Master's Thesis: Behaving expressions (Morten Winther)

Morten Winther

B E H A V I N G EXPRESSIONS A form study of shape-chang ing interfaces

Master’s Thesis IT University of Copenhagen

BEHAVING EXPRESSIONS A form study of shape-changing interfaces Morten Winther mwla@itu.dk www.mortenwinther.dk Master’s Thesis | Digital Design and Communication IT University of Copenhagen Copenhagen, June 2015

Supervisor: Anna VallgĂĽrda IxD Lab | IT University of Copenhagen

B E H A V I N G EXPRESSIONS A form study of shape-chang ing interfaces

Abstract

Objects with the ability to change their physical shape are receiving increased attention due to their new potentials for design. Within Human–Computer Interaction (HCI) and interaction design, current research in shape-changing interfaces is primarily dominated by 1) technological advancements, 2) understanding cognitive dimensions of the user experience, and 3) development of smart materials with shape-change as their core property. As a complementary perspective to these, this thesis proposes a form approach to draw attention to the immediate and visceral expressive qualities of computationally actuated objects.

With the basic assumption that any object always tells an abstract story about itself through its expressions and the way it holistically presents itself in the world, a form approach is an aesthetic inquiry into better understanding the expressional dimensions of dynamic forms.

Drawing on design traditions from architecture and the Bauhaus School, the form approach seeks to qualify design literacy to help interaction designers gain a sensibility for the aesthetics of form. Moreover, the approach advocates for aesthetic explorations as an alternative path to imagine new applications and functionalities of shape-changing interfaces, rather than relying on existing conventions for technology.

Formgiving is the designerly practice of shaping an object, and through a series of formgiving experiments, I have unfolded the form approach and explored the expressive qualities of various form compositions. As a synthesis of the experimentation, I have produced two prototypes, Tilting\Plate and Bending\Arches.

The main contribution of the thesis is the identification of five primary elements in a form language for shape-changing interfaces: The point, the line, the plane, the volume, and the force.

Based on my experiments and prototypes, as well as three analytical form descriptions of related works, I illustrate the potentials of the form language as 1) a generative facilitator, 2) a communicative framework, and as 3) a descriptive lens for understanding the form dynamics of actuated objects. The purpose of a form language is to develop interaction designers’ aesthetic sense of the form dynamics of shape-changing interfaces by improving our ability to better articulate nuances in form compositions.

i

Preface and acknowledgements

Throughout my studies at the IT University of Copenhagen, I have continuously sought to explore new – and challenge current – ways of using technology. Rather than developing screen-based solutions to help optimise or improve our everyday lives, my agenda has been to design for more playful, beautiful, and sensory ways of engaging with technology. In this sense, my work is driven by an aesthetic curiosity towards creating alternative ways of embedding technology into our everyday lives.

Shape-changing interfaces are therefore naturally interesting to me because they entail physical actuation, which can allow for new and more bodily interactions with technology. Also, dimensions of physicality and materiality become central to attend to in the design process.

This thesis, thus, describes my efforts towards an aesthetic form approach to shape-changing interfaces. However, as with any other design project, it has been a messy and sometimes confusing journey. When I first began to work on the project it looked rather different from what I present here. Inspired by my participation in a research project on temporal form in interaction design together with Anna Vallgårda, Nina Mørch Pedersen, and Edit Emese Viser during fall 2014, my initial ambitions were to build a full-scale shape-changing wall to explore temporal forms on a larger architectural scale. However, I quickly experienced a lack of language for articulating and explaining relations between spatial movements and temporality. This paved the way for a new focus to develop a more basic understanding of dynamic and shape-changing forms. As such, I moved away from an isolated perspective on temporal form as well as the architectural scale, to instead focus on the form compositions and expressive qualities of computational actuation as a way to improve the design literacy for grasping form dynamics of shape-changing interfaces.

Throughout the process I have felt privileged to be surrounded by great people. Without them, I would not have been able to write this thesis.

First of all, I would like to thank my supervisor Anna Vallgårda for her support and encouragements – not only during my thesis, but throughout the last three years, in which she has made my head spin more than once trying to grasp material and temporal aspects of interaction design.

iii

Victor A/S has made the prototype manufacturing possible by contributing with both financial and material resources. I would also like to acknowledge Norisol and Frederikshavn Duco for their vast know-how and for producing and painting the steel constructions. In particular, I would like to thank my farther, Finn J. Larsen, for facilitating and managing the production process – and most of all for playing along with my weird requests and wishes.

I would also like to thank Lars Toft Jacobsen – for his immense patience and for helping me with the technical set-ups of my prototypes – as well as Cathrine Zebbelin and Lars Yndal Sørensen for their lab companionship.

I am appreciative of Hanne-Louise Johannesen, Louise Söholt Larsen, Kirsten Andersen, Nina Mørch Pedersen, and Henrik Svarrer Larsen for their invaluable input and for proofreading earlier versions of this thesis. Also, I would like to thank Cameline Bolbroe, Tomas Sokoler, and Lars Toft Jacobsen for their insightful comments and thoughts on my project.

I owe so much to Nina Mørch Pedersen who has been my partner-in-crime for the last many years. Writing a thesis can be a lonesome task, but Nina has been there the entire way to share frustrations with, laughs, as well as (often lavishing) lunch dates.

Most of all, I am grateful to my two favourites, Kirsten and my daughter Agnes, for keeping me grounded – and for making me realise how great LEGO bricks are for form analyses!

iv

Contents

page 1

page 7

page 27

page 45

Chapter one Introduction Knowledge contribution Thesis outline

Chapter two A form approach to shape-changing interfaces Shape-changing interfaces A functional approach to shape-changing interfaces A user experience approach to shape-changing interfaces Shape-change as a material property Design space proposal: Shape-changes as aesthetic form expressions

Chapter three Behaving expressions The form gestalt of computational objects A form language for shape-changing interfaces Giving form to computationally actuated objects

Chapter four Methodology Designerly form experimentation Prototype production Descriptive form exercises

page 51

page 67

page 81

page 89

page 101

Chapter five Formgiving experiments Plate series Textile series Arches series Box series

Chapter six Prototype production Tilting\Plate Bending\Arches Communicating with the production company

Chapter seven Descriptive form exercises Topobo: Points on a line and rotating volumes Morphees: A bending plane with internal forces Thrifty Faucet: Compositions of forces

Chapter eight Ramifications of a form approach Form language Design expertise Form studies Closing remarks

References

List of figures

Figure 2.1

Lumen is a basic investigation of how shape-change can be used to communicate information through a physical display. . . . . . . . . . . . . . . . . . . . 10

Figure 2.2

The eight types of shape-change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Figure 2.3

SpeakCup is a digital voice recorder that uses its shape to communicate its functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Figure 2.4

Morphees is a prototype series of actuated mobile interfaces . . . . . . . . . . . . . . 13

Figure 2.5

The Thrifty Faucet is a living interface and explores actuated home appliances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Figure 2.6

Topobo and its kinetic sketch up language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Figure 2.7

The magical feeling in Aerial Tunes results in an engaging experience for spectators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Figure 2.8

The soft mechanical alphabet proposed by Coelho and Zigelbaum (2011) . . 18

Figure 2.9

PneUI is created from silicon material and can change shape when air pressure is applied to it . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Figure 2.10

A polymer material is in ShapeShift to create actuated architectural surface . 20

Figure 2.11

The computational composite PLANKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Figure 2.12

The Lotus Dome reacts to human presence by opening its hundreds of smart flowers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Figure 2.13

The Thematic Pavilion uses shape-change to depict the dynamics of the sea. 24

ix

Figure 3.1

The three form components in interaction design: Physical form, temporal form and interaction gestalt. The shaded area of the figure maps the main focus in this thesis; that is, the relation between physical and temporal form. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Figure 3.2

Examples of form studies from Itten’s course in basic composition . . . . . . . . . 31

Figure 3.3

The four primary form elements in architecture: Point, line, plane, and volume. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Figure 3.4

The proposed basic elements in a form language for shape-changing interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Figure 3.5

Multiple forces are in play in Spatial Sounds, e.g. gravity, a force the counters gravity in the steel construction, as well as the forces involved in the circular movement. However, from an perceptional perspective only one force seem relevant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Figure 3.6

A digital camera redesigned based on meaning-making in the actual form of the object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Figure 4.1

Example of how a set-up and reflections from the experiment were articulated in a reflection sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

Figure 5.1

The form experiments in the Plate series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Figure 5.2

The temporal potentials of the moving plate set-up . . . . . . . . . . . . . . . . . . . . . . 54

Figure 5.3

The experiment series that explored less controllable expressions . . . . . . . . . . 57

Figure 5.4

Top row: Vertically shifting the two contact points relatively Bottom row: Twisting the textile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

Figure 5.5

The bending arches experimental set-ups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Figure 5.6

Although the set-up only has two attachment points it is still possible to produce a rather complex organic expression . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Figure 5.7

These two experiments were an alternative exploration of the bending arches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Figure 6.1

Tilting\Plate is a box with a dynamic top. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 x

Figure 6.2

Both the inner plate and the second frame have axes of symmetry perpendicular to each other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Figure 6.3

There are two motors inside the box. The red lines indicate the strings used to move the plates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Figure 6.4

The temporal dynamic of Tilting\Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Figure 6.5

Tilting\Plate described through the design vocabulary . . . . . . . . . . . . . . . . . . . 72

Figure 6.6

Bending\Arches is made from ten flexible arches. . . . . . . . . . . . . . . . . . . . . . . . 74

Figure 6.7

The strings are attached to the arches in different attachment points. This gives a more complex dynamic form as the temporal form is expressed in the prototype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Figure 6.8

Two linear actuators are run by each their stepper motor, which pulls down the crossbar and thus the arches. The red lines indicate the metal wires that are connected to the crossbar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Figure 6.9

Bending\Arches described by the design vocabulary . . . . . . . . . . . . . . . . . . . . 78

Figure 7.1

An image of Topobo (top) as well as a form analysis based on the design vocabulary (bottom) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Figure 7.2

The forces in Morphees can be described in multiple ways. . . . . . . . . . . . . . . 84

Figure 7.3

The faucet consist of joint planar elements that can each be manipulated by a force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

List of tables Table 3.1 – The four causes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Table 4.1 – Adaptation of Kozel’s phenomenological inquiry to my design practice . . . . . . . 48

xi

Chapter one

Introduction Technological advancements and new smart materials are expanding the possibilities for what we can design. For example, objects and surfaces that can change their physical shape are currently receiving a lot of attention in various design, art, and research communities due to their interesting and new potentials. In Human–Computer Interaction (HCI) and interaction design, physical actuation has been particularly explored to develop new forms of shape-changing interfaces, for instance for information displays or mobile communication (e.g. Roudaut et al., 2013; Poupyrev et al. 2007; Rasmussen et al., 2012). Shape-changing interfaces are interfaces “[…] that use physical change of shape as input or output” (Rasmussen et al., 2012, p. 735). In this thesis, I will argue that current research within shape-changing interfaces has primarily been preoccupied with the technological advancements, better understanding the cognitive dimensions of the user experience of interacting with actuated interfaces, as well as the development of smart materials with shape-change as their core property.

Besides the need for knowledge about the technical possibilities we also need to explore other areas of shape-change, such as aesthetics and expressive dimensions, if we are to fully exploit its prospects. In fact, Rasmussen et al. (2012) argue that more systematic investigations of new perspectives on shape-changes are necessary. In their article, they describe the current landscape of interaction design and HCI research in shape-changing interfaces. They find that several existing studies have focused on functional and technological aspects and they therefore wish to draw attention to the expressive qualities of shape-changing interfaces as a yet under-explored domain (Rasmussen et al., 2012, p. 736). To address such a domain, this thesis is the tentative beginnings of an aesthetic inquiry into shape-changing interfaces. It attempts to qualify the work with the expressive dimensions of computationally controlled actuation as an output modality.

In the last decade, the physical gestalt of computational objects has received increased attention within interaction design (Mazé and Redström, 2005; Robles and Wiberg, 2010; Vallgårda and Redström, 2007). This has led to a holistic focus on the overall compositions of digital and non-digital components. Consequently, renewed attention is given to the dialectic relation between an object’s dynamic form and its aesthetic expression. As temporal dynamics in interaction design have previously been articulated as an object’s behaviour (Arvola and 1

Artman, 2007; Ross and Wensveen, 2010), this thesis then seeks to better understand the aesthetic potentials of the behaving expressions in shape-changing form compositions. My attention will be on the immediate and sensory experience of actuated interfaces; or put it in another way, this thesis focuses on the visceral means designers have available to give an object a certain expressional meaning: “Forms, either abstract or concrete always carry meanings. It is the responsibility of the designer to make good use of these meanings, for example to make products beautiful, exploiting all the sensorial aesthetic languages, to stress the importance of certain values, or to improve a product’s ease of use to create and facilitate richer experiences” (Feijs & Kyffin, 2006, p. 70).

As a result of the holistic focus on the computational object’s form, interaction design has been argued to be a design practice of formgiving (Vallgårda, 2009). As in other design disciplines, formgiving is the activity in a design process that deals with the concrete form creation of an object (Zainal Abidin et al., 2008). In interaction design, form encompasses not only the physical gestalt but also temporal dynamics and interaction. In design in general, the formgiving process if often driven by the intended use of the object and the designer’s aesthetic decisions in the sense that “[a]esthetics concerns the formal reasons explaining and describing the appearance of given things” (Hallnäs and Redström, 2002a, p. 105). While this definition is simple and may trivialise aesthetics to a reductionistic kind of logic, it does offer a graspable conception of its implications for design. It emphasises aesthetics as the reasoning for the designer’s choices for a given form. As such, the designer exercises aesthetic judgements as a way of deciding on one form composition over another. To qualify a form approach to shape-changing interfaces, we must actively train our skills in aesthetic judgements. This can be done by bringing awareness to the expressional means that are available for designers to consciously consider the aesthetic meanings that are always present in an object. This is a key competence for any designer (Hove, 2010) and can be developed through compositional experiments (Itten, 1975) or analytical exercises (Ching, 2014), as known from other design disciplines like industrial design and architecture.

However, whereas these other design disciplines have already identified the basic elements with which they can explore compositional qualities (Zainal Abidin et al., 2008), interaction design has yet to develop a comprehensive form approach and a accompanying form language for enabling us to design, articulate, and critique form and expressional qualities of computational objects. The main objective of this thesis is therefore to propose a designerly approach to shape-changing forms as well as a design vocabulary, consisting of five basic elements, to help interaction design gain a sensibility of the expressive dynamics of actuated interfaces.

To unfold this form approach, I have made a form study of a series shape-changing 2

Introduction

compositions. As part of this, I have performed 14 formgiving experiments with different simple material set-ups. With them, I have iterated on various form compositions and expressional qualities to explore the generative potentials of the design vocabulary. By combining conventional materials, like textile or acrylic plastic, in different compositional set-ups, I have explored various dimensions of form in shape-changing objects. As a synthesis of these experiments, I have produced two prototypes, Tilting\Plate and Bending\ Arches, which are used to point to both descriptive and communicative capacities of the form language. Furthermore, I will describe three existing shape-changing objects in order to challenge the proposed form language beyond my own work and to explore its analytical applicability.

In my design practice, I have explored abstract forms with no regard to their functional application. Throughout the thesis I will therefore denote them interchangeably as interfaces, surfaces, objects, forms, as well as expressionals. However, to position my work in interaction design and HCI, I will primarily refer to shape-changing interfaces (albeit interfaces in its widest sense).

With my thesis, I argue how a form language can assist formgiving practice in interaction design to deal with expressive dimensions of shape-changing interfaces. The main objective is to look at how we can understand the temporal dynamics in an object’s physical form.

Therefore, this study does not focus on the user’s interaction with such objects. Being an interaction designer myself, I will always carry an appreciation of interaction with me and I will naturally consider interaction possibilities in my process. However, I will not explore and cultivate these, but will instead focus on the physical and temporal forms of shape-changing objects and their expressions.

The ambition is to propose a design vocabulary to inform design practice, and therefore it is essential that the form language is easily graspable and available. Rather than relying on merely abstract concepts, the form language must have the potential to become a vernacular that is relevant for practitioners.

Knowledge contribution With this thesis, I propose a theoretical approach to form and expressions, which I begin to make meaningful through my own design practice and formgiving process. The thesis is resting on a practice epistemological assumption, in which my own design practice can inform my investigations. Schön’s (1983) theory on the reflective practitioner has been widely used in design as a way to frame the insights produced by practice. Through reflections on my own formgiving, prototyping, and form analyses as design exercises, I unfold the potentials and explore the limitations of the proposed form language.

3

My method for this reflective practice is highly dominated by phenomenology, in which my experimental and designerly experiences are taking centre stage to inform the study. In her book Closer, Kozel (2007) introduces a phenomenological method for a performative setting.

She argues that through reflection of our own performances (or in this case, design practice), we are able to reveal knowledge through even mundane aspects of lived experience (Kozel, 2007, p. 2). The phenomenological method, therefore, allows me to elevate the tacit banalities of my design practice in order to construct meaning about the form language and uncover its potentials and limitations. The purpose of such an inquiry is not only to improve my own design skills, but also to reveal and develop insights that I hope can resonate and seem relevant for other designers as well.

My thesis will not present a final or contiguous account of aesthetics as part of a practice for shape-changing interfaces. Instead, it zooms into aspects of designing computationally actuated objects and begins to explore the dynamics of form and expressions through a design vocabulary. As such, it does not encompass all the complexities of an entire interaction design practice. For instance, less attention is given to the actual interaction with shape-changing interfaces in this thesis. The focus is on the object’s physical form and the temporal changes within it. Furthermore, I do not address dimensions of functionality or user experience – areas that are usually pivotal for successful interaction design – but delimits the thesis to the visceral perspectives on behaving expressions.

Thesis outline This thesis is divided into eight chapters including this introduction. Each chapter seeks to inform the project from different perspectives.

