means of design. We think that meta-design offers such a framework. As we illustrate with examples of our students’ work, meta-design can enable designers to exploit heuristics that are foreign to analysts (so-called “design thinking”) in a manner that complements analytical methodology and enhances their ability to shape the complex systemic processes that are intrinsic to big data.

While the inherent value of user-centered design is broadly acknowledged, the dynamics of socio-technical interaction receive relatively little attention in the context of design education. The importance of defining users’ needs is emphasized, but ways of understanding the systemic dynamics from which these needs arise are rarely articulated. In effect, this leaves designers ill-equipped to address situations in which the final form of their work is determined by forces that cannot be adequately addressed at the individual level of analysis – situations that are becoming increasingly common as a result of the widespread adoption of networked technologies with interfaces that are inherently plastic. This paper highlights the importance of educating designers to address complex processes that extend across multiple levels of analysis. Current limitations of user-centered methodologies are articulated in order to clarify ways in which the role of the designer is changing and why it is important for us to reconsider how we train design students. The development of curricula in which processes of innovation are tightly coupled to those of mapping is suggested as a means of addressing these concerns. This systems-oriented approach is exemplified by coursework in which students create interfaces to map socio-technical dynamics of urban environments.

This paper concerns the development of tools that enable teams of culturally dissimilar and geographically dispersed students to leverage endogenous group discord as a means of gaining valuable insight into systemic dynamics that shape their work. We describe a methodology, the cultural quest, which we developed in the context of a course on design research which we offered to students located in the United States and Germany. Derived from an ethnographic technique known as the cultural probe, the cultural quest enabled these students to juxtapose local, indigenous impressions of cultural dynamics with perspectives generated from afar. After describing this process in detail, we exemplify its pedagogic value with student-generated work. We then discuss ways in which the cultural quest sheds light on the evolving relationship of research to design.

In the weeks following Anthony Dunne’s Stephen Weiss lecture, Design + Management faculty member Raoul Rickenberg conducted the following interview with Dunne, in which they expand several themes that arose in the lecture itself. Among the topics discussed are the hermeneutical context within which Dunne situates his body of work―which spans domains of art, science, and politics―and the nature of Dunne’s collaboration with Fiona Raby and his colleagues and students at the Design Interactions Department at the Royal College of Art.

Generically used in reference to the allocation and management of design-related resources, design planning denotes the conceptualization, specification, and articulation of goals and processes that are used to organize design efforts. Typically, design planning is most salient in the early stages of projects, when strategies and tactics are addressed in a formal manner. It is not unusual, however, for designers to adjust such structural frameworks throughout the course of a project in response to unforeseen developments. For this reason, design planning is best viewed as an ongoing activity that addresses organizational aspects of design at multiple levels of granularity in formal as well as informal ways.

Designers use a broad range of methods in the course of their work, many of which are also employed by practitioners in other disciplines. It is the generative manner in which designers press such methods into service that is distinctive.

For the purposes of this definition, it is useful to examine design methods in relation to scientific methods. In both scientific and design practices, methods serve as an infrastructure through which information is conveyed and knowledge is codified, and are thus used to delineate legitimate forms of engagement. In other words, they are the rules and routines with which practitioners develop common perspectives and build upon lessons learned by others. Although scientific and design methods can be seen as sharing the same overall function, the processes through which they are employed differ. In descriptive pursuits such as the sciences, accurate depictions of reality are of primary concern; as such, practitioners must adhere to reliable protocols and apply methods in a uniform, consistent manner in order to calibrate their work against external criteria. Such calibrations have less significance in the course of design, which is oriented toward the conception of that which is new. Design practitioners have more latitude to employ methods in an improvisational manner, and are able to adopt, discard, and tailor methods to address the specific circumstances in which they are used. Designers often define their methods on the basis of project briefs, for example, while scientists must contend with universal laws and previously established principles. It is in the terms that are used to assess quality, however, that the differences between these approaches to methods are most salient. Scientific findings are evaluated on the basis of validity and reliability, whereas the work of designers is judged on the basis of ingenuity.