Chapter two, A form approach for shape-changing interfaces, positions this thesis’ form approach in relation to other perspectives on shape-changing interfaces based on related work within interaction design, HCI, and architecture. The chapter discusses the qualities of three current approaches to shape-change: A functional, a user experience, and a material approach. I further argue how a form perspective can be complementary to these other approaches in order to inform a design practice for actuated interfaces. Moreover, special attention is given to the language with which each of the approaches are described, to be able to argue for a need for a dedicated design vocabulary to deal with the expressional dimensions of shape-changing objects.

Chapter three, Behaving expression, explores the theoretical foundations for a form perspective on shape-changing interfaces. Based on writings within interaction design, the

4

Introduction

chapter draws in concepts from architecture and classical aesthetics in order to further qualify both an aesthetic approach as well as an accompanying form language for the behaving expressions of computationally actuated interfaces.

Chapter four present the methodology for exploring how the proposed form language serves as a qualified grasp for an aesthetic inquiry into shape-changing interfaces. The three following chapters then unfold the insights from this methodical approach to shed light on different uses of the form language.

Chapter five, Formgiving experiments, introduces the designerly form experiments I have performed to explore different dimensions of the proposed form language. It is a thematic walk-through of the iterations of the design set-ups and shows how the form language can be used generatively.

As a synthesis of the formgiving experiments, chapter six, Prototype production, presents two prototype concepts through the form language as a descriptive lens. A company has been involved in the manufacturing of the two prototypes and this production process will also be investigated in this chapter to indicate a communicative potential of the form language.

Chapter seven, Descriptive form exercises, is a form analysis of three projects presented in chapter two. This chapter seeks to hint the potential of the form language beyond my own design work. Furthermore, it shows the promising aptitudes as well as its limitations to deal with the form dynamics of shape-changing objects.

Chapter eight, Ramifications of a form approach, further discusses how a form approach is relevant for design practice. In particular, it looks at the form language as a manifestation of a design tradition for shape-changing interfaces, just as it ties the form language together with other aspects of the object, such as material selection. It also discusses how a form approach requires a certain degree of expertise due to the central role of designers in such processes, and moves on to elaborate how this expertise can be developed through collective form studies.

5

Chapter two

A form approach to shape-changing interfaces Interaction design is the design practice of developing interactive products. Thus, the interaction designer must be able to make design decisions regarding multiple aspects of these objects; for instance, functional purposes, experiential qualities, and ethical values (Löwgren & Stolterman, 2004). Thus, interaction design deals with complex relationships between various dimensions that must be beautifully tied together to produce successful designs. Furthermore, design is not defined by a natural logic, where a design solution is given based on the situation we are designing for. Instead, “[d]esign work is given form and structure by designers’ own thoughts, considerations, and actions” (Löwgren & Stolterman, 2004, p. 8). To assist interaction designers’ ability to navigate this complexity, design research continuously seeks to develop new frameworks, strategies, and methods for working with the multiple facets of design.

Additionally, design is in its nature forward-looking; design is not about making descriptions of the existing world, but rather to suggest future improvements (Hove, 2010; Löwgren & Stolterman, 2004): “Design forces us to challenge the present and makes us think about the basic conditions of our society. This holds for interaction design as well as for any other design field.”

(Löwgren & Stolterman, 2004, p. 11) If shape-changing modalities are to become part of future products, we must be able to deal with these different dimensions of the design process surrounding their creation. As Rasmussen et al. (2012) have argued, more design theory about the expressional qualities of shape-changing interfaces is therefore needed.

To develop a strategy for dealing with the aesthetic dimensions of shape-changing interfaces, I want to propose a form approach as a way to address the immediate, expressive dimensions of computationally actuated objects. In this chapter, I wish to lie out the ways that a form approach can complement current research in shape-changing interfaces. Concurrently, I will also point to the limitations of current approaches in order to qualify the need for this new design space within interaction design research.

7

As I will present later, the basis for the form approach is a form language with which we can begin to discuss form compositions and their aesthetic qualities. The purpose of a form language is not only linguistic, but does also serve to help designers bring awareness to their use of expressional means as well as to act as a generative framework for designers to develop new forms: “Designers inevitably and instinctively prefigure solutions to the problems they are confronted with, but the depth and range of their design vocabulary influence both their perception of a question and the shaping of its answer. If one’s understanding of a form language is limited, then the range of possible solutions to a problem will also be limited” (Ching, 2014, p. IX).

At the same time, language is what makes it possible for us to share our internal experiences with others (Wittgenstein, 1968). For instance, my tacit experiences of a visceral form compositions can be shared with others through my articulation of the experiences in either written or spoken language. Wittgenstein argues that our use of language must always be understood holistically in relation to its situated use; as such, language is not private, but exists as a social tool with which we can articulate elusive areas of our individual experiences to others (Ehn, 1989). Wittgenstein further uses language-game as the rules that surround our way of using different words for particular situations. In his discussion on design philosophy, Ehn turns to the notion of language-games to understand how carpenters make meaning of different words for tools and materials: “Understanding the professional language of chair making, as any other language-game, is to be able to master practical rules which we did not create ourselves. They are techniques and conventions for chair making as part of a given practice” (Ehn, 1989, p. 104).

The language-game is the context in which we must understand the words we use. The form language I will propose in this thesis works within in a language-game of aesthetics. To qualify this, it is helpful to look at the language-games following the other research approaches for shape-changing interfaces. Therefore, I will pay special attention to the ways in which current research articulates the dynamics and temporal dimensions of actuated objects. I do this to illustrate how these current descriptive vocabularies fall short in terms of encompassing expressional dimensions of shape-change; that is, that the scopes of their language-games do not include expressional qualities.

8

A form approach to shape-changing interfaces

Shape-changing interfaces Shape-changing interfaces have been conceptualised under multiple concepts, like actuated interfaces, kinetic interfaces, organic interfaces, and even robotics. Poupyrev et al. were some of the first to theorise actuation and changes of shape in interface design and defined it “[…] as interfaces in which physical components move in a way that can be detected by the user” (Poupyrev et al., 2007, p. 206). Based on various projects and their own design, Lumen (figure 2.1), they presented the initial articulation of shape-changing displays as a way to embody digital information in physical transformations.

Rasmussen et al. (2012) have reviewed 44 articles to make a thorough attempt to map the current design space of shape-changing interfaces. They lay out four overall dimensions of shape-changing interfaces to better understand their potentials: 1) Types of shape change, 2) types of transformation, 3) interaction, and 4) purposes of shape change. For example, they identify eight types of transformation: Orientation, form, volume, texture, viscosity, spatiality, adding/subtracting, and permeability (figure 2.2). As a design exercise, Nørgaard et al. (2013) used these eight types of shape change as a framework for students to develop interactive toys. Based on one of the eight types, groups of students were to design interactive sketches of shape-changing tumbling objects. The study showed the generative potential of the framework, but they also note how it may not be ideal to approach shape-changing design based on only one of the categories: “Even though, for the sake of the experiment, we wanted to pursue the types of shape change oneby-one, in more natural design cases, this would probably not be a fruitful approach. However, systematically trying out alternatives, inspired by the breadth of the design space as illustrated by the framework seems to be a very fruitful approach to more diverse and rich shape-changing interfaces” (Nørgaard et al., 2013, p. 259).

Rasmussen et al.’s framework does indeed create an overview of the landscape of shapechange, but the different types can be hard to understand and utilise in one’s own design practice. For instance, Rasmussen et al. use the Thrifty Faucet (Togler et al., 2009) as an example for orientation. The faucet does change its orientation as they state, but at the same time its form changes from bent to stretched states, which means it could also fit in the form category. Similarly, the Morphing Harddisk (Horev, 2006) does indeed change form, but could just as easily be described as changing its volume. The challenge then seems to be how do we decide which category to use to describe a certain change in shape, as the proposed types are not mutually exclusive. I do not seek to criticise the work of Rasmussen et al. (2012), but to argue for a need for a better language to describe actuation in interfaces. As I will argue, it may be better to develop a more basic vocabulary for shape-changes, rather than relying on categorised distinctions.

9

Figure 2.1 Lumen is a basic investigation of how shape-change can be used to communicate information through a physical display Credit: Ivan Poupyrev

Figure 2.2 The eight types of shape-change Credit: Rasmussen et al. (2012)

10

Figure 2.3 SpeakCup is a digital voice recorder that uses its shape to communicate its functions Credit: Jamie Zigelbaum

Parkes et al. (2008) discuss how kinetically actuated interfaces could learn from comparable disciplines like kinetic art or robotics in order to qualify this category of interfaces. Among other things, they argue how we need a basic language for describing kinetic movements and motion based on descriptions of the spatial motion of the individual elements in the interface and how this description should be based on parameters such as speed, direction, spatial extent, and similar (Parkes et al., 2008, p. 63).

In order to develop this basic vocabulary, they argue how shape-changes cannot simply be approached from a purely technological approach, but must be investigated outside the boundaries of HCI domains and include other expertise from e.g. robotics, haptics, design, and architecture (Parkes et al., 2008, p. 65). Similarly, Zigelbaum and Labrune (2009) also point to the lack of a physical language as one of the key challenges for successful design practices for shape-changing interfaces. In their article, they present the simple design concept SpeakCup (figure 2.3). Based on the quite complicated process of designing their rather simple interface, they argue that the design of advanced functionalities, like those found in an audio editing software, will be highly complex to embed into a shape-changing interface. While I do not believe the purpose of shape-changing interfaces should necessarily be the same as for screenbased software, I do accept that interaction design must indeed find a way to “[‌] scale up complex interactions with dynamic physical materialsâ€? (Zigelbaum and Labrune, 2009, p. 4).

11

Both Poupyrev et al. (2007, p. 209), Rasmussen et al. (2012, p. 741), and Parkes et al. (2008) all outline multiple types of approaches for research within shape-changing interfaces.

In the following, I wish to look deeper into three of these: 1) A functional approach, 2) a user experience approach, as well as 3) a material approach. From this review of the current research and design practice landscape of shape-changing interfaces, I wish to uncover a form approach as a yet not so explored perspective, which I strongly believe will help evolve the shape-change vocabulary.

A functional approach to shape-changing interfaces A large body of research within HCI and interaction design has investigated functional purposes of shape-change. This type of research often focuses on how shape-changes can be used to communicate information in an efficient manner, how dynamic affordances can be used to communicate interaction possibilities, or how haptic or tactile feedback can be used to improve system usability (Rasmussen et al., 2012, p. 740). In general this type of research has functionality and technology as its main aim and technological advances are thus its end goal.

In the following, I describe three research projects to exemplify this line of research.

Naturally, a lot of interest has been on how shape-change can be used as a new modality in mobile phones. One study (Hemmert et al., 2010) has investigated how precisely a tilting interface in a mobile phone can be used as a haptic feedback system. Another study (Roudaut et al., 2013) has further explored this in a more elaborate manner. Together with their series of prototypes called Morphees (figure 2.4), Roudat et al. (2013) propose the concept of shape resolution as a way to describe shape-changing interfaces. The shape resolution concept understands form as a mesh of physical control points, similar to how a display is made of pixels. They further present a set of features of shape resolution to articulate as the ways in which the mesh can deform, e.g. area, curvature, or speed. The framework goes beyond specific types of shape-change towards a more general way of describing shape-changing forms. While I believe this is a step in the right direction for the development of a shapechange language, the complexity of the shape resolution framework makes it more appropriate for form analysis rather than as a generative vocabulary for design practice.

Other domestic appliances have been explored besides the mobile phone. The Thrifty Faucet (figure 2.5) is an experimental design to explore the potential of shape-changes in everyday objects. In their paper, Togler et al. describe Thrifty Faucet in the following way: “The Thrifty Faucet is constructed around a stiff yet bendable plastic tube. Seven plastic rings mounted to its outside are used to guide three pairs of steel wires that, when exerted a force on,

12

A form approach to shape-changing interfaces

Figure 2.4 Morphees is a prototype series of actuated mobile interfaces Credit: Roudat et al. (2013)

bend the tube […] The very characteristic, organic movement is here achieved by allocating the force on many hinges, so that normally the whole body is affected, also by a single movement” (Togler et al., 2009, p. 43) To describe the faucet’s movements they articulate it as compositions between a material with certain properties (i.e. a bendable plastic) and a mechanic system (i.e. the steel wires). To describe the shape-changes, they explain how a force exerted on the steel wires will result in bending movements. Based on this description they show how the faucet in its entirety appears organic. This way, they use a formal description of the relations between individual elements and how this can be affected through the exertion of a force on given points, to indicate how the faucet as a complex whole has certain expressional qualities. This seems as a comprehensible way to articulate shape-changes and is easier to adapt to design practice compared to the previous abstracted form vocabularies.

Topobo (figure 2.6 top) is an assembly toy for children. While it does not serve the same functional purpose as the previous projects, it is still included in this section as it “[…] can help children to understand certain physical principles affecting kinematic systems” (Raffle et al., 2004, p. 654). As such, Topobo is not a purely hedonic toy for free play, but a system to support educational goals. Users can assemble the pieces of Topobo to make it act in certain ways. Based on the Topobo project, Parkes and Ishii (2009) have developed Kinetic Sketchup, 13

Figure 2.5 The Thrifty Faucet is a living interface and explores actuated home appliances Credit: Jonas Togler

which is a series of physical modules with distinct properties that can be combined in any way to sketch physical prototypes and motion of design ideas. As with the Thrifty Faucet, the focus here is on how individual parts with distinct properties are combined into a whole. The prototyping tool is based on a design language for articulating movement and transformative properties of interfaces (figure 2.6 bottom). Instead of focusing on a purely mechanical way of describing motion, Kinetic Sketchup is based on how the categories are perceptually different. This focus on the way motion is expressed in a material seems as a fruitful approach to making an easier adaptable vocabulary for design practice. As their language is tied closely together with the physical modules of the prototyping tool, it is relevant to develop a more general language independent of specific technologies that can be utilised in design practice for describing shape-changes.

As these projects illustrate, the strength of the functional approach is its closeness to application. It demonstrates how shape-changes can be used to obtain well-known applications and uses of technology. In turn, this also limits its scope to the confinements of our current understandings of technology and ways of using it, rather than imagining radically new potentials for shape-change. Also, it is not concerned with the user’s perspective, which another line of research has instead begun to investigate.

14

Figure 2.6 Topobo and its kinetic sketch up language Credit: Parkes and Ishii (2009)

A user experience approach to shape-changing interfaces In recent years, the user experience of shape-changing interfaces has received increasing attention within a number of studies. This line of research is characterised by an assumption that if we are to develop meaningful actuated interfaces, we must better understand perceptual and experiential qualities of such modalities (Grönvall et al., 2014; Kwak et al., 2014; Pedersen et al., 2014; Rasmussen et al., 2013). The scope is not only to use technology to expand knowledge about it in itself, but also to use it as a means to better understand the potential relations between humans and shape-changing interfaces.

To better understand the user experience of shape-changing handheld devices, Pedersen et al.

(2014) have evaluated 51 videos of different versions of a shape-changing smartphone with 187 participants. In an online questionnaire, the participants are asked to rate these versions on different pragmatic and hedonic parameters, like predictable/unpredictable or tacky/ stylish. The aim of the study is to bridge the gap between research in modelling shape-change and studying user experience. Their approach is to alter parameters in the shape-change video models to understand how these changes are perceived. While the study may be of importance for technical and engineering aspects of developing shape-change, it seems less viable for a more holistic understanding based on qualitative measures, due to its quantification of design parameters.

More similar to my perspective, Rasmussen et al. (2013) have begun studying shape-changes because “[…] It is a necessity to move beyond the (understandable) technological fascination, towards understanding Actuated Interfaces’ strengths and weaknesses from a human perspective” (Rasmussen et al., 2013, p. 63). Based on three design cases, they explore how people experience shape-changes. They find that actuation can be used to design for rich experiences as people ascribe multiple interpretations to the designs. For example, Rasmussen et al. describe how some experienced the designs as magical. On one of their design cases, the Aerial Tunes (figure 2.7), they note how people experience a “[…] breach of our familiar ‘physical reality’ […] by the ball hovering in mid-air seemingly defying gravity” (Rasmussen et al., 2013, p. 69). As such, shape-changes allow for new and engaging experiences. However, as the authors also note themselves, this perceived magical experience may also be the result of exposing people to novel and unknown technologies. It may be that this magical sensation will fade as this type of technology and temporal dynamics become more common in our society (Vallgårda et al., in press).

The coMotion bench, which is also one of the three design cases from the study above, has elsewhere (Grönvall et al., 2014) been used to begin to understand how people make sense of shape-changing interfaces. By expanding McCarthy and Wright’s framework for sensemaking processes, they embrace both sensory and cognitive means of making sense. By 16

A form approach to shape-changing interfaces

Figure 2.7 The magical feeling in Aerial Tunes results in an engaging experience for spectators Credit: Akosta.dk

placing the bench in three different contexts, they investigate how shape-changing interfaces are experienced “in the wild” – that is, outside a lab and in a real context – as a key to making meaningful shape-changing interfaces.

As the body of research in shape-changing interfaces grows, Kwak et al. (2014) remind of the importance of including the user’s point of view as an integral part of the design space. Based on six abstract interfaces with similar physical form, they interviewed 11 participants to “[…] identify the constructs that are useful for describing end-user’s experience of interacting with shape-changing interfaces and in doing so provide an empirically grounded characterization of shape-changing interfaces” (Kwak et al., 2014, pp. 184–185). Their analysis is build upon deriving consensus and making categorisations based on the interviews with the participants.

In this sense, they sought to normatively describe people’s understandings of shape-changing interfaces.

The projects described here all focus on people’s articulations and reflections on their experiences. These studies capture the more cognitive dimensions of how people perceive this type of technology. While this is indeed important for the advancement of relevant and meaningful interactions with shape-change, explorations of more immediate sensations may further help to inform aesthetic and bodily dimensions of engaging with actuated interfaces.

A form approach will help to inform the understanding of the experiential qualities through a focus on more visceral and immediate dimensions of shape-change.

17

Figure 2.8

The soft mechanical alphabet proposed by Coelho and Zigelbaum (2011)

Shape-change as a material property While the projects described above use a variety of mechanics and technologies to produce actuation as well as beginning to understand how users make meaning of such interfaces, one line of research focuses more specifically on shape-change as an inherent property of “[...] materials that undergo a mechanical deformation under the influence of direct or indirect electrical stimuli� (Coelho and Zigelbaum, 2011, p. 162). For instance, a shape memory alloy (SMA) is a metal alloy that can deform its shape when heat is applied to it and return to its original shape when cooled. In these materials, actuation is an additional core property on par with the static properties we find in conventional materials.

Coelho and Zigelbaum (2011) argue how these types of materials are yet a rather unexplored area due to technical challenges and a lack of knowledge of its potentials. To remedy this they have made notable efforts to bring shape-changing materials into HCI and interaction design, by beginning to describe their properties for this context. They present a soft mechanical alphabet for working with actuating materials. Based on two basic ways a material can deform, compression and elongations, they propose a vocabulary for form transformation. As it is evident in figure 2.8, Coelho and Zigelbaum base this soft kinetic alphabet on an indication of conceptual lines onto which forces of either elongation or compression are applied. It is therefore possible to describe form transformation through these simple means without loosing the breadth of possible expressions.

18

Figure 2.9 PneUI is created from silicon material and can change shape when air pressure is applied to it Credit: Yao et al. (2013)

Figure 2.10 A polymer material is in ShapeShift to create actuated architectural surface Credit: caad-eap.blogspot.dk

A project that focuses on new actuating materials is PneUI (Yao et al., 2013). This project explores how to utilise pneumatically-actuated (actuated by air pressure) materials to create soft and organically shape-changing interfaces. The PneUI uses a composite silicon material that is created with multiple layers, which gives it its shape-changing properties when air pressure is applied to the air channels within it. The framework for the project is defined based on Rasmussen et al.’s (2012) and the soft kinetic alphabet (Coelho and Zigelbaum, 2011) described above. For instance, Yao et al. (2013) have created a shape-shifting lamp (figure 2.9) with which they explore surface curvature. Based on the positions of internal airbags, which can be either compressed or elongated, the surface will start bending and eventually curl.

ShapeShift (figure 2.10) is an experimental project at an architectural scale that explores the potential of an electro-active polymer (Kretzer, 2010). While it has not been made in a HCI or interaction design research context, it is introduced here to show a range of smart materials. The material used here is a thin polymer that converts electric power into an internal mechanical force. ShapeShift is made from an architectural approach and is explored to develop responsive environments and surfaces. Architecture is an area that is currently beginning to build design practices around actuated surfaces.

Besides developing new material properties, these projects also begins to explore new expressional potentials of such materials. They take a step towards aesthetics and have less emphasis on the functional aspects of shape-change. However, we have yet to fully understand how these material properties can be utilised in interaction design practice.

20

Figure 2.11 The computational composite PLANKS Credit: Anna Vallg책rda

A step towards such an understanding has been proposed by Vallgårda and Redström (2007), who have argued that computers can be seen as a compositional material with distinct material properties, such as temporality and computed causality (Vallgårda and Sokoler, 2010).

These properties can be brought to use when combined with other materials in which the computer’s material properties can be expressed. While computational composites are not limited to actuation, PLANKS (figure 2.11) is an example of a shape-changing composition. It is made from two-meter long planks that can bend outwards as a result to sound input: “The PLANKS composite comprises a negotiation between each element (or material) and it is this negotiation, which makes up the properties of the composite. Properties which are different from the sum of the properties of the parts—some are restrained (the computer) and others are challenged (the pine planks)“ (Vallgårda, 2008, p. 573).

The concept of the computer as a composite material turns the focus towards understanding the compositions and the holistic relation between the properties of the individual parts.

The material approach to shape-change is a promising and interesting way of exploring new uses of actuation. Rather than staying within the confinements of traditional modern technology, which often focus on improving and making it more efficient, these projects take a step back and examine new potentials and expressions of shape-change. They are not driven by functional aims, but by the advancement of new material properties. To exemplify this, I will briefly present two works that have utilised actuation as a hedonic quality.

Outside the research community, artists and architects have begun exploring these new material potentials in shape-change to develop new art practices. Art and architecture have embraced shape-change as a hedonic strategy for producing new artistic expressions. The purpose of these kinds of projects is not to develop technology, but to use technology to produce dynamic expressions. In this sense, they come closer to the aim of a form approach, as they incorporate these new modalities in their design practices.

Lotus Dome (figure 2.12) is an interactive art installation by Dutch artist Daan Roosegaarde.

It is a spherical dome created from “[…] hundreds of smart flowers which fold open in response to human behaviour” (Studio Roosegaarde, 2011). It is described as an organic interface that can change its permeability depending on human presence. Inside the dome is a rotating lamp that reacts to the position of people. The flowers in Lotus Dome are created with a heatsensitive smart foil and they will curl as the warm light from the lamp illuminates them.

On a larger scale, the Thematic Pavilion (figure 2.13) by soma architects uses shape-change as a way to incorporate the dynamics of the building’s surroundings: “Towards the sea the conglomeration of solid vertical cones defines a new meandering coast line, a soft edge that is in constant negotiation between water and land” (soma architecture, 2011).

22

Figure 2.12 The Lotus Dome reacts to human presence by opening its hundreds of smart flowers Credit: Studio Roosegaarde

Figure 2.13 The Thematic Pavilion uses shape-change to depict the dynamics of the sea Credit: soma architecture

A form approach to shape-changing interfaces

While the purpose of this thesis is not the create art or inform architectural practices, it is relevant to investigate how other practices are working with shape-change as a more integrated aspect of the overall design. Rather than focusing on its technology, shapechange is understood as part of a complex, similar to Vallgårda and Redström’s proposal of computational composites. They are therefore included here as exemplary to how interaction design could be inspired to include expressional dimensions of shape-change to its practice.

What is then needed to further develop the material approach towards design practice is moving beyond a sole focus on the material properties themselves, and rather begin to understand how such properties can be used in computationally actuated objects. A form approach does not dissociate itself from the focus on the expressional dimensions that work within smart and computational materials has established. Instead, it seeks to complement this work by offering a way to utilise these material properties in more elaborate form composition. A form approach focuses less on the specific materials, but emphasises the composition of properties. It is of less interest whether actuation is a property within the material or if it is an effect produced by compositions of both traditional materials as well as computational technology (e.g. motors).

Design space proposal: Shape-change as aesthetic form expressions Throughout this chapter we have seen how there have been many different attempts across research approaches to define a vocabulary for describing shape-changes. Parkes et al. (2008) began to explore the spatial motion of the individual elements of the Topobo prototype.

Togler et al. (2009) further advanced a similar approach but with a hermeneutic focus on how forces applied to single elements in the design resulted in an overall expression. This attention to the composition of elements is also the underlying principle for Vallgårda and Redström’s (2007) work on the computer as a material that is brought together with other materials. Another approach to a language for shape-change is Roudat et al.’s (2013) notion of shape resolution, where a surface is understood as a mesh of points that can be manipulated to make topological changes. To describe actuated materials, Coelho and Zigelbaum (2011) uses conceptual lines that result in shape-change when forces are applied to them.

While these attempts have enabled a more qualified discussion of actuation, a more basic and general language is needed to further strengthen shape-change literacy in interaction design. Moreover, current studies in the user experience of shape-change has focused on more cognitive dimensions of meaning-making and thus a focus on more immediate and sensory levels could potentially inform the field further.

25

In his book Emotional Design, Norman (2005) distinguishes between three dimensions of how people reacts emotionally to objects: 1) The Visceral level (the immediate, sensory response to the object) , 2) the behavioural level (the experience of the functionality of the object), and 3) the reflective level (a reflection on the social or cultural meanings of the object).

The visceral level is connected to more subconscious ways of being in the world and focuses on our immediate sensory attraction to objects and their form (including shape, material, colour, and texture). While attention must naturally also be on the behavioural and reflective levels, a focus on the visceral experience of form is relevant for shape-changing interfaces in order to improve a basic understanding of the expressional qualities of actuation.

With this thesis, I wish to unfold a design space that focuses on the immediate and sensory experiences of shape-changing interfaces. With a focus on shape-changing interfaces as forms, we can begin to explore these visceral expressions as well as unfolding a potential vocabulary for this. The design space is in its essence an aesthetic inquiry into understanding the basics of shape-changing form compositions.

I use the concept of a design space to hint to the open-ended nature of a form approach, and because it entails a model that acknowledges design work that is not problem-oriented, but is instead used in knowledge generation: “This model claims that all design work supports the understanding of the ‘design space‘. This means that all the different methods and techniques used during the design process will result in some knowledge about the design space” (Westerlund, 2005, p. 1) Besides contributing with a way to deal with the visceral aesthetics of shape-changing interfaces, a form approach also seeks to uncover new potential applications of shape-change.

In their paper Abstract Information Appliances, Hallnäs and Redström propose aesthetics explorations as a way to develop new functionalities and purposes of technology: “[T]he idea that ‘function resides in the expression of things’ suggests that we could try to discover functionality in expressions and rediscover the hidden aesthetical choices in the expressiveness of things in use“ (Hallnäs and Redström, 2002a, p. 108).

The proposed design space is part of a growing interest in interaction design on form and expressions (Hallnäs et al., 2001; Hallnäs and Redström, 2002a; Jung et al., 2010; Jung and Stolterman, 2011; Mazé and Redström, 2005; Vallgårda, 2013), which has been conceptualized as form-driven interaction design that “[…] can enrich current user-centered and functionoriented perspectives in HCI by incorporating more of a designerly approach with emphasis on material, meaning, and making of form” (Jung and Stolterman, 2011, p. 406).

26

Chapter three

Behaving expressions As described in the previous chapter, the purpose of a form approach is both to draw attention to the visceral form dynamics of computationally actuated objects, and to use form explorations as a driver for developing new purposes and functionalities for shape-changing interfaces. In this chapter, I wish to articulate the theoretical foundations for a form perspective for the behaving expressions of shape-change. Moreover, I will propose five basic elements of a form language as a means for dealing with shape-changing form and expressional dynamics.

The first part of the chapter looks at the computational object as form compositions and lays out its constituting form elements. It describes how these form compositions can be regarded as abstract expressionals, with which we can uncover new functionalities and purposes for actuated objects. I will then present the basic elements of a potential form language for shapechanging interfaces. The last part of the chapter positions the form approach as a formgiving practice and draws the contours of how to embrace such practices in a designerly way.

The form gestalt of computational objects The physical form of a shape-changing object is, like any other object, the physical configuration of both its three dimensional shape, the materials from which it is made, its colour as well as other characteristics that defines its appearance. The form is the way an object exists in the world; it is, the holistic composition of its entire gestalt. However, shape-changing objects are different from many other objects due to their ability to dynamically change their shape.

As such, their forms are not static, but changes as a result of their inherent computational capacities.

Traditionally, interaction design has had less focus on the physical form dimensions of interactive artefacts since most of these were confined to computer screens. However, ever since computational technology has taken a leap beyond the screen and is now embedded into all kinds of settings and products, form and materiality have also started to receive more attention. While interaction design has yet to develop a strong form tradition, some work has been done in applying a form perspective to computational objects (cf. Baskinger and Gross,

27

2010; Hallnäs et al., 2001; Hallnäs and Redström, 2006, 2002a; Jung and Stolterman, 2011; Mazé and Redström, 2005; Vallgårda, 2013). For instance, Mazé and Redström (2005) wish to re-establish a focus on the designed object itself. This is especially the case for objects that are designed with “[…] ‘surfaces’ that extend beyond three-dimensional form, where the main expressiveness comes with the temporal form elements” (Mazé and Redström, 2005, p. 16) as it is the case with shape-changing interfaces. They believe that a form perspective will help designers to navigate the complexity of designing computational objects. Additionally, they stress how interaction design is often concerned with designing unknown types of objects.

Interaction designers rarely design commonly known objects, like a chair or a teapot, but do instead explore entirely new categories of objects, for which it is not possible to rely on wellestablished uses or traditions (Mazé and Redström, 2005, pp. 8–9). Mazé and Redström argue how a strengthened form literacy can enable interaction design to navigate these unfamiliar design spaces, by shifting the attention towards the physical composition of the objects, rather than focusing on its purposes and potential experiences. In their paper, they argue how discourses within design have come to focus more on the experience of an object instead the object itself. As such, their proposition is a shift in focus from process (of experience) to the designed object (Mazé, 2005, p. 7). However, a focus on form will inevitably also include a process. But instead of being a process of meaning-making for the user, it is a process of defining form by the designer; that is, a form approach entails a formgiving process.

Since computational objects, including shape-changing interfaces, are not static entities, interaction design cannot simply rely on classical notions of form. Instead, Baskinger and Gross (2010) have argued that interaction design must develop its own form literacy in order to truly achieve relevance beyond design of screen-based solutions. They further argue that understanding how to design the computational object’s form is an essential aspect of interaction design, as it can be used to support and guide the user’s interaction possibilities with computational objects, just as the form “[...] visually signals and physically embodies functionality” (Baskinger and Gross, 2010, p. 8).

To improve the form literacy in interaction design, Vallgårda (2013) has proposed a trinity of form elements in the practice of interaction design to address the design of computational objects – see figure 3.1. The physical form is the three-dimensional shape of the object, its material, scale, texture, colour, etc. The temporal form “[…] is the pattern of the state changes that the computer will produce” (Vallgårda, 2013, p. 2). Finally, the interaction gestalt is the ways in which a user can interact with and affect the object.

For shape-changing interfaces in particular, their temporal forms are expressed as physical form changes. The physical form is not simply a container for the computation, but is what enables computation to be expressed at all. Shape-changing interfaces are by definition

28

Behaving expressions

Physical form Temporal form Interaction gestalt

Figure 3.1 The three form components in interaction design: Physical form, temporal form and interaction gestalt. The shaded area of the figure maps the main focus in this thesis; that is, the relation between physical and temporal form.

dependent on a physical form in which spatial changes can occur. As such, there is a direct relationship between an objects physical and temporal form and it is therefore pivotal that we improve our understanding of how they can be combined in meaningful compositions.

The interaction gestalt with shape-changing interfaces is in turn the ways in which a user can affect the temporal – and thus also the physical – form.

VISCERAL NARRATIVES On a visceral level, we have an emotional relation to objects based on their form compositions.

For instance, we feel that one flower vase is more beautiful than another. Maybe it is because of its materiality or maybe it is because of the way it is shaped. In her report for the Danish Design School, Hove (2010) argues that this is due to the fact that objects tell a certain abstract story about themselves through their form composition. To exemplify, she describes two desks.

On a functional level, they both serve the same purpose. A person can sit by it and do some kind of work. However, while the functions of the tables may be similar, the appearances of the desks can be different; one table may appear light and dynamic, whereas the other may seem heavy and stable. Each of the tables has a certain expression, which is a result of its form composition as a whole.

Similarly, shape-changing interfaces also tell abstract stories, or have visceral expressions, depending on their form compositions. And since the form of these objects is not static, the 29

expression will neither remain the same, but will behave according to the temporal changes in the object’s form.

As it has been noted by Hallnäs and Redström (2002a, p. 105), it is not that decisions about the aesthetics of computational objects will not be made without a form perspective. However, awareness of the expressional dimensions will allow us to more thoughtfully design form compositions that entail desirable expressional narratives of the object.

One of the basic principles in the introductory courses at the Bauhaus school was to teach students to attain a heightened awareness of the compositional means that are available to produce certain aesthetic expressions (Itten, 1975). In his book about the basic course about design and form, Itten presents the foundations for his teachings at the Bauhaus. The purpose of his teaching was to nurture the creative talents of his students and present them for the principles of creative composition (Itten, 1975, pp. 7–8). Through analytical and compositional exercises, the students were to gain an understanding of the aesthetic qualities of different form compositions in order to qualify their design skills. The basis for many of these exercises was to make the students work with contrasting expressions (figure 3.2), which they then had to study in detail. Through multiple sketches and attempts, the students began to inform their own design intuitions and thereby got a sense of the compositional rules for combining different elements. In this sense, it trained their holistic view on how the individual parts influenced the complex whole of the composition to qualify their sensibility towards the visceral narratives of different compositions.

In this light, this thesis can be said to be a study of the expressive potentials of shape-changing forms (that is, physical forms that encompass temporal changes). And through this, the purpose is then to bring attention to form literacy as a key skill for a better understand the visceral, aesthetic dimensions of shape-changing interfaces.

ABSTRACT EXPRESSIONALS While one aspect of the form approach is to bring attention to the designer’s expressional means to convey visceral narratives through the object’s form composition, a form approach will also allow to explore new applicational purposes of shape-change. Through their work on an aesthetic approach to computational technology, Hallnäs and Redström (2002a) wanted to refocus attention towards the expressions of computational objects. With the notion of the function-expression circle, they introduced a lens for understanding the complex relationship between functionality and form. On one side of the circle, an objects form is determined by its functionality. On the other side of the circle, the expression of an object is what communicates its functionality. The circle seeks to articulate the dialectics between a form’s functionality and its form.

30

Figure 3.2 Examples of form studies from Itten’s course in basic composition Credit: Itten, 1975

They also proposed to use the distinction as a methodological exercise to not only derive the form to support specific functionality but also to explore potential functionalities from different expressions through their leitmotif “function resides in the expression of things” (as a counter-leitmotif to the established “form follows function”) (Hallnäs and Redström, 2002a, p. 3).

In fact, Hallnäs and Redström have on multiple occasions addressed the notion of expressions of computational objects (Hallnäs et al., 2001; Hallnäs & Redström, 2006, 2002a, 2002b).

Throughout their writings they advocate for an aesthetic approach to computational technology based on a material understanding of the computer. By focusing on the expression of objects, they shift the attention from the function of objects towards an appreciation of our relationship to the object on a more phenomenological level, or as they write: “A thing always presents itself through its expression. The expressions of a thing are its pure appearances as we disregard–or ‘bracket’–functional and existential definitions. It is what defines the thing as an abstract expressional, a bearer of the properties of expressions that are invariant across the many different existential definitions, that is an expression-identity. Similarly to how we may thing of a thing as an appliance–a thing designed to perform certain functions–we may think of the bearer of this expression-identity as an ‘expressional’–a thing that is designed to be the bearer of a certain expression” (Hallnäs and Redström, 2002b, pp. 112–113).