Since designers do not need to produce accurate depictions of reality, they are thus free to experiment with methods in a manner that other disciplines do not tolerate. Methods are often applied with little regard to the parameters for which they were developed, for example, and immediate feedback is generally valued over procedural competence. Such practices can obscure lessons that are embedded in methods in ways that appear to be at odds with convention, but this cannot be decoupled from design’s own disciplinary expectations concerning creativity – improvisation is often the only way to proceed when innovation is the objective.

At a basic level, most descriptions of the design process concern ways in which form is derived from interactions between actors and their environments. In practice-oriented disciplines, such as Architecture, Product Design, or Engineering, the design process is generally viewed as the means by which people shape their surroundings. Designers are expected to define problems that can be solved in a step-wise manner. They are trained to conceptualize the process of design as a series of activities that unfolds over time, and to view the completion of each activity as a step toward some pre-defined goal. In other words, designers are expected to model futures that can be realized through strategic engagements with their environments. Implicit in this formulation is the assumption that designers are rational actors. Moreover, it is assumed that environments are stable enough to be modeled, yet pliable enough to be shaped. While such situations may arise, they are far from common. In practice, the process of design only approaches this ideal when rationality is tightly bounded. Architecture, for example, may be an efficient way for people to solve some problems that are tightly coupled to the built environment, but such efficiencies soon dissipate when these problems are located in broader social or economic contexts.

In other disciplines, such as the natural sciences, the agency of the actor is rarely so privileged. Instead of addressing the design process as a means of solving problems, the process is usually described in terms of the structural relationships that exist between actors and their equally instrumental environments. It is the alignments and misalignments of these two factors that give rise to form. In Darwinistic theories of evolution through natural selection, for example, there is no need for rational actors because the process of design is motivated solely by environmental “fitness.” There is no need for rational actors because form develops in the absence of pre-defined goals. It is precisely this designer-less conception of design that distinguishes evolutionary approaches to the process from those of most practice-oriented disciplines. The fact that design occurs in the absence of rational actors and their strategic plans does not mean, however, that the two approaches are at odds. Everyday experience clearly suggests that both processes co-exist. What is required is a perspective from which the two approaches to design can be viewed on a continuum – a definition that addresses the process of acting on an environment as well as that of acting in an environment.

Every design initiative is situated in its own complex environment of dynamic and interrelated requirements. One of the ways in which designers deal with the difficulty of working in such situations is by employing higher-order frameworks such as rules and heuristics. These frameworks codify lessons gained from prior experience in ways that enable designers to model and stabilize dynamics that are at play in unfamiliar environments. The effectiveness of this practice is severely compromised, however, when the environments in which lessons are applied differ significantly from those in which they were learned. It is useful, therefore, to articulate the nature of relationships between dynamic environments and, in particular, to address such relationships in the form of “hierarchically-nested systems.” Such hierarchies enable designers to focus on “niches” that are characterized by the features of local situations (such as their particular social, economic, or political dynamics). By identifying common conditions in which such features are expressed, designers can consider the viability of deploying practices across niches in advance, and thereby generate estimates of potential fitness.

Clearly, such forethought can be advantageous in that it permits strategic planning. Designers need not plan in advance, however, in order to engage complex environments effectively. Parallel and complementary approaches can be used that allow problems and solutions to emerge without such rational analysis. This is particularly relevant in the pursuit of innovation, where the end goal is not explicitly defined at the outset of the process. For instance, many elegant designs are the result of tacit lessons that designers have learned by continually testing their intuitions and aptitudes as they experiment with alternate solutions. This approach to design is significant in that it echoes the concept of “local optimization” that scientists use to explain the development of form in the natural environment. In both cases, the design process may appear to be random and open-ended, but it is still path-dependent in that each step of the process inherently limits the range of possible subsequent steps.