The scope of a form approach is therefore also to explore shape-changing forms as abstract expressionals, from which, we can then begin to investigate the potential, new applications and uses of such expressions. A form-approach allows us to explore the possible expressional potentials of shape-changing interfaces in order to broaden the design space and to derive novel and new ways of utilisting it in design objects.

A form language for shape-changing interfaces Hove (2010, p. 8) argues that a form language is essential for designers to be able to distinguish nuances in form compositional differences and to understand the expressional means we have available for deciding on a given form. We need a form language for articulating compositional qualities and to bring awareness of the designer’s form decisions, in order to qualify a form approach in interaction design to shape-changing interfaces. Other design disciplines, like industrial design and architecture, have spent years developing a form language for their design practices to help them qualify their form work. Their form languages consist of both basic form elements but also compositional rules and principles as well as theories for e.g.

32

Behaving expressions

use of colours and materials. These design disciplines have developed their form traditions through years of refinements, explorations and reflection. However, Baskinger and Gross (2010) argue that interaction design has yet to commence a vocabulary for form compositions.

Consequently, a form language for shape-changing interfaces does not yet exist either.

In architecture, form can be described by the use of four primary elements: Point, line, plane, and volume; see figure 3.3 (e.g. Ching, 2014). These elements are the basic vocabulary that the architect can use to develop more elaborate forms. While the form language of architecture is more complex than those four elements, the identification and study of this basic vocabulary enable the architect to qualify his or her design practice as it increases awareness and broadens the architects ability to image new design possibilities. Ching even argues that the study of form should be one of the initial focal points for any new architect: “The analogy may be made that one must know and understand the alphabet before words can be formed and a vocabulary developed; one must understand the rules of grammar and syntax before sentences can be constructed; one must understand the principles of composition before essays, novels, and the like can be written. Once these elements are understood, one can write poignantly or with force, call for peace or incite to riot, comment on trivia or speak with insight and meaning. In a similar way, it might be appropriate to be able to recognize the basic elements of form and space and understand how they can be manipulated and organized in the development of a design concept, before addressing the more vital issue of meaning in architecture” (Ching, 2014, p. IX) Similarly a form approach to actuated interfaces is about identifying the primary elements of shape-changing forms in order to understand the various potentials and from here proceed to understand how these can be used in interaction design. In Wittgensteinian concepts, a form approach for shape-changing interfaces attempts to propose a new language-game, in which the use of a design vocabulary qualifies the understanding of expressional dimensions of computationally actuated objects. A form language is not in itself what will further develop form literacy about shape-changing interfaces. Instead, it is the way the form language is given meaning through practical use: “One cannot guess how a word functions. One has to look at its use and learn from that” (Wittgenstein, 1968, para.340). As such, we must begin to use the form language to expand its meaning for the language-game of shape-changing expressions in order to inform interaction design practice. First, however, we must consider the concepts relevant for such a form language.

The primary elements of form in architecture are not sufficient when it comes to describing the entire form gestalt of shape-changing interface as these are also composed of temporal dynamics that may be affected by some kind of human input. The form vocabulary of shapechanging interfaces must encompass other elements to sufficiently describe them.

33

Figure 3.3 The four primary form elements in architecture: Point, line, plane, and volume Credit: Francis Ching (2014)

Behaving expressions

By definition, the temporal form in shape-changing interfaces results in spatial changes in the physical form. In the previous chapter, we saw how some projects used mechanical setups with motors to create shape-changing interfaces, while other projects had developed materials with an inherent ability to change form. Despite the different actuation strategies, all projects were based on the idea that they could be computationally controlled. Depending on the set-up, the temporal form will be shaped accordingly; a set-up based on motors will need a temporal form (that is, a program) that can control the capabilities of the motor, whereas a pneumatic set-up will require a temporal form that can alter the air pressure.

Regardless of set-up, the interest here is not on the temporal form in itself, but rather how the temporal form translates into physical changes in the shape-changing interfaces. This does not entail that interaction designers should not be able to develop temporal forms, but from a form perspective, temporal forms are more interesting in the way in which they enter into compositional relationships with the rest of the object’s gestalt. Consequently, I propose that a form language for shape-changing interfaces deals not only with the physical and the temporal form separately, but also focuses on a common vocabulary for the way in which the temporal dynamics are expressed in the physical form.

As we have seen, architecture is able to describe physical forms based on the four primary elements: Points, line, plane, and volume (Ching, 2014) – once again, see figure 3.3. I propose that we adopt these basic form elements from architecture to describe the physical form of shape-changing interfaces, and combine these four with a fifth element of force to encompass the temporal dynamics within the physical form. The basic elements for a form language for shape-changing interfaces is the point, line, plane, volume, and force (figure 3.4).

The point is a position in space that has no direction, spatial dimensions or dynamics. A point can serve to mark different intersections between elements, the center or the end of an object. The expansion of one point to two points produces a line, which has a beginning and an end, which gives it a length as well as a direction. A line can be used to describe the path of a movement, a conceptual line within an object or to articulate the edges of a shape. Two non-parallel lines form a plane. Planes can be used to describe the two dimensional shape of an object. Besides length, the plane therefore also has width. This gives it a spatial expanse, which allows it to hold properties of colour, texture, or patterns. A plane extended becomes a volume, which has both width, length and height. It has a three dimensional extent and is what encases an object’s form.

Both Togler et al. (2009) and Coelho and Zigelbaum (2011) have used the concept of force to describe how a temporal structure can alter the physical form of an interface or material. In physics, a force can cause changes to an object. It is denoted as a vector, which gives it an entry point on the object, a size that describes its strength, as well as a direction. Therefore, forces

35

The point

The line

The plane

The volume

The force

Figure 3.4 The proposed basic elements in a form language for shape-changing interfaces From top to bottom: Point, line, plane, volume and force

Behaving expressions

Figure 3.5 Multiple forces are in play in Spatial Sounds, e.g. gravity, a force the counters gravity in the steel construction, as well as the forces involved in the circular movement. However, from an perceptional perspective only one force seem relevant Credit: Edwin van der Heide

contain information about both where on the object changes occur, the magnitude of the changes in the object, as well as the orientation of these changes; all attributes that are relevant for describing interactive objects as well. However, I do not in any way advocate for a physics perspective on the concept of force in design. Instead, I believe a more perceptually driven use of force is more easily adopted in design practice. For instance, a physicist would describe multiple forces applied to an object that are not relevant to design. Instead, we should focus on resultant force (that is, the sum of all force vectors on an object; see Young and Freedman, 2004) or maybe more intuitively, the perceived force. For example, a physicist would describe multiple forces in the art installation Spatial Sounds by Edwin van der Heide (figure 3.4).

While Spatial Sounds is not a shape-changing interface I use it as an example as it includes a multitude of different forces from a physics perspective; that is, a centripetal force, gravity, rotational force and others. For design, however, it is not relevant to consider all these forces at play in the installation; instead, we are only interested in the experienced force that drives the speaker around in its circular motion in order to describe and comprehend its temporal dynamics. The force therefore helps us to denote the transformations that occur within the object’s shape.

It is important to note that we cannot simply adopt design theory from architecture.

Interaction design and architecture are two different disciplines with each their perspectives and traditions. For instance, interaction designers most often deals with rather different 37

physical scales than architecture; they focus on objects, whereas architecture is primarily preoccupied with buildings. Therefore, practices and theories relevant for architecture are not necessarily helpful for interaction designers.

In this sense, the proposed primary elements are not yet relevant for a form language for computationally actuated objects. This is only possible if they can be shown to have relevance for practice, which is what I will seek to do in the remainder of this thesis. The elements do not make up an entire form language. If they succeed as the primary form elements, we must then begin to explore them in order to develop a form tradition that entails a deeper understanding of interactive forms. Moreover, we must also elaborate on other aspects of a form language; for instance, the compositional principles for shape-changing interfaces.

Giving form to computationally actuated objects In this chapter, I have argued for the purposes of a form approach as well as proposed the primary form elements of a form language for shape-changing interfaces. Lastly, I want to draw the attention to how interaction designers can embrace a form perspective in their work through formgiving.

Formgiving is the practice of arranging and creating the shape of an object (Hallnäs and Redström, 2001; Vallgårda and Sokoler, 2010; Zainal Abidin et al., 2008). It has been introduced in interaction design on multiple occasions. Some of the first to use the concept of formgiving were Smets et al. in 1994 that used it as a way to understand how visual forms in graphical interfaces convey information. Based on three case studies from industrial design, they explored designers’ ability to create forms that convey non-obvious, complex information, e.g. flavour-inspired packaging for desserts or interpreting unfamiliar scents into physical forms. They sought to draw attention to the visceral narratives of graphical elements in order to bring awareness of how this could be better utilised in the design of user interfaces.

Ten years later, after the rise of tangible computing, formgiving was used as a lens in interaction design “[…] as the way in which objects appeal to our senses and motor skills” (Djajadiningrat et al., 2004, p. 1). Instead of relying on abstract mappings and semantics to create meaningful interaction possibilities for the users, they argued for a direct approach where meaning is created in the actual interaction with the object. With strong emphasis on affordances (in a Gibsonian sense) and bodily skills, they advocated for a formgiving practice with focus on creating meaning-carrying action possibilities. For instance, they designed a digital camera to explore how an everyday information appliance could be manipulated through a physical interface and not through the typical screen-based menu structures (figure 3.6). The point

38

Figure 3.6 A digital camera redesigned based on meaning-making in the actual form of the object Credit: Djajadiningrat et al., 2004

here is that they used formgiving to draw attention to the richness of the material world and the potentials it offers for interaction design. Djajadiningrat et al. focus on information appliances and they do not exceed the domain of function-driven design. Adopting their view on understanding materials to design for bodily richness, this thesis further wishes to readdress the purposes of technology and shape-change. Instead of simply making the currently available functions and information available in new forms, there is a potential in re-imagining the ways in which technology can be part of our lives.

This is also a central aspect in Vallgårda and Sokoler’s (2010) article on a material strategy for interaction design where they write: “[T]he expression or the form is pivotal to the functionality and [one] cannot be designed independently of the other. Thus, the material strategy leaves both form and function in a state of ongoing negotiation throughout the design process. The strategy is thereby closely linked to the craft-related notion of formgiving […] Formgiving is the act of deliberately manipulating a material into a form in which functions resides” (Vallgårda and Sokoler, 2010, p. 2) This refers to a process of developing function and form dialectically. They further argue for a view on computers as a compositional material that through its combination with other materials allow for various physical and temporal expressions. Formgiving here is the craft of making compositions of computational and traditional materials by skillful practice and material knowledge.

A formgiving approach to shape-change is concerned with developing the actual form as well as the dynamics of the changes within in. On one hand, formgiving is therefore the process in which we define the abstract narrative of the object through its visceral expression, but also the process that allow us to actively explore new applications and “[…] to take a step backwards in order to enable steps in new directions” (Vallgårda and Sokoler, 2010, p. 5).

A DESIGNERLY APPROACH TO FORMGIVING While they are not HCI or interaction design scholars, Zainal Abidin et al.’s (2008) writings are still relevant and will help to further position formgiving as a design activity. In their paper they outline two formgiving traditions: Industrial design and engineering design.

The two traditions carry with them two sets of practices and understandings of the design process. Industrial design, as they present it, involves combining basic entities into a desirable form. The form is not predetermined, but is developed throughout the process based on the quality or character of form. The form emerges through the process, similar to Vallgårda and Sokoler’s descriptions of ongoing form negotiation. Engineering design, on the other hand, is the transition of a defined function into form: 40

Behaving expressions

“Formgiving, when used in engineering design, relates sometimes to a specific phase in the design process: the part in which a solution-principle is developed into a materialized design [...]

Here, the emphasis is on the embodiment; the determination of form and material, as well as the process of bringing both in line with each other” (Zainal Abidin et al., 2008, p. 1) Zainal Abidin et al. also note how engineering design is characterized by preferring quantitative data for form evaluation, while industrial design has an art and design perspective on form and prefers qualitative measurements. Their work shows how formgiving is no singular practice, but is deeply rooted in specific design traditions.

Though formgiving has been a recurrent concept in HCI and interaction design, the traditions from which it has been adopted have varied. Some uses of formgiving has drawn more on the traditions from engineering design (e.g. Djajadiningrat et al., 2004), whereas others (e.g.

Vallgårda and Sokoler, 2010) are closer related to the more designerly tradition in industrial design.

Although Heidegger (1977) is a philosopher and not a design practitioner, he has addressed formgiving in his writings on technology as a mode of bringing-forth an object. Based on an example of how a silversmith brings forth a chalice, he introduces the classical concepts of the Aristotelian four causes (table 3.1) as a way to discuss how the silversmith’s practice takes place. To free technology from purely functional purposes, Heidegger feels the need to challenge the classical notion of these causes: ”For a long time we have been accustomed to representing cause as that which brings something about. In this connection, to bring about means to obtain results, effects. The causa efficiens, but

Table 3.1 – The four causes

Causa materialis Causa formalis Causa finalis Causa efficiens

The material out of which an object is produced The shape and composition of the object The function or end of the object The practitioner or designer who brings about the object

41

one among the four causes, sets the standard for all causality. This goes so far that we no longer even count the causa finalis, telic finality, as causality. Causa, casus, belongs to the verb cadere, ‘to fall,’ and means that which brings it about that something falls out as a result in such and such a way” (Heidegger, 1977, p. 5).

What Heidegger argues is how a designer – the cause efficiens – often works from an established instrumental ideal for technological objects. The function – the cause finalis – of the object has already been defined prior to the creation process. This perspective is similar to the engineering design formgiving tradition described above (Zainal Abidin et al., 2008). Here the object is predetermined, making formgiving the process of ensuring a certain result. As an alternative, Heidegger appropriates the four causes towards a practice similar to that of industrial design. He releases the four causes from predeterminated categories to modes of occasioning in a process of revealing an object. By using “revealing” instead of “manufacturing”, the focus shifts from something predetermined to a process of bringing something into appearance or helping it “on its way” (Heidegger, 1977, p. 6). The practice of formgiving proposed by Heidegger is not about forcing the material into a particular form, but rather to reveal the potentials of the materials, the compositions between them, and the overall abstract concept framing the object.

As an approach to shape-changes this means that rather than making interfaces that communicate certain information or serves a specific purpose, the formgiving practice is what reveals the shape, composition and even function through explorations of material qualities and potentials, similar to the Hallnäs and Redström’s (2002a) notion of abstract expressionals.

As outlined in the previous, formgiving in this thesis is driven by a designerly mode of revealing form through a holistic understanding of the complex relationship between basic form entities. The expression and function of the object as well as its interaction possibilities emerges through the composition of, for instance, these basic entities, affordances, and material selections. These compositions emerge based on the decisions of the designer, who therefore plays a central role in formgiving. This is also apparent in Heidegger’s work, where the designer constitutes one of the four causes. A form approach requires that designers have a certain expertise, in order to be able to make appropriate decisions and to be aware of their expressional means.

As a method for qualifying this design expertise, Hove (2010) proposes three concepts (lens, holistic take, and appraisal) to help designers articulate the way they grasp their formgiving.

Through this articulation, designers can begin to reflect upon their own work, but it will also serve as an informed basis for others to critique and discuss this work. In this sense, it will expose the premise for how the designer has given form to an object and through this, it will qualify the way it can be critiqued.

42

Behaving expressions

While Hove’s work is not focused on design of interactive objects, Larsen (2015) has introduced her work to an interaction design context, as well as made an effort to translate Hove’s Danish terms to English.

Lens (in Danish: blik) is a way of perceiving and comprehending form. For instance, lenses can come from perception psychology or design traditions. However, as Hove (2010, p. 16) argues, design often works with particular problems, where established lenses have not yet been articulated. A lens is something the designer develops through practice and is not something that can be taught, but must be learned through the designer’s own compositional work. In this sense, Itten (1975) sought to facilitate that the students developed their lens for compositional work.

Holistic take (in Danish: greb) is “[…] an emergent and overarching guiding principle permeating how one addresses (senses/ views/ reflects/ acts on) a design space” (Larsen, 2015, p. 145). The holistic take indicates a designer’s way of doing design. For instance, Itten (1975) used contrasting expressions as a way to expose aesthetic qualities of various compositions.

Appraisal (in Danish: værdsættelse) is a designer’s more general “[…] holistic set of values, stances and judgements integral to the designer’s ‘register’/way of conduct” (Larsen, 2015, p.

145). Articulation of a designer’s appraisal is central to understand design choices and applied aesthetic strategies.

As we saw in the writings of Heidegger, the designer plays a central role in a form approach to shape-changing interfaces. The way a designer gives form to an object is not simply neutral, but it tacitly embodies the ways and values of the designer. Hove’s three concepts have been presented as part of a formgiving practice to shape-changing interfaces because they will help to expose the embodied values in the object, which have permeated the formgiving process.

FORMGIVING OF SHAPE-CHANGING OBJECTS When a glassblower shapes glass or when Heidegger’s silversmith gives form to a new chalice, they work with a single material in which they reveal an object’s final form. But when architects give form to a new house, they do not work directly with the bricks or other materials that will make up the house. Instead, architects make drawings or scale models as a way to reveal the overall form compositions of a new building. It is also possible for the architect to delve into details about the new house. How should the doorway look? How will the wall be combined with the ceiling. As a result of these overall and detail processes of revealing a final form for a house – or a design, in general – will emerge. This final form can then be prepared and interpreted into building instructions for actual construction. Formgiving for the architect deals with the complex expressional unity rather than specific details on construction (Hove, 2010).

43

While formgiving in interaction design has previously been compared to the practice glassblowing (Vallg책rda and Sokoler, 2010), this may not embrace all the complexities of giving form to computational objects. Interaction designers cannot give form to the entire gestalt in one confined process, but must work with different elements in different stages of the process. In the process of designing shape-changing interfaces, designers have to consider both the three form components individually as well as their compositional relationship. The physical form must be developed; not in isolation, but with the temporal form in mind. Also, the technological set-up and the appropriate temporal form must be configured to match the other form components. As such, the formgiving of shape-changing interfaces is highly complex, and is therefore more likely to resemble the architectural practice of exploring both the overall as well as subparts of the overall design gestalt.

Moreover, formgiving processes in interaction design will go through different stages. The initial formgiving may be characterised as more exploratory investigations, whereas later stages of the process closer to an actual prototype or product will include more normative and defining decisions. As such, the formgiving process does not always only appear as a process of revealing and releasing material potentials, but can shift between different modes and purposes.

In the following, I will first, in chapter four, present my methodology for investigation the proposed form approach. In chapter five and six I will present my formgiving process , which has consisted of both a phase of experimentation as well as a phase of prototype production.

44

Chapter four

Methodology Construction is the cornerstone of my practice. While the form analyses are descriptive design exercises, both the formgiving experiments and the prototype production rely on the construction of physical objects. The primary subject for my methodology is my relationship to these objects and their creation process. I then use these physical constructions to explore the form approach to shape-changing computational forms, as well as to investigate the potential design vocabulary. This way, the physical constructions become physical hypotheses (Koskinen et al., 2011, pp. 60–62).

The methodology in this thesis is, thus, based on my own formgiving process, in which I have explored a series of form compositions and expressions. As a synthesis of these exploratory form experiments, I have produced two prototypes. To challenge the form language, I have applied the proposed design vocabulary to three of the projects presented in chapter two to make descriptive form analyses. In the following, I will briefly elaborate further on the heuristic methodical approach, which has been the basis for both informing and developing an understanding of a form approach to shape-changing interfaces.

Designerly form experimentation To explore the theoretical framework for a designerly approach to shape-change I have performed a series of material experiments. The purpose of the experiments was to explore the described form language and to investigate the ways in which it could be used generatively in a design practice. Design decisions and experimental set-ups for the expressional possibilities of various form compositions were based on my designerly decisions and curiosity, rather than on predetermined logical steps. For instance, this can be seen in the judgements I made in order to decide if a given visceral form expression felt well-balanced or interesting. Based on these judgements, I made new iterations of the form compositions to change a minor detail or was inspired for completely new set-ups. The experimental process was guided by my aesthetic judgements based on my visceral experience of the relation between the given form properties and the resulting aesthetic expression.

45

To capture and document the formgiving experiments, I have used Kozel’s (2007) method of phenomenological inquiry to reflect upon my own experiences in order to inform the design space and my theoretical contribution. Thus, the embodied experience of working with materials and technology, reflecting on these experiences, and synthesising them into new experimental iterations are central elements in the knowledge production. Through numerous low-fidelity experiments, I have explored different basic dynamics as well as potential mechanics that could be used to achieve various expressions.

My experiments are attempts to work with dynamic and changing forms in a tangible way.

In this sense, each experiment is a physical manifestation of more abstract concepts. For instance, my various experiments then seek to explore expressional complexity and how it may manifest in physical form. My sketches are, in this sense, physical hypothesis (Koskinen et al., 2011, pp. 60–61) that through their making inform and develop my design space and theorising.

Phenomenological inquiry is a method based on (hyper)reflection on lived experiences, and Kozel argues how “[…] potential dense or difficult concepts can be demystified and given a sort of intuitive fluidity once they are read through the body” (Kozel, 2007, p. XV). Through reflections on my experiences of various dynamic expressions, I begin to give substance to the proposed form language for shape-changing interfaces. Kozel describes a series of steps to her phenomenological method, which I will here roughly summarize in three overall steps: › Awareness of the experience itself (being in it; sensing through the body) › Describe and document the felt experiences › Revisit the descriptions (Kozel, 2007, pp. 53–55) All experiments were low-fidelity and did not include any technology, but were a way to explore basic dynamics and mechanics. As I constructed each experiment, I made a series of design decisions that manifested the hypotheses I wished to investigate. To document these decisions, I described the three form components of computational objects (see chapter three) in a reflection sheet. With each experiment, I performed small enactments with them to better understand dynamic aspects of the sketches (Arvola and Artman, 2007). In these performances, I would improvise with the physical sketch and control its temporal behaviour through simple mechanics, such as strings or simply by hand. This way, I was able to easily make changes to the way the sketch behaved and experience their effects first-hand.

Where Kozel refers to the bodily awareness of the actual performance, my work has also included a heightened awareness and reflection on the actual construction of the experiments 46

Figure 4.1 sheet

Example of how a set-up and reflections from the experiment were articulated in a reflection

prior to the actual performance of them. Being aware about the design decisions in each experiment has been an essential aspect in my creative process and has allowed me to qualify my work. Reflections from each experiment were also noted in the reflection sheets.

Table 4.1 schematically presents my design and research practice based on Kozel’s (2007) phenomenological framework.

Prototype production As a result of the formgiving explorations, I designed two prototype concepts that were based on steel constructions. These steel constructions were manufactured by a production company. While prototypes are often the result of a process towards a solution to a problem (Buxton, 2007; Houde and Hill, 1997), prototypes are used here “[…] in their generative role in enabling designers to reflect on their design activities in exploring a design space” (Lim et al., 2008). The two prototypes have been used as a tool for synthesising the form experiments based on reflection on their qualities. I call them prototypes as they have a more defined and specific nature, compared to the exploratory and tentative nature of the form experiments.

Moreover, the prototypes added new dimensions to the explorations of the design space.

Where the experiments has focused particularly on revealing interesting forms, the prototypes

47

Table 4.1 – Adaptation of Kozel’s phenomenological inquiry to my design practice

Phenomenological inquiry

My design practice

Framing study through abstract concepts

The design space: Exploring shapechange from a form perspective and the design vocabulary

Constructing the experiments with awareness of the design decisions and their purposes

Phenomenological inquiry – method

1. Awareness of the experience itself (being in it; sensing through the body)

Enacting and performing the experiments to explore the dynamics and possible expressional qualities and potentials.

Each experiment has its own reflection sheet, where one column was allocated to the description of the experimental setup and the decisions for each form component.

2. Describe and document the felt experiences

Another column was used to articulate the felt experiences of performing and enacting the experiment.

3. Revisit the descriptions

Based on the notes on the reflection sheets, new ideas for experiments were synthetized.

Overall conceptual insights and revelations

Two fully functional prototypes were constructed, based on qualities and insights from the various experiments.

48

Methodology

also insisted for attention to other important aspects of the framework; for instance, material selection, colour, and scale. My prototypes has particularly focus on the visceral look and feel aspects of the design space (Houde and Hill, 1997).

Lim et al. (2008) also stress how prototypes are important in communicating design ideas between designers. The two prototypes are a way of exemplifying my work in a way that can be more accessibly shared. In this sense, the prototypes are physical manifestations of the proposed theoretical framework. This is similar to Vallgårda’s (2008) use of PLANKS as a physical manifestation of the, at the time, innovative approach to computers as a computational material. In chapter six, I will go into a deeper description of the two prototypes based on the form language to illustrate its descriptive potential.

Initially, the purpose of prototypes were to serve as syntheses of the experiments. However, as it turned out, the actual production process has given interesting insights regarding the form language. The need for articulating construction instructions to the production company proved to be a valuable exercise in using the proposed design vocabulary to communicate design decisions to others. Thus, the prototype production process will be included in the thesis as a way to illustrate the usefulness of the form language as a communicative tool.

Descriptive form exercises The formgiving and prototype production processes have been used to explore and give meaning to the form language. However, to challenge the durability of the proposed framework beyond my own work, I have used it to analyse other works within shape-changing interfaces.

By using the design vocabulary on a number of the projects presented in chapter two, I seek to support the viability of the form language. With no intention of arguing for a universal applicability, I still hope that this use of the vocabulary beyond my own work is a step towards a more basic and general form language for shape-change in interaction design. Similar to how Ching (2014) uses the work of other architects as the argument for his illustration of the basic elements and compositional rules of architecture, the form descriptions of other’s projects serve as a perspectival exercise in illustrating its potentials.

The purpose of the exercises is, on one hand, to show the descriptive capacities of the form language, but, on the other hand, also to illustrate how the form language can inform compositional exercises to qualify the work of the design practitioner. In Itten’s (1975) basic design course, the students were not only asked to create their own formgiving experiment, but also to analyse the works of others as a way to strengthen their design skills.

Whereas the form experiments and prototype production were characterised by the 49

construction of physical objects, this is a more classical analytical exercise, where a theoretical framework is used as a lens to shed lights on another matter. The projects in chapter two were used to argue for the need of a new design space with focus on form and expression of shapechanging interfaces. With them, I argued for the basic position of this thesis. Returning to them after the presentation of a theoretical approach to the design space can be regarded as a hermeneutic manoeuvre to challenge the proposed framework in order to understand its potentials and limitations. The form exercises will be presented in chapter seven.

50

Chapter five

Formgiving experiments During a period of three weeks, I have constructed and performed 14 formgiving experiments.

With the framing of phenomenological inquiry, the experiments took their beginning in conceptual starting points. The form approach as a design space is the conceptual space that frames the overall research agenda. However, this design space cannot be directly translated into actual design work as it is a conceptual representation of all possible solutions and cannot be fully described (Westerlund, 2005). It has therefore been necessary to define a way to navigate my journey into the design space.

As a way to articulate this I will return to Hove’s three concepts (lens, holistic take, and appraisal) presented in chapter three.

My lens in the experimentation was guided by the five primary elements of the proposed form language. The notions of the point, the line, the plane, the volume, and the force were, on one hand, the consistent theme that served as an inspirational framing, but on the other hand was informed themselves through the experiments. For instance, I would begin to imagine how the plane could serve as a generative concept for imagining interesting forms, but in turn the resultant composition of the experiment would informed the plane as a form element.

In general, the aim of the formgiving was to develop the primary elements and to explore their generative capacities. However, I have not explicitly worked with the vocabulary in a literal way. Instead, it has been the underlying perspective, which has framed the formgiving experiments.

To further frame my explorations, I was inspired by Itten’s way of using contrasting parameters to explore expressional qualities of different compositions. My holistic take was therefore to explore expressional control and complexity as two contrasting parameters. I further limited the explorations to forms and expressions that could be realised in constructions with two motors.

Expressional control refers to the degree of control the designer has of the overall expression.

A high level of control means that the composition behaves tightly to changes in the temporal form; whereas, a low level of control occurs when the composition does not respond tightly in accordance to the temporal form. For instance, if there is a high material resistance in the composition this can result in more degrees of freedom in the way the composition responds to temporal form changes. The second parameter, expressional complexity, is related to the 51

experience of the expressions. I will by no means argue that these are necessarily the two optimal or most relevant parameters to explore. However, while they naturally limit the extent of the experimentations, they have made it possible to act within the design space.

As a designer, I value abstract interfaces as a way to explore new expressions of computation.

As such, my appraisal is closely related to the entire purposes of the form approach. Since my formgiving in this thesis is not driven by the desire to give form to new objects, but as a way to explore and unfold the form approach, my use of Hove’s concepts differ from how they would be put to use in a design practice. Here they would serve as a way to define a final object, whereas I only use them as ongoing explorations of a design space.

With this overall framing of the experimentation, I began to explore and investigate multiple form compositions. Based on my reflective experience of each experiment and my designerly judgements, the experiments were continuously developed based on qualities and insights from the previous. The experiments are all interrelated, although four thematic threads emerged during the experimentation, which I wish to explain in more detail in this chapter.

Plate series As a starting point, I wished to work with a set-up that gave me full control of the expressions, while having a rather low complexity. I was inspired by the mechanics of a wooden labyrinth game, as it would allow moving a plate both vertically and horizontally independently. I did this as a way to work with better understanding the plane as a form element.

The goal of the first experiment (Plate I) was to understand the mechanics of the labyrinth game. In order to be able to explore expressional qualities of moving planar elements, I had to understand the mechanisms that could be used to achieve this. Therefore I made a setup with less attention to material selection and expressive qualities, but simply made a quick laser-cut sketch (figure 5.1 a). From this I was able to begin exploring other elements of this type of expression. Figure 5.1 shows five iterations of investigating various compositional elements of the set-up. I sought to make a set-up that would draw attention to the inner frame (what I consider as the actual moving plate). The two outer frames are necessary for the mechanism, but the goal was to create the illusion of a single moving plate. To attain this, I narrowed the two outer frames from Plate II (figure 5.1 b) to the third experiment Plate III (figure 5.1 c). Although the two outer frames are still visible in the third experiment, I believe the focus has shifted towards the centre-plate compared to the previous experiment with different proportions. Figure 5.2 shows the potential temporal dynamics of Plate III where the main focus is on the inner plate, rather than the two outer frames.

52

a)

b)

c)

d)

e)

Figure 5.1 a) Plate I

The form experiments in the Plate series b) Plate II c) Plate III d) Plate IV

e) Plate V

Figure 5.2 The temporal potentials of the moving plate set-up Read from top left to bottom right

Both the Plate II and Plate III experiments were laser-cut from white acrylic and assembled with bolts. As a way to explore complexity in the overall expression and to challenge the qualities I had experienced from especially the Plate III experiment, I wanted to replace the hardness of the plate with a more soft and light textile. Figure 5.1 d and e portray two experiments (Plate IV and Plate V) where two types and layouts of textiles have been attached to an inner frame instead of the plate in the previous experiments. My surmise for the experiments was that the textile would bring a more dynamic element to the expression as it would move with a higher degree of freedom. I especially thought that the layered set-up in Plate IV (Figure 5.1 d) would produce an expressional complexity and dynamic. However, I did not at all experience that the textile brought anything desirable to the plate experiments. Rather it seemed to diminish the qualities from the expression from the mechanism, which was much more apparent in the set-ups with an actual plate. Here, the movements and rotation in both vertical and horizontal direction were more visible, resulting in a more balanced expression.

Textile series To explore contrasting expressions, another line of experimentation dealt with a higher degree of freedom in the material, resulting in less control of the expression. Also, the complexity of the expression was intensified. This series of experiments were all based on set-ups of layered textile (Figure 5.3). The initial experiment was simply a piece of felt that I twisted with my hands. Due to the stiff materiality of the felt, the textile would fold and sit in a structured layout (Figure 5.7). The changes in the structure of the textile layers gave an interesting play due to the way the light was able to shine through the fabric. I responded positively towards the way the felt was able to be flexible and almost organic, while maintaining an overall structure. When the textile was twisted, its shape changed. If it was then twisted back to the original position, it would not attain the exact same shape, but instead relax into a different structure. It had a certain degree of freedom in the material, which gave interesting element s to the expressional potentials of the set-up.

Inspired by the textile work of the Berlin-based designer Elisa Strozyk (2011), I made another setup where twenty-or-so strips of felt were cut out and attached together in each end. While I liked the initial experiment with a whole piece of felt, I wanted to explore versions with a higher degree of expressional control. By combining the pieces of cutout felt, I increased the control of the shape of the composition, but still with more material resistance compared to the plate experiments. Thus, this set-up allowed for a flexible structure that would dramatically change shape as the two attachments points were either vertically displaced (Figure 5.4 top row) or twisted (Figure 5.4 bottom row).

56

a)

b)

c)

Figure 5.3 The experiment series that explored less controllable expressions a) Textile I b) Textile II c) Textile III

Figure 5.4 Top row: Vertically shifting the two contact points relatively Bottom row: Twisting the textile

Arches series In an attempt to explore an area of the design space between the plate and the textile series, I made a number of experiments based on curved plastic. This way I could take a starting point in a more stiff material, closer to the plate, while remaining a more flexible quality, as in the textile. Where the planar elements in the plate series were rigid, I wanted to explore how planar elements could be used to enter into other compositions as well; for instance, as curved elements. Inspired by the two last textile experiments, I also created the arches series as cutout pieces (Figure 5.5). For the first three experiments in this series (Figure 5.5 a, b, and d), I did not confine to the principle of two motors, but more openly explored different expressions. Towards the end of this experimental line, I began to understand how the twomotor principle could be applied to this set-up (Figure 5.5 d). While this experiment still had more than two arches, I connected the arches to an inside crossbar, which was then controlled by two strings (potentially motors).

In the initial experiment, Arches I, I wanted to investigate the material qualities of the plastic and whether it could result in an interesting form. The first set-up was therefore made quickly from a transparent plastic sheet that was sliced with a knife to create four arches (Figure 5.5 a). Each arch was attached to one or two strings. One of the arches had a string attached in its geometrical summit. When the string was pulled it would bend downwards in the middle, creating a soft waving shape. Another of the arches had two strings attached to it at similar distances from its axis of symmetry. This also resulted in a waving dynamic, but with a larger degree of expressional possibilities due to the multiple combinations of tensions in the two strings. However, when the strings were pulled with similar force at the same time, the plastic opposed the movements in a less natural flowing movement, but with a more broken motion.

This resulted in a strained, almost unpleasant, expression. This was probably due to the fact that the material had to respond to two forces simultaneously, whereas the arches with only one string only had to follow one force.

By pulling the strings, I was able to bend the arches and make dynamic movements with focus on both the individual arch, but also in the relational movements between them. When all the arches were in a relaxed state (that is, the string were not pulled), the arches formed an even surface. However, as soon one or more strings were pulled, this uniform surface would be broken and the individual movements of the arches would result in a somewhat dramatic effect. I was drawn to this tension between the calmness of the even surface and the sudden changes into individual movements of the arches.

In this first set-up, I had not sliced the arches completely from the plastic. This way, they were held together in each end by the plastic sheet. This meant that the movement from one arch would influence the nearby arches as well. To reduce this mutual influence, I made a 60

a)

b)

c)

d)

Figure 5.5 The bending arches experimental set-ups a) Arches I b) Arches II c) Arches III d) Arches IV

Figure 5.6 Although the set-up only has two attachment points it is still possible to produce a rather complex organic expression

second iteration of the transparent plastic sheet, where the arches were completely cutout of the plastic (Figure 5.5 b). This allowed for a more tight relation between the pull in the string and the resulting movement without impact from the movement of the other arches. This resulted in more expressional control.