The goal-oriented practice that informs many approaches to design can be at odds with the systemic focus that is inherent in evolutionary perspectives on the design process, but this need not be the case. While the desire to downplay the importance of individual-level initiative can be compelling when addressing the design from an evolutionary perspective – perhaps because this stance is often taken in the context of scientific disciplines – there is no need to view the design process in a dichotomous manner. As a human activity, the process of design can be goal-oriented as well as fitness-driven, it can be motivated by rational choice as well intuition, and it can occur in environments that are stable as well as dynamic. Regardless of whether the generation of form is considered from an evolutionary perspective or from that of the practice-oriented disciplines, richly nuanced definitions of the design process can be derived from the interaction between actors and their environments.

It is difficult to imagine how one could teach design without asking students to articulate how they expect whatever they are developing to be used. Conceptualized in terms of functionality, the importance of addressing artifacts from the perspective of use has grounded the practice as well as the pedagogy of design since the early years of the industrial revolution [Beniger, 1989; Nobel, 1984]. Until relatively recently, however, the human dimension of usability – the fact that it is people who contextualize and use artifacts – was largely ignored. The term “human factors” did not enter the discursive formation of design until the early years of last century, when Frederick Taylor and his colleagues drew attention to the relationship between efficiency and the differing levels of knowledge and ability that people bring to their interactions with artifacts [Taylor, 1947; Rabinback, 1990]. And even with this insight into efficiency, designers rarely addressed usability in a manner that accounted for the ways in which people actually perceive of their interactions with artifacts until the final twenty years of the century [Chapanis, 1996; Norman, 1988]. It was only with the wide-spread adoption of a conceptual framework generally known as user-centered design that designers expanded their focus to explicitly address ways in which people’s interactions with artifacts are informed by needs and desires as well as by knowledge and abilities [Alexander, 1977; Winograd, 1996; Laurel, 1999]. Many factors motivated the adoption of this framework, but the social sciences clearly played a formative role in the process – it was only by appropriating the methodologies of psychology, anthropology, sociology, and other social sciences that designers gained insight into the motivations of those who use their designs. Today, it is not unusual for clients to expect the designers that they hire to be conversant, if not fluent, in social science methodologies.

The process of mapping is a form of decentralized research. It is also a form of decentralized design. Maps reflect the specifics of situation without inhibiting interpretation; they define perspectives and relationships without articulating implications. By making maps, social entities shape technology in situ, as the forms and meanings of their organizations unfold. If we address ubiquitous computing as a form of mapping – as a way of distributing the capacity to design – we can conflate research and design in a manner that will enable useful technologies to emerge in the context of use.

It can be difficult to gain perspective on social dynamics. So it is fitting that the busiest hub in New York State's vast network of public transportation, Penn Station, is concealed beneath Madison Square Garden, itself a nondescript building. But this site is more than a metaphor for the conceptual difficulty of understanding public activity. The architecture of Penn Station affords few tangible opportunities for individuals to observe the manner in which their journeys through the building affect those of others. Without such feedback, it is difficult to appreciate the value of public space, much less steward its development. This project demonstrates how technology that is typically used to maintain social order can be repurposed to provide the feedback necessary for new conceptions of public space to emerge.

Animated characters are common in user interfaces, but important questions remain about whether characters work in all situations and for all users. This experiment tested the effects of different character presentations on user anxiety, task performance, and subjective evaluations of two commerce websites. There were three character conditions (no character, a character that ignored the user, and a character that closely monitored work on the site). Users were separated into two groups that had different attitudes about accepting help from others: (1) people with control orientations that were external (users thought that other people controlled their success) and those with internal orientations (users thought they were in control). Results showed that the effects of monitoring and individual differences in thoughts about control worked as they do in real life. Users felt more anxious when characters monitored their website work and this effect was strongest for users with an external control orientation. Monitoring characters also decreased task performance, but increased trust in website content. Results are discussed in terms of design considerations that maximize the positive influence of animated agents.