To explore the overall expression of the arches series, I wanted to use a different material that was not transparent. Therefore, I made two versions with white plastic (Figure 5.5 c and d).

The first version in white plastic, Arches III, (Figure 5.5 c) was similar in composition to the previous version. However, where the strings on the first iterations were placed in different attachment points on the arches, the strings in this version were all placed in the geometrical summits of the arches. I did this to investigate the influence of the attachment points on the overall expression. When all strings were attached to the same point on the arches, the movements of the individual arches were more similar. This gave it a more coherent overall expression, with an almost rhythmic element to it. However, in the two first iterations, where the string were positioned in different points, the change between the overall relaxed surface to the topological changes in the surface was more surprising. It felt less predictable which gave it a higher expressional complexity. I even experienced it to feel more organic as the arches suddenly came to have their own individual ways of behaving.

In the last iteration of the arches series (Figure 5.5 d) I increased the number of arches to ten. This was an attempt to increase the complexity of the overall expression by adding more moving elements. However, to avoid also increasing the complexity of a potential mechanical set-up and to return to the principle of two motors, I attached the arches’ strings to a crossbar.

In each end of the crossbar I attached two strings, which I could then use to change the shape of the set-up. Of course, this meant that the arches could not be individually controlled, but were depending on the movement of the crossbar. Consequently, this reduced the number of ways the set-up could behave. Despite this, I did not think the experiment seemed to loose its expressional complexity. Although it could only change in the two ends of the crossbar, the fact that it has this increased number of arches made it appear just as interesting as the versions where the arches could move independently. In fact, due to the relatively large number of ways the crossbar can be positioned (one end could be completely pulled down, while the other end was relaxed or any other kind of combinations), the overall expression of the set-up was quite manifold (see Figure 5.6). It seemed like an ideal way of having a high level of expressional complexity with a relatively simple set-up.

64

Formgiving experiments

Figure 5.7 These two experiments were an alternative exploration of the bending arches Left: Box I Right: Box II

Box series Inspired by the three previous series of experimentations, I wanted to explore some of their qualities in a new composition. Particularly, I was curious about the potential of using the dynamics of the arches in another way. Where the arches were in a curved shape from the beginning, I wondered if it would be interesting to have a planar shape that would curve when a force was applied to it. Therefore, I made an acrylic box, where the top was replaced with a piece of flexible textile (Figure 5.7 left). Four strings were attached to the textile and as the strings were pulled, the textile would curve inward. However, as I did not respond positively to the set-up, I wanted to take a different approach. Inspired by the cutout pieces of textile in the last iteration of the textile series, I made another box where the top consisted of four textile strips, each attached to a string (Figure 5.7 right). Also this set-up was not as interesting to me as the other experiments. Maybe it was due to the fact that the use of a string attached in one point in the textile did not create a continual curved shape, but instead seemed boxy and strained. As I had not encountered any interesting or inspirational qualities from these two experiments, I decided not to make any further explorations of this set-up.

Based on these four experimental series I have develop two prototypes, which will be explained in the following chapter.

65

Chapter six

Prototype production Based on the form experiments, I have created two prototypes. The constructions have been produced by a company which I have then added motors and computational capacities to.

With my holistic take in mind, I wanted to make two prototypes that pointed to different qualities in terms of expressional control and complexity. The design decisions for the prototypes were based on my experience of expressional qualities from the various form iterations described in the previous chapter. While the design experiments allowed me to quickly explore different shapes and compositions as well as their potential expressions, the prototypes required me to concretise design ideas and to attend to other details; for instance, measurements, compositional ratios, material and colour.

Both prototypes have been made in steel and painted in a matte white. Steel was chosen as it has material properties that can embrace the form requirements of both prototypes; it can be used both as a rigid plane and both as bended arches. It would, for instance, be less appropriate to create both prototypes from wood. Moreover, steel can be bent and welded, which makes it easier to put together in a more seamless manner. Inspired by architectural scale models, I decided to have the two prototypes painted in a standard white colour (RAL9010). As nothing is ever truly neutral, the white colour is not meant to strip away any meaning, but rather emphasise the two prototypes as having minimalistic and coherent styles.

Tilting\Plate Tilting\Plate is a 600x600x300 mm steel box. The top of the box consists of two frames, each 15 mm wide, and an inner plate that is 520x520 mm (figure 6.1). The outer frame is part of the steel box itself, where the second frame and the inner plate are connected to the box with bolts. The inner plate is connected to the second frame with two bolts, which in turn is connected to the outer frame with two bolts opposite the inner plate’s bolts. This makes it possible for the inner plate to rotate around an axis of symmetry, and for the second frame to rotate around an axis perpendicular to the other axis (figure 6.2).

67

Figure 6.1

Tilting\Plate is a box with a dynamic top

Axis of symmetry

Axis of symmetry

Figure 6.2 other

Both the inner plate and the second frame have axes of symmetry perpendicular to each

Figure 6.3 plates

There are two motors inside the box. The red lines indicate the strings used to move the

Two servomotors are placed in the centre on the bottom of the box and controlled with an Arduino board. One motor is connected to the second frame in the opposite sides of the axis of symmetry, while the other motor is connected to the inner plate (figure 6.3). As the motors rotate, they will apply a force in the attachment points on the second frame and the inner plate respectively. Since the inner plate can rotate around one axis and is attached to the second frame, which can rotate around the perpendicular axis, the inner plate appears to be able to move in all directions (figure 6.4). This gives it a highly dynamic expression and it appears almost to be floating, due to the fact that its movements appear to be entirely independent from the rest of the box.

As already argued, the focus of this thesis is on the way in which the temporal form is expressed in the physical form composition. The scope is not to explore the actual interaction with such forms. However, to include minimal elements of interaction I have attached two potentiometers to the Arduino board, each controlling one of the two motors. This way, the interaction defines the movements of the plate. Due to the symmetrical dynamics of the prototype, the motors have a natural point of origin around which it could rotate. Similarly, I wanted to include this in the interaction. The potentiometers have an input range from 0 to 1023. When the value of each potentiometer is around 511, the two motors will be in a position, where the plate and frame makes a levelled top of the box. Alternating the values below 511 will result in movements in one direction, whereas values above 511 will make the frame or plate move in the other direction. While much more complex and interesting interaction modalities could be incorporated in the design, I wanted to keep the interaction very simple to maintain focus on the form dynamics.

70

Figure 6.4

The temporal dynamic of Tilting\Plate

In the description of the prototype above, I have already used some of the basic elements of the proposed form language. I have referred to bolt attachments between the plate and the frames as points. Also the attachment of the motors to the plate and second frame are points in which the force vector is applied resulting in the tilting movements. The symmetry axes can be drawn is lines. Both the plate and the second frame are conceptual planes that are angularly displaced to the box and to each other. Volume can be used to understand to the form in two ways. The box itself has a volume of a specific scale. But the space in which the planes are tilting can also be used to define an imaginary volume. This is relevant to do, in order to discuss the extent of the movements. Figure 6.5 illustrates the use of the design vocabulary to describe the prototype.

Attachment points Points

Line and plane

Line: Axis of symmetry

Line and plane Line: Axis of symmetry

Volume

Figure 6.5

Tilting\Plate described through the design vocabulary

Forces and volume

Tilting\Plate

Figure 6.6

Bending\Arches is made from ten flexible arches

Figure 6.7 The strings are attached to the arches in different attachment points. This gives a more complex dynamic form as the temporal form is expressed in the prototype

Figure 6.8 Two linear actuators are run by each their stepper motor, which pulls down the crossbar and thus the arches. The red lines indicate the metal wires that are connected to the crossbar

Bending\Arches The second prototype, Bending\Arches, was particularly inspired by the arches experiments.

It consists of ten pieces of flexible steel that are connected by an angled mounting. The arches are then fastened onto a 670x500x70 mm steel base (figure 6.6). The total height of the prototype is 370 mm. Each arch is connected to a crossbar with fishing line. The position of the attachment points varies from the geometric nexus of each arch to a displaced position further down the arch. This way the prototype has a uniform surface in its relaxed state and a complex, uneven surface when the arches are bent; similar to the expression explored in the previous formgiving explorations (figure 6.7).

Two linear actuators inside the steel base, both controlled with stepper motors, are used to pull down each end of the crossbar. By controlling the rotational force (or torque) of the motors it is possible to determine the amount each end of the crossbar is pulled down (see the inside view of the steel base in figure 6.8). This way it is possible to control the slope of the crossbar, which is used to change the overall shape of the prototype.

The temporal form of the prototype is based on a simple algorithm. As a start-up routine, the motors are calibrated to a zero; that is, the position where the arches are fully relaxed. From this, the motors have a maximum number of steps they can move, which is determined by the position where the arches are fully bent. In between this minimum (zero) position and maximum number of steps, the motors can move freely. In an arbitrary order, the motors are assigned new positions based on a randomized principle between the minimum and 76

Prototype production

maximum positions. The interaction in Bending\Arches is also kept simple. A potentiometer is connected to the prototype’s Arduino board. With this it is possible to control the speed motors as steps per second. Again, an analogue input (1024 values) has been used, and at such the input values are mapped based on a minimum and maximum speed. Only the speed of the performance of the motors is affected by the input.

Similar to Tilting\Plate, this prototype can also be described by the form language; see figure 6.9. The point can be used to describe the position of the strings’ attachment to the arches, as well as the metal wires’ position on the steel crossbar. The crossbar itself can be regarded is a line, which is helpful in the description of its slope. In architecture, Ching (2014, p. 25) illustrates how planes can also be used to describe curving elements; for instance, vaulted ceilings. Each arch can thus be considered a curved planar element. Forces are applied to the crossbar through the metal wires, which are pulled by the linear actuators. The magnitude and direction of the forces determine the movements of the prototype.

Communicating with the production company The steel division in a Danish Offshore, Marine, and Energy Company has produced the two prototypes. As a design brief, I drew two construction drawings with measurements, which they could use to produce the steel constructions. To utilise their expert knowledge on the properties of steel, I also sent them video material from my formgiving experiments. This way they were better able to understand the purpose of the constructions, which enabled them to suggest constructional changes and material selection to best possibly support expressions I was aiming for. For instance, they tested several types of steel to determine the kind that was best fitted for Bending\Arches (as it required high flexibility and less material memory than regular steel).

While I was unaware of it at the time of production, I used the proposed form language extensively to explain the requirements to the construction during the production process.

For instance, the construction drawings were made almost identical to illustrations in figure 6.5 and 6.9, and we discussed the arches as curved planes of steel or the attachments in the Tilting\Plate as points. Also, the notion of force was used on multiple occasions. For instance, in the description of the movements in Tilting/Plate where forces applied to the edges of the frames would result in movements around the symmetry axes. Especially, when determining the appropriate type of steel for the arches, we explicitly discussed how the reach the lowest levels of necessary force to bend the arches. This was important in order to ensure that the motor capacity was sufficient to actually produce shape-changes in the prototype. As such, the form language seems as an easily adopted vocabulary that can be readily shared with others.

77

Point

The strings’ attachment points

Crossbar

Line

Strings

Plane Planar element

Curved planar element

Volume

Figure 6.9

Bending\Arches described by the design vocabulary

Forces

Bending\Arches

Chapter seven

Descriptive form exercises In the previous chapters, I have shown how the form language has framed my design practice.

However, to further develop the form language and to challenge it beyond my own work, I will, in the following, apply it to three of the projects presented in chapter two. It is by no means an attempt to argue for the universality of the design vocabulary, but instead I hope to make its potential probable. Furthermore, it is important to stress that the form language is not meant for thorough geometric analyses, but rather as concepts to help designers make meaning regarding shape compositions. The purpose of the following analyses is to use the vocabulary as a framework for looking at the objects and understanding their compositional structures as well as temporal dynamics. That is, the point, line, plane, volume, and force are relevant for interaction design, only if we can use them in our design practices and analyses to grasp the form dynamics of shape-changing interfaces.

Topobo: Points on a line and rotating volumes The interactive toy, Topobo, consists of modular pieces that can be connected to create animal-like robots: “By snapping together a combination of static and motorized components, people can quickly assemble dynamic biomorphic forms like animals and skeletons� (Parkes et al., 2008, p. 64). Figure 7.1 shows both an image of Topobo as well as a schematic drawing of how the object can be regarded as a composition of the basic form elements.

The static components are denoted as points that are attached on a line. While the static components are not simply points in the physical object, it is a way to use the design vocabulary to make meaning of the form dynamics of Topobo. In order to understand the mechanism and dynamics in Topobo, it is less relevant which specific shape the static components have.

The interesting aspect is instead more on the relation between these elements. This way, the form language can be used on multiple levels depending on the scope of the analysis. This is also apparent in architecture, where the point is sometimes used to denote a specific nail in a construction, but at other times is used to position monuments on city scale (Ching, 2014, p. 7).

Collectively, the points on lines make up volumes, which I have separated depending on 81

Rotational force

Rotational force

Force that drives the object forward Figure 7.1 (bottom)

An image of Topobo (top) as well as a form analysis based on the design vocabulary

Descriptive form exercises

their internal relationships. The “neck/head” of the object is one volume whereas the “tail” is another. By defining the static elements as a volume it is possible to describe the collection of these as a whole, which makes it simpler to understand the dynamics of the object. Each of these volumes is attached to a motorised component (the light grey spherical volumes), which applies a rotational force to the neck/head, tail and to the four legs respectively. This way, Topobo is able to turn its parts, and depending on the way it has been assembled it moves accordingly.

While the form language could have been used to describe the object in more detail (for instance, the specific form of the static elements or the different ways they could be attached), I have used the vocabulary here as a way to understand the overall form composition and logic as well as using the concept of volume to articulate the moving parts of Topobo.

Morphees: A bending plane with internal forces Roudaut et al. (2013) understood their Morphees interfaces as meshes of physical points that could be controlled to make them change shape. Instead of a mesh, I have made a simpler description of one the Morphees prototypes, which is a flexible and bending interface (figure 7.2).

Overall, I understand the interface as a planar element that is affected by forces, which creates its curling movements. However, in both Topobo as well as my two prototypes, the applied forces came from rather specific points. In my prototypes, the forces came directly from the attachment of strings that were driven by motors. The rotational force in Topobo comes from the attachment point of the static elements to the motorised components. In the Morphees prototype, however, the force does not stem from a particular point, but is rather some kind of distributed force within the material.

Coelho and Zigelbaum (2011) explained a similar shape-change by distinguishing between compressional and elongating forces that are applied to each side of the affected material (see figure 2.8 on p. 18). It may be that my proposed force concept is too simple and should possibly include a more differentiated palette of attributes; already now, we have touched upon direct/ linear forces (in my prototypes), rotational forces (Topobo), elongating and compressional forces.

Another approach could be to understand the movement as being driven by the application of a number of forces. These forces are then positioned across the surface with increasing strength towards each ends. However, as I have also done in figure 7.2, the forces can also be simply understood as affecting the end points of the planar surface. In the middle of the interface 83

Forces with varying strenghts are applied across the material

A single force is applied to each end of the surface or

Line around which the material bends

Line around which the material bends

a)

b)

A volume emerges as the material is bent c) Figure 7.2 The forces in Morphees can be described in multiple ways. c) A volume emerges as the interface bends. Here one end is moving downwards, while the other end moves upwards

Descriptive form exercises

there is then a conceptual line, around which the bending occurs. What becomes central here is that the form language will no longer be sufficient to describe the movements alone. The material in which the form is created is what determines exactly how the planar element can curl. The form then requires a material that has properties, which will allow the forces to be expressed in a manner that creates the curving surface. Similarly, when an architect wants to include a large arching structure in a building, this arch must be constructed from a material, which can hold this structure without bursting. Therefore, it may not be inappropriate to have the same approach in interaction design practice. Thus, it is not enough to have a language for describing the forms, but we must also understand how the forms can be made possible with certain materials. Consequently, the form language should not exist in isolation, but must always be understood in the holistic relationship with other dimensions of the composition; such as material, colour, texture, scale and similar.

When the planar element of the Morphees prototype is completely relaxed it appears to have almost no volume. On a perceptual level it is simply a plane. However, as the surface begins to curl a third spatial dimension emerges, which gives the object a volume. The size of the volume in Morphees is not a fixed entity, but develops according to how it dynamically changes. This points to a need for understanding the relationship between the elements. How does the force relate to volume? And how does a changing volume relate to any planer elements in the object?

Questions like these are yet to be explored to fully embrace all the facets of a form approach to shape-changing interfaces.

Thrifty Faucet: Compositions of forces The shape-changing faucet by Togler et al. (2009) can be regarded as body with multiple joints (figure 7.3). Each joint can be manipulated by a force, which in turn will change its shape. As the shape is altered, the overall volume of the faucet changes accordingly.

The way the faucet moves seems organic due to the flexibility of the joints. It can both curl in a coherent movement throughout its entire body, as well as it can move some parts in one direction, while the rest of the faucet moves in another direction. Since the parts of the faucet can move in multiple directions at the same time, it is not sufficient to describe the movements with only one force. Instead, multiple forces must be used to encompass this more complicated movement patterns (figure 7.3). In the figure, I show how several forces applied to two points of the faucet can make the tip of the faucet move in one direction, while the rest of it moves in the opposite direction.

Similarly, the Tilting\Plate prototype also contain multiple forces simultaneously (see figure

85

6.5 on p. 74). A set of forces makes the inner plate move around the axis of symmetry, where another set of forces makes the second frame tilt. While the two forces applied to the inner plate are equal in size, they are opposite in direction. One force makes one end of the plate move upwards, while the other force in opposite direction pulls the other end of the plate downwards. However, the two different sets of forces have no relation and the two plates can therefore move independently. Also, multiple rotational forces influence Topobo. This makes it possible for the various “body parts” of the robot to move independently at the same time.

Therefore, it is rarely only a matter of a single force that results in shape-changes in an object.

Instead, it is a composition of forces that result in the dynamics in an object’s form. Just like the other form elements, force must then be understood as an element, which is in a compositional relationship with the other elements as well as. In fact, the temporal changes in the form are a result of the composition of the forces. This composition is not only the spatial points in which the forces are applied, but also the temporal rhythm with which they are applied.

With the previous three projects we have seen that the form language, as a lens, has indeed made it possible to discuss the objects’ actuated form dynamics. Moreover, we have seen how the language offers flexibility in the sense that it can be used on various levels, which allows it to describe the appropriate and meaningful details for the task at hand. However, it has also become apparent that the force element may need further development if it is to encompass all types of actuated transformations. No matter what, it is evident that the form language cannot exist in isolation, but must always be understood in a holistic relation with the given materials, their properties and other dimensions of the object’s gestalt. Lastly, in order to explain more elaborate temporal dynamics of shape-change, like for instance those in the Thrifty Faucet, it is helpful to look at the compositions of forces as well as the rhythm with which forces are applied to the shape-changing object.

86

The faucet has several points of joints across its body

Multiple forces affacts the faucet

a)

Changes result in varying volumes b)

Figure 7.3

The faucet consist of joint planar elements that can each be manipulated by a force

Chapter eight

Ramifications of a form approach With my thesis, I have introduced a form approach to address the immediate and sensory expressions of shape-changing interfaces. I have made a theoretical account of what such an approach entails for design practice and how it complements current research approaches. I have also begun to explore shape-changing interfaces as a form practice through a series of formgiving experiments, prototype productions and form compositional analyses. However, this is just the beginning steps towards a more elaborate form tradition for shape-changing objects. Or to quote Vallgårda: “The work is not finished—it cannot be. Incompleteness is a premise when doing research for design” (Vallgårda, 2009, preface). In the following chapter, I wish to take a step back and revisit my practical and theoretical work in order to develop the form perspective further and to look at its wider ramifications for interaction design.

Form language In chapter three, I proposed a design vocabulary as the primary elements of a form language for shape-changing interfaces. With these elements as my starting point, I have to explore different form compositions and experimental set-ups. While I did not work with the five elements in a literal manner, they still served as a lens for my work. On the basic elements in architecture, Ching writes: “While they do not actually exist, we nevertheless feel their presence” (Ching, 2014, p. 2). In a similar way, the point, the line, the plane, the volume, and the force were present in my formgiving experiments without being consciously articulated.

The concept of the force helped me to deal with the ways in which the forms behaved. For instance, I explored how the different compositions of forces applied to the arches set-up would change the expression of the experiment. In some set-ups I had attached four strings to the arches, whereas other set-ups had fewer strings attached them. As such, I investigated the results from having different number of forces in play. I found that two forces applied to a crossbar gave an interesting expressional quality when the number of arches was increased to ten instead of only three or four arches. With the crossbar, I was able to manipulate each arch with only two overall forces.

89

In the set-ups with multiple strips of textile, I was interested in their spatial arrangement. It was a matter of how the planar elements of the strips would dynamically settle in its spatial volume. Also, as the strips of textile changed their spatial position, the volume itself would change accordingly (see figure 5.4, p. 58). I furthermore explored two types of forces with the textile series. In some set-ups, the force was rotational, resulting in twisting movements in the textile, whereas other set-ups had a linear force.

The way the form elements were apparent in my design work illustrates the generative capacity of the design vocabulary. Although not explicitly articulated in the experimentation, the design vocabulary inspired for new moves, posing “What If?” questions to the compositional set-ups. What happens if the string’s attachment point is moved on the arch? What happens if the planar elements have a larger spatial displacement? What happens if a large piece of textile is cut into identical planar elements that are connected in two end points? The design vocabulary offered a grasp for the different form variables that could be altered and explored.

Besides this generative potential of the form language, I have also shown other ways it can be relevant for design practice. For the prototype production, it served as a communicative framework, which helped along the manufacturing process. Through schematic drawings of the prototype concepts, I was able to point to details important for the overall expressions. For instance, the positions of the connecting bolts in Tilting\Plate were decisive for its symmetric movements. Therefore, the bolts’ positions were clearly drawn onto the drawings as points.

Another example was how I discussed the scale and volume of the box with the production company. By describing the dynamics of the movements in Tilting\Plate and the magnitude of the planar elements’ spatial movements, we were able to determine the necessary height of the prototype. The volume of the containing box simply had to be large enough in order not to decrease the potential oscillations of the plate and frame.

The force concept was particular in play in Bending\Arches. To make sure that the required force would not exceed the capacities of the stepper motors, I continuously discussed with the production company how to improve the bended steel constructions. For instance, it was important that the construction would not require a force that would result in very slow movements, as I wanted it to behave in a quickly noticeable manner. Thus, the form language was not only important to describe and communicate the form composition of the construction, but also to illustrate the expressional and aesthetic goals of the prototypes.

The design practice with the form language has therefore been intertwined with the design values and expressive qualities I wanted the objects to incorporate. In more general terms, the form language helps to articulate a design tradition for shape-changing interfaces. The design vocabulary must not simply help to describe the dynamic forms, but also help to do so in a manner that takes the expectations and conventions for shape-changing interfaces

90

Ramifications of a form approach

into account. Shape-changing interfaces are characterised by physical actuation and we must therefore develop a form language, which encompasses these dynamics. The point here is that the form language cannot be arbitrarily defined, but must exist in a coherent relationship with design practice.

A form language is therefore an essential component in a form approach because it articulates the form tradition for a given discipline. A form language for shape-changing interfaces does not only include the basic form elements nor only the compositional principles and guiding rules for form compositions, but it is also a formalisation of the form tradition for actuated interfaces. The form language is a manifestation of the language-game for computationally actuated objects, because it frames the ways in which we can discuss shape-change. As Wittgenstein (1968) discussed, the language-game is the rules for our communication in our practice; or as Ehn writes: “To follow the rules in practice means to be able to act in a way that others in the game can understand. These rules are ‘embedded’ in a given practice from which they cannot be distinguished. They are this practice. To know them is to ‘embody’ them, to be able to practically apply them to a principally open class of cases” (Ehn, 1989, p. 106).

The language-game is what defines our way of conducting ourselves in practice. However, the language-game is, on the other hand, also created and continuously developed through our practice. Hence, the language-game has two dialectic sides, where it, on one side, is the result of practice, but on the other side, is also what constitutes practice. The form language, as a language-game, is what it is because of its use.

The way to further develop the form language is to use it. Through use, we can begin to expand it and appropriate it. With my design practice, I have begun to give meaning to the design vocabulary and come to realise its strengths and limitations.

For instance, I have realised how the form language is a flexible tool for understanding form dimensions. As we have just seen above as well as in the design experiments, it serves as a generative framework for posing questions and for offering inspiration. With the form compositional exercises it was apparent how the form language could also be used as an analytical entry point for grasping the form dynamics within the shape-changing objects.

Instead of using the design vocabulary in a strict and literal sense, it performs better as a flexible lens where the basic elements are used relevant to the task at hand. In this sense, the form components exist as conceptual elements (Ching, 2014) that the designer can call upon for understanding the dynamic compositions in given shape-changing interfaces.

Besides being conceptual, the form elements do not exist in isolation, but all are related.

We have seen how the force element can change the volume of an object. This was apparent 91

in both Bending\Arches as well as in the analysis of the Morphees interface (figure 7.2, p.

84). In turn, we have also seen how the position of the joints, as points, in the Thrifty Faucets determine how multiple forces applied to the object could curl its shape. It is not simply a matter of dealing with the basic elements in themselves, but rather to focus on the compositions of them. And it is in the attention to compositions that the design vocabulary becomes relevant for actuated interfaces. To explain the constructional requirements for Bending\Arches to the production company, video helped to capture these compositions of elements at play in the prototype. The video much clearer communicated how the planar elements should respond to applied forces, which made it possible for the craftsmen to make a suitable construction. It also helped the craftsmen to more qualified select the appropriate steel for the arches. Hence, the form language, here embodied by the video, does not only help designers deal with the expressional dynamics of an object, but is also concerned with the selection of appropriate materials. Or said in another way, the form language must be considered in relation to the materials from which the object is made.

Heidegger described these interrelationships between the different dimensions of the object through the four causes (see table 3.1, p. 41): “They differ from one another, yet they belong together” (Heidegger, 1977, p. 7). One dimension can therefore be discussed in itself, but is always connected to the three others; they exists holistically as complex wholes (Hove, 2010).

The form language must be considered in relation to the other dimensions of an object and visa versa. For instance, in the form analysis of Morphees we saw how the force element (as part of the causa formalis) was either insufficient to describe the actuation within the interface, or how the forces applied could only be understood based on the specific qualities of the materials (as part of the causa materialis).

Similarly, the ability of the arches to bend (the causa finalis) in Bending\Arches is possible due to the combination of its compositional form (the causa formalis) and the material properties of the steel (the causa materialis).

The fourth cause, the cause efficiens, is equally important in this relationship, since it is the designer who brings the three other causes together. In fact, in his example with the silver chalice, Heidegger writes: “The silversmith considers carefully and gathers together the three aforementioned ways of being responsible and indebted […] The silversmith is co-responsible as that from whence the sacrificial vessel’s bringing forth and resting-in-self take and retrain their first departure. The three previously mentioned ways of being responsible owe thanks to the pondering of the silversmith for the ’that’ and the ’how’ of their coming into appearance and into play for the production of the sacrificial vessel” (Heidegger, 1977, p. 7).

92

Ramifications of a form approach

This entails that a form approach to shape-changing interfaces requires the designer to be able to skilfully bring together these different elements. In the following, I will further elaborate on this necessity of design expertise.

Design expertise The purpose of a form perspective is to complement other approaches to shape-changing interfaces by drawing attention to the aesthetic and visceral expressions of computational actuation. The form language, however, does not in itself encompass the aesthetics of such objects. The design vocabulary merely describes the geometrics of the form as well as the temporal changes to these geometric compositions. With the form language we can describe the planar elements of the arches and how they are connected. We can also describe how it moves as its motors apply forces to the crossbar. We can, however, not determine with the form language if the form composition is balanced or if its movements are graceful. The form language is only able to describe the non-aesthetic properties of an object (Goldman, 2005).

Drawing on Sibley’s (1959) work on aesthetic concepts, Goldman (2005) uses the notions of non-aesthetic and aesthetic properties to be able to discuss the relation between an object’s composition and its expression. Where the non-aesthetic properties are the objective features, like the shapes, colours or scale, the aesthetic properties are the expressive qualities of an object.

Examples of aesthetic properties are beautiful, balanced, delicate, graceful, powerful, or even aggressive. What Goldman argues is that the non-aesthetic properties can be observed by any person, but the ability to point to the aesthetic properties of an object requires taste. What ties the non-aesthetic and aesthetic properties together is a person’s ability to make aesthetic judgements. However, the relationship between the two is not logical and is not universal: “That a painting contains pale colors and curved lines does not entail that it is graceful, although one might well point to those features of the painting to support a claim that it is graceful” (Goldman, 2005, p. 256).

We can therefore use non-aesthetic properties to support a claim about the aesthetic qualities in an object. Similarly, the form language can be used to articulate the aesthetics of a shapechanging object, but this does not entail that two moving plates will always form an interesting expression. Still, I have used the dynamics and the composition of the planar elements in Tilting\Arches to describe its interesting, fluid-like movements (cf. p. 72).

Throughout my design process, my aesthetic judgements have been central to the creative progress. I decided to have ten arches in Bending\Arches instead of only three, because I felt it gave it a more complex and interesting expression. I did not explore the box series 93

further, because I did not find it appealing or well-balanced. Design expertise, in this sense, is the ability to make aesthetic judgements as well as understanding the expressional means we have, as designers, to make a desirable form composition. However, Hove (2010) argues that designers often base their aesthetic work on tacit knowledge; that is, knowledge beyond what we can verbally articulate. Sharing and communicating aesthetic dimensions of form can therefore be difficult for designers. As an example, Hove writes: “To a large extent, formgiving deals with giving products ‘faces’ and expressions, and the difficulty with expressing your knowledge about this can be compared to the difficulties of articulating what is in a face you recognize, even though you can easily see the differences and perhaps even reproduce it visually” (Hove, 2010, p. 14, my translation).

In this light, the form language serves as a grasp for designers to begin to share their knowledge about form and compositions. With the basic elements, the designers are offered concepts to help them articulate their work. Moreover, Hove’s three concepts of lens, holistic take and appraisal is also an attempt to offer designers a framework for sharing their point of views. But whereas Hove’s three concepts articulate more abstract and advanced aspects of a designer’s practice, the proposed design vocabulary deals more closely with the objects at hand.

It is important that we recognise design expertise as a valid argumentation and as an essential aspect of the form approach if we are to improve form literacy about shape-changing interfaces.

Similarly, Pierce et al. (2015) have recently argued how expertise, or design authorship, is essential to a design practice within HCI. Through their discussions of the role of critical design, they argue that design expertise has to be a recognised as legit argumentation: “[D]esigners should not only be allowed an authorial voice, they should be encouraged to embrace, develop and enact one. By design authorship, we mean that design explicitly and significantly embodies intentions or ideas arising from a concern or curiosity of the designer. The product or service is a clear expression of the “voice” of the designer and her interpretation of social or cultural themes and concerns. While all design to some extent “embodies” such concerns and intentions, what’s unique here is that this is explicitly acknowledged and presented as such” (Pierce et al., 2015, p. 2088).

Design expertise must be acknowledged as part of design practice if the form approach is to become relevant for the domain of shape-changing interfaces. As argued previously, it is not that design decisions about objects’ forms and expressions are not already being made; thus it is not a matter of if shape-changing interfaces should encompass aesthetic dimensions, but rather how qualified we deal with the expressions of such objects.

The form language is only relevant for the language-game of aesthetics if we accept design expertise as an integral part of its practice. The role of the form language is to serve as the 94

Ramifications of a form approach

supporting claim for an aesthetic judgement. Also, Ehn notes how “[w]ith mastery of the practice of a certain language-game comes the freedom to follow the rules in totally unforeseen but still ‘correct’ ways” (Ehn, 1989, p. 107). As such, design expertise is not only a prerequisite for the practice in a form approach, but also for the continuous development of it. The expansion of the form language and the development of the understanding of aesthetics of shape-change can occur when designers use the design vocabulary in their practices to make and study forms.

Form studies The foundations for Itten’s basic design course at the Bauhaus were to make students explore and investigate different form compositions because “[e]xercises in composition of abstract forms serve in the improvement of thinking and at the same time the study of new means of representation” (Itten, 1975, p. 62). Form studies were essential in Itten’s teaching as a way to train the designers’ expertise in recognising compositional qualities and to develop their aesthetic sense. Similarly, form studies have to become an essential part of a form practice for shape-change.

In this thesis, I have engaged in different methods for working with form. I have made a series of low-fidelity formgiving experiments to explore various form compositions and to begin to expand my understanding of behaving expressions. The experiments were essential, not only to qualify the two prototypes, but as part of my specific form studies. I hope, by sharing my practice through this thesis, I can inspire other designers and show concepts, compositions, and expressions that can resonate with others and help develop their own practice.

Based on Schön’s writings, Hove (2010) argues how specific examples are crucial for designers to develop their form expertise. The use of examples should, however, not attempt to generalise beyond the specific examples themselves, but rather, through their specific circumstances and peculiarities, add to the designer’s repertoire of examples: “The reason for using cases is thus not because they can only say something about themselves, but rather, because aesthetic questions are best illustrated in complex wholes and because the breadth of examples on complex wholes is what in unison gives an understanding of a [form problem]. Paradoxically, it seems that the best way to general knowledge about aesthetic issues is through single cases” (Hove, 2010, p. 22, my translation) Löwgren has also noted how “[…] one of the most important elements of design ability is a repertoire of abstracted examples that is used to spawn ideas in new design situations. On a general level, it seems straightforward to claim that design can create knowledge of value to 95

other designers.” (Löwgren, 2007, p. 3). Therefore, designers must both engage in their own design experimentation as a way to develop this repertoire, but also share their examples with others. In this sense, we must establish a community for collaborative knowledge construction (Löwgren, 2007) to collectively develop form literacy about actuated interfaces.

Also inspired by the Bauhaus, Franinovic (2009) has proposed Basic Interaction Design to draw attention to the aesthetics of interaction. While her attention is on the interaction, which is beyond the scope of this thesis, our ambitions are overlapping: “Here, I have discussed the analytic and creative methods of Basic Interaction Design, which may offer an opportunity to better understand qualities that shape an aesthetic experience in interaction. This may allow designers to become more familiar and aware of their creative practice and of the aesthetic choices they make. As they are mastering their materials, they must not forget to consider the personal, social and cultural experiences that their products might engender.“ (Franinovic, 2009, p. 58).

Franinovic points to both analytical and creative methods as central for a basic approach for dealing with the aesthetics of interactive objects. In this thesis, both types of methods have been apparent and they are the basis of my form study. Both the formgiving experiments and the prototypes were creative methods, where my work was the driver of exploring form dynamics of shape-change. The form compositional exercises, on the other hand, were an example of an analytical method for better understanding other’s work. While there may be other relevant methods, the point here, is that designers should both contribute to the collective form knowledge as well as use others’ work to become better designers themselves.

Franinovic further emphasises the personal and social values of designers. I have introduced Hove’s three concepts as a framework for articulating the designer’s personal way of doing design. However, social and cultural aspects of the aesthetics of shape-changing interfaces still need to be addressed to further deepen the understanding of how to best possibly utilise actuation in interactive objects.

The form approach is still in its early development. I propose the beginning of what I hope can be an on-going debate about the form and expressional dimensions of shape-changing interfaces. New form studies and theoretical work must engage in this debate to further develop it and to qualify the language-game of aesthetics within actuated interfaces. It would be particularly relevant to combine the focus from this thesis with Franinovic’s work on aesthetic interaction. This could help us to understand how the form language relates to users’ interactions with shape-changing objects. It seems straightforward to claim that there is a direct relation between the interaction and the force elements in the object. This relation is characterised by an input-output causality where a users’ interaction would translate

96

Ramifications of a form approach

into changes in the force components. Albeit not being the primary focus of this thesis, my prototypes did include basic interactions. For instance, Tilting\Plate had two potentiometers.

The interaction gestalt, here, was configured to mimic the dynamic form composition of the prototype. The inner plate and the second frame are able to rotate symmetrically around each of their point of origin; and similarly, the potentiometers were configured to have a range corresponding to the two outer positions of the oscillations. This leads me to believe that there may also be other aspects of the form language that could relate to the interaction. In this example, it was the overall dynamic of the form composition that was recurring in the use of potentiometers. Of course, other interaction gestalts could have been made; still, it seems appropriate to further explore if the form language can help inform design decisions about the interaction as well. For instance, Djajadinigrat et al. (2004) argued how a form focus would allow to improve the design of the interaction gestalt through attention to physical affordances. Similarly, form studies must begin to explore the potential relations between the form language and interaction.

Furthermore, I believe it is important for the language-game of aesthetics to appreciate and encourage qualitative argumentation and the holistic focus on the complex wholes of forms.

Similarly, Hove argues: “One of the main points […] is the influence of the entirety and nexus on the experience of a phenomenon. This means that studies that try to get general answers about the individual instruments will have very limited validity” (Hove, 2010, p. 12, my translation) Rather than searching for truths – or general insights – the purpose of form studies is to strengthen designers’ sensibility or intuition about aesthetic problems and to qualify their language for articulating their design decisions. Instead of building a community around what we now for certain, we need to share our experiences and knowledge from our practices to collaboratively expand our understanding of the expressive qualities of shape-changing forms.

Closing remarks Current approaches to actuated interfaces have focused on functional and technological aspects of shape-change, as well as actuation as a core property of smart materials. Also, recent studies have begun to investigate the user experience of shape-changes. However, these current approaches do not encompass the immediate aesthetic dimensions of shapechanging interfaces. In an attempt to draw attention to this, I have proposed a form approach as a strategy for qualifying the work on behaving expressions in actuated objects.

97

Shape-changing objects always contain visceral narratives about themselves depending on their form compositions. The purpose of a form approach is therefore to improve designers’ awareness of form and to help them make qualified aesthetic choices in the design process.

Moreover, explorations of shape-changing interfaces as abstract expressionals also serve to imagine new applications and functionalities of shape-changing interfaces.

The core of a form approach is a form language because it can both serve as a generative framework for developing shape-changing objects, as well as function as a communicative and descriptive tool for understanding others’ design work. In this thesis, I have begun to define the basic elements with which we can describe the temporal and physical dynamics of shape-changes in a computational object. Furthermore, I have described how the design process in a form approach is formgiving, in which the designer reveals the form of an object in a holistic process of bringing together the designer’s visions, the materials properties, its purpose as well as other dimensions of the object.

The last half of my thesis has exemplified a methodology for a form practice in interaction design; that is, formgiving experimentation, prototype production as well as descriptive form analyses.

My thesis is the preliminaries to a form practice for shape-changing interfaces and other aspects of it have yet to be explored. In general, more form studies are needed to further develop the form language. A better understanding of the basic elements and examples on how they inform design practice are needed, and in particular, a better understanding of their interrelationship. In particular, our understanding of the force element must be developed.

The force element has been useful in this thesis to point to shape-changing objects potential of actuation, rather than their actual movements. As I have hinted, the force may need additional attributes, such as speed, acceleration, types of forces, to get closer to describing the concrete shape-changes within an object. Furthermore, we must also move beyond the primary elements and begin to develop compositional rules and guiding principles for shapechanging interfaces. Not rules in a normative sense, but as rule-of-thumbs (Hove, 2010) or best practices. Such inquiries could entail open-ended questions, like how do we address the temporal rhythms of forces in a shape-changing interfaces? or are certain form compositions more appropriate to produce harmonious expressions than others?

As I have argued, a form approach must acknowledge the expertise of the designer. However, it is important to stress that the form approach should be considered as complementary to current other approaches; and as such, “[...] acknowledging design authorship in no way demands a rejection of user-centered design. Users are and should remain an important consideration. […] When designers are allowed a clear voice, they are liberated to speak about topics and issues that are difficult to address in a [user-centered design] framework” (Pierce 98

Ramifications of a form approach

et al., 2015, p. 2088). The form approach seeks to liberate designers to be able to discuss the aesthetics of objects based on their own aesthetic judgements, without neglecting the importance of both technological, material and user experience advancements.

To discuss how this proposed approach is governed by different norms and traditions than the other approaches to shape-changing interfaces, I have found Wittgenstein’s concept of language-game useful. With language-game, it has been possible to point to the need for certain rules for the design discourses to allow to work, describe, and discuss aesthetic dimensions of computationally actuated objects. The language-game for a form approach allows designers, based on the aesthetic judgements, to describe expressive qualities of an object by pointing to compositional structures within it.

I have proposed the point, the line, the plane, the volume, and the force as the primary elements for a form language for shape-changing interfaces. To begin to unfold the potentials of these primary elements, I have engaged in an exploratory design process, where I have experimented with multiple form compositions, which have resulted in two prototypes.

While the design vocabulary has proved useful, more work is needed to further develop the form approach. However, I have shown the ways a form language can help interaction design better deal with the expressional dimensions of actuation in interfaces. In particular, I hope that my work can be an inspiration for others and serve as a stepping stone for the further development of form literacy about shape-changing interfaces in interaction design.

99

References

Arvola, M. & Artman, H. (2007) Enactments in Interaction Design: How Designers Make Sketches Behave. Artifact. 1 (2), 106–119.

Baskinger, M. & Gross, M. (2010) COVER STORY: Tangible interaction = form + computing.

Interactions. 17 (1), 6–11.

Buxton, W. (2007) Sketching user experiences : getting the design right and the right design.

Amsterdam; Boston, Elsevier/Morgan Kaufmann.

Ching, F. (2014) Architecture: form, space, & order. Fourth edition. Hoboken, New Jersey, Wiley.

Coelho, M. & Zigelbaum, J. (2011) Shape-changing interfaces. Personal and Ubiquitous Computing. 15 (2), 161–173.

Djajadiningrat, T., Wensveen, S., Frens, J. & Overbeeke, K. (2004) Tangible products: redressing the balance between appearance and action. Personal and Ubiquitous Computing. 8 (5).

Ehn, P. (1989) Work-oriented design of computer artifacts. Doctoral Dissertation. Stockholm, Arbetslivcentrum.

Feijs, L. & Kyffin, S. (2006) A taxonomy of semantic design knowledge. In: Loe Feijs, Steven Kyffin, & Bob Young (eds.). Proceedings of DeSForM 2006. 2006 Eindhoven, The Netherlands, Koninklijke Philips Electronics N.V. pp. 70–83.

Franinovic, K. (2009) Toward Basic Interaction Design. Elisava Temes de Disseny Journal.

(25), 52–58.

Goldman, A. (2005) The Aesthetic. In: The Routledge Companion to Aesthetics by Lopes, D.M. and Gaut, Berys. Routledge. pp. 255–266.

101

Grönvall, E., Kinch, S., Petersen, M.G. & Rasmussen, M.K. (2014) Causing commotion with a shape-changing bench: experiencing shape-changing interfaces in use. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2014 ACM Press. pp. 2559–2568.

Hallnäs, L., Jaksetic, P., Ljungstrand, P., Redström, J., et al. (2001) Expressions; Towards a Design Practice of Slow Technology. In: Proceedings of Interact 2001. Tokyo, Japan.

Hallnäs, L. & Redström, J. (2002a) Abstract information appliances. In: DIS ’02 Proceedings of the 4th conference on Designing interactive systems: processes, practices, methods, and techniques. [Online]. 2002 ACM Press. pp. 105–116.

Hallnäs, L. & Redström, J. (2002b) From use to presence: on the expressions and aesthetics of everyday computational things. ACM Transactions on Computer-Human Interaction.

9 (2), 106–124.

Hallnäs, L. & Redström, J. (2006) Interaction design: foundations, experiments. Borås, The Interactive Institute ; The Textile Research Centre, The Swedish School of Textiles, University College of Borås.

Hallnäs, L. & Redström, J. (2001) Slow Technology – Designing for Reflection. Personal Ubiquitous Comput. 5 (3), 201–212.

Heidegger, M. (1977) The question Concerning technology. In: The question concerning technology, and other essays. New York, Garland Pub. pp. 3–35.

Hemmert, F., Hamann, S., Löwe, M., Zeipelt, J., et al. (2010) Shape-changing mobiles: tapering in two-dimensional deformational displays in mobile phones. In: CHI ’10 Extended Abstracts on Human Factors in Computing Systems. 2010 ACM Press. pp.

3075–3080.

Horev, O. (2006) ‘TALKING TO THE HAND’ An exploration into shape shifting objects and morphing interfaces. Master’s Thesis. Italy, Interaction Design Institute Ivrea.

Houde, S. & Hill, C. (1997) What do Prototypes prototype? In: M Helander & P. Landauer (eds.). Handbook of Human Computer Interaction. Amsterdam, Elseveier Science. pp.

367–381.

Hove, H. (2010). FORMSPROG: Formidling af designerens formgivningserfaring. Copenhagen Working Papers on Design, No. 4. Copenhagen, Denmark: The Danish Design School.

102

References

Hubert, D. (2009) Heidegger on Gaining a Free Relationship to Technology. In: David M. Kaplan (ed.). Readings in the philosophy of technology. 1st ed. Lanham, Rowman & Littlefield Publishers. pp. 53–62.

Itten, J. (1975) Design and form: the basic course at the Bauhaus and later. New York, Wiley.

Jung, H., Altieri, Y.L. & Bardzell, J. (2010) Computational objects and expressive forms: a design exploration. In: CHI ’10 Extended Abstracts on Human Factors in Computing Systems. 2010 ACM Press. pp. 3433–3438.

Jung, H. & Stolterman, E. (2011) Form and materiality in interaction design: a new approach to HCI. In: CHI ’11 Extended Abstracts on Human Factors in Computing Systems. 2011 ACM Press. pp. 399–408.

Koskinen, I.K., Zimmerman, J., Binder, T., Redström, J., et al. (2011) Design research through practice : from the lab, field, and showroom. Waltham, MA, Morgan Kaufmann/ Elsevier.

Kozel, S. (2007) Closer: performance, technologies, phenomenology. Leonardo. Cambridge, Mass, MIT Press.

Kretzer, M. (2010) SHAPESHIFT. [Online]. Available from: https://dl.dropboxusercontent. com/u/1325890/shapeshift_booklet.pdf [Accessed: 19 April 2015].

Kwak, M., Hornbæk, K., Markopoulos, P. & Bruns Alonso, M. (2014) The design space of shape-changing interfaces: a repertory grid study. In: Proceedings of the 2014 Conference on Designing Interactive Systems. 2014 ACM Press. pp. 181–190.

Larsen, H.S. (2015) Tangible participation-Engaging designs and design engagements in pedagogical praxes. Doctoral Dissertation. Lund University.

Lim, Y.-K., Stolterman, E. & Tenenberg, J. (2008) The anatomy of prototypes: Prototypes as filters, prototypes as manifestations of design ideas. ACM Trans. Comput.-Hum. Interact. 15 (2), 1-27.

Löwgren, J. (2007) Interaction design, research practices and design research on the digital materials. [Online]. Available from: http://webzone.k3.mah.se/k3jolo.

Löwgren, J. & Stolterman, E. (2004) Thoughtful interaction design a design perspective on information technology. Cambridge, Mass., MIT Press.

103

Mazé, R. & Redström, J. (2005) Form and the computational object. Digital Creativity. 16 (1), 7–18.

Nørgaard, M., Merritt, T., Rasmussen, M.K. & Petersen, M.G. (2013) Exploring the design space of shape-changing objects: imagined physics. In: Proceedings of the 6th International Conference on Designing Pleasurable Products and Interfaces. 2013 Newcastle, ACM Press. pp. 251–260.

Norman, D.A. (2005) Emotional design: why we love (or hate) everyday things. New York, Basic Books.

Parkes, A. & Ishii, H. (2009) Kinetic sketchup: motion prototyping in the tangible design process. In: Proceedings of the 3rd International Conference on Tangible and Embedded Interaction. 2009 ACM Press. pp. 367–372.

Parkes, A., Poupyrev, I. & Ishii, H. (2008) Designing kinetic interactions for organic user interfaces. Communications of the ACM. 51 (6), 58–65.

Pedersen, E.W., Subramanian, S. & Hornbæk, K. (2014) Is my phone alive?: a large-scale study of shape change in handheld devices using videos. In: CHI ‘14 Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, ACM Press. pp. 2579– 2588.

Pierce, J., Sengers, P., Hirsch, T., Jenkins, T., et al. (2015) Expanding and Refining Design and Criticality in HCI. In: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. 2015 Seoul, Korea, ACM Press. pp. 2083–2092.

Poupyrev, I., Nashida, T. & Okabe, M. (2007) Actuation and tangible user interfaces: the Vaucanson duck, robots, and shape displays. In: Proceedings of the 1st international conference on Tangible and embedded interaction. 2007 Louisiana, USA, ACM Press. pp. 205–212.

Raffle, H.S., Parkes, A.J. & Ishii, H. (2004) Topobo: a constructive assembly system with kinetic memory. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2004 ACM Press. pp. 647–654.

Rasmussen, M.K., Grönvall, E., Kinch, S. & Petersen, M.G. (2013) ‘It’s alive, it’s magic, it’s in love with you’: opportunities, challenges and open questions for actuated interfaces.

In: Proceedings of the 25th Australian Computer-Human Interaction Conference: Augmentation, Application, Innovation, Collaboration.2013 ACM Press. pp. 63–72.

104

References

Rasmussen, M.K., Pedersen, E.W., Petersen, M.G. & Hornbæk, K. (2012) Shape-changing interfaces: a review of the design space and open research questions. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2012 New York, NY, USA, ACM Press. pp. 735–744.

Robles, E. & Wiberg, M. (2010) Texturing the ‘material turn’ in interaction design. In: Proceedings of the fourth international conference on Tangible, embedded, and embodied interaction. 2010 ACM Press. pp. 137–144.

Ross, P.R. & Wensveen, S. (2010) Designing Behavior in Interaction: Using Aesthetic Experience as a Mechanism for Design. International Journal of Design. 4 (2), 3–13.

Roudaut, A., Karnik, A., Löchtefeld, M. & Subramanian, S. (2013) Morphees: toward high ‘shape resolution’ in self-actuated flexible mobile devices. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2013 ACM Press. pp.

593–602. soma architecture (2011) soma architecture - theme pavilion. [Online]. 2011. somaarchitecture.com. Available from: http://www.soma-architecture.com/index. php?page=theme_pavilion&parent=2# [Accessed: 23 April 2015].

Schön, D. A. (1983). The reflective practitioner: how professionals think in action. New York: Basic Books.

Sibley, F. (1959) Aesthetic Concepts. Philosophical Review. 68 (4), 421–450.

Strozyk, E. (2011) Elisastrozyk.de. [Online]. 2011. ELISA STROZYK. Available from: http:// www.elisastrozyk.de/seite/embossed.html [Accessed: 3 February 2015].

Studio Roosegaarde (2011) Lotus info. [Online]. 2011. Studioroosegaarde.nl. Available from: http://www.studioroosegaarde.nl/project/lotus/info/ [Accessed: 17 May 2014].

Togler, J., Hemmert, F. & Wettach, R. (2009) Living interfaces: the thrifty faucet. In: Proceedings of the 3rd International Conference on Tangible and Embedded Interaction.

2009 ACM Press. pp. 43–44.

Vallgårda, A. (2009) Computational composites : understanding the materiality of computational technology. Doctoral Dissertation. Copenhagen, IT University of Copenhagen.

105

Vallgårda, A. (2008) PLANKS: a computational composite. In: Proceedings of the 5th Nordic conference on Human-computer interaction: building bridges. 2008 Lund/IKDC, Sweden, ACM Press. pp. 569–574.

Vallgårda, A. (2013) Giving form to computational things: developing a practice of interaction design. Personal and Ubiquitous Computing. [Online] 18 (3), 577–592.

Vallgårda, A., Pedersen, N.M., Vizer, E.E. & Winther, M. (in press) Temporal Form in Interaction Design. Manuscript, Recently submitted to a journal.

Vallgårda, A. & Redström, J. (2007) Computational composites. In: CHI ‘07 Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2007 ACM Press. pp. 513–522.

Vallgårda, A. & Sokoler, T. (2010) A Material Strategy: Exploring Material Properties of Computers. International Journal of Design. 4 (3), 1–14.

Westerlund, B. (2005) Design space conceptual tool – grasping the design process. In: Proceedings for ‘In the Making’, Nordes, the Nordic Design Research Conference. 2005 Copenhagen.

Wittgenstein, L. (1968) Philosophical investigations. Oxford, Basil Blackwell.

Yao, L., Niiyama, R., Ou, J., Follmer, S., et al. (2013) PneUI: pneumatically actuated soft composite materials for shape changing interfaces. In: Proceedings of the 26th annual ACM symposium on User interface software and technology. 2013 United Kingdom, ACM Press. pp. 13–22.

Young, H.D. & Freedman, R.A. (2004) University physics: with modern physics. AddisonWesley series in physics. 11th, International Ed. San Francisco, Pearson, Addison Wesley.

Zainal Abidin, S., Sigurjonsson, J., Liem, A. & Keitsch, M. (2008) On the role of formgiving in design. In: A Clarke, M Evatt, P Hogarth, J Lloveras, et al. (eds.). New Perspectives in Design Education, Proceedings of the Engineering and Product Design Education.

2008 Barcelona, Spain.

Zigelbaum, J. & Labrune, J.-B. (2009) Some Challenges of Designing Shape Changing Interfaces. In: Proceedings of CHI 2009. 2009 Boston, USA.

106

B E H A V I N G EXPRESSIONS Morten Winther www.mortenwinther.dk