Second Year Report Jose Rojas 1 August 2007

1

2

The Social-cultural Construction of Ubiquitous Computing Environments In this report we explore the nature of Ubiquitous Computing and introduce a definition that seeks to harmonise current views in the field. We discuss the role of culture in HCI and identify some popular approaches to this issue by the community. We also introduce a view of culture that adds a dynamic dimension to this concept and loosens its associations with highly defined geographies. We describe our ongoing research split into three interventions aimed at understanding the process whereby ICTs fulfil one of the conditions of our view of Ubicomp environments.

1 Introduction “The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it.” Mark Weiser [29]. Perhaps no other statement has received more attention in the Ubicomp community than this introductory paragraph in Mark Weiser’s seminal article “The Computer for the 21st Century” [29]. We take this statement to emphasize what some authors have suggested [7, 16] and what constitutes the underlying thesis of our research, that is, that the dominant view in the development of computing technology in general, and Ubicomp in particular, seems to be North American oriented. This suggestion leads us to question the suitability of the standard “North American view” of Ubicomp to other socio-cultural environments. Are there any alternative visions for the development of Ubicomp because of different circumstances, needs and expectations towards computing technology? How do we understand those other socio-cultural contexts? Through what process are computing technologies appropriated by their users? What is the role, if any, of the socio-cultural context in shaping this process of appropriation? And how can we apply that understanding to the construction of alternative views of Ubicomp?

In this report we present our research aimed at exploring these questions. We begin with a discussion of the nature of Ubicomp. We highlight the central role of human actions in achieving Ubiquitous Computing environments and how the socio-cultural context is an important element of this achievement. We identify several approaches used in the HCI community to study the relationship between culture and computing technologies. Later, we describe four interventions in our ongoing study aimed at understanding the process whereby Ubiquitous Computing environments may be achieved. We finish highlighting the limitations and some preliminary observations from our first intervention. 2. What is Ubiquitous Computing? For the past ten years we have seen an increased interest in the technological ideal known as Ubiquitous Computing initially proposed by Mark Weiser in his article “The Computer for the 21st Century” [29]. From a simplistic point of view, Ubicomp suggests a world where computers are so abundant in our lives and so specialized in their functions that they become part of the very landscape of our everyday life enhancing it through some form of computation [29]. This view seems to arise from a narrow reading of Weiser’s Ubicomp vision, which, having been formulated from anthropological studies of work life [29, 32], is commonly characterized as an infrastructure fitted to satisfy the communication and information needs of an office environment. A deeper search for the meaning of Ubiquitous Computing provides us with more elements of analysis as we struggle to clarify the meaning of Ubiquitous Computing. For instance, Weiser’s statement on the intrinsic qualities of “profound technologies” has stirred different interpretations within the HCI community. His untimely death precluded him from further elaborating on his particular vision of Ubicomp. Weiser suggested, however, some guidelines as to what Ubicomp should achieve. In his later writings he insisted on the invisibility and calmness of Ubicomp and how, in his view, Ubicomp achieves this unobtrusiveness by moving from the periphery to the centre of attention and back [31]. The nature of the invisibility and calmness proposed by Weiser is one that so far remains elusive, if not misunderstood. We believe, however, that it is of the utmost importance to arrive to a better understanding of what Ubicomp should be. To explore the nature of Ubicomp and support our thesis that this ideal is 2

socio-culturally constructed, we present below two perspectives of what has been argued is one of its more fundamental elements, the context. From a pure technical perspective, the nature of Ubicomp was already explored by Weiser [30] who identified low power devices, network protocols, interaction substrates, applications, privacy, and computational methods as some of its constitutive elements. Lyytinen and Yoo [23] further synthesize this view affirming that Ubicomp is achieved through the integration of two types of computation: mobile computing that suggests the ability to carry a computational device everywhere one goes, and pervasive computing or computational devices populating the environment and provided with models of the very environment they occupy in order to react and provide ad-hoc resources to other computational devices in their vicinity. However, while this view recognizes the challenge of building “virtual or physical models of the physical and social/cognitive environment” [23], it seems to a) reduce Ubicomp to a mere technical problem, that is, to a problem of having the proper means to capture the context where computers may operate, and to b) downplay the role of a most important element of this equation, the human variable. We argue that humans and their everyday practices are essential elements to achieve Ubicomp by defining the very form that Ubicomp may embody. The ongoing discussion on the importance of capturing the context to achieve Ubicomp seems to have polarized the research community. Broadly speaking, we could distinguish two schools of thought on this issue. We will simple call them the the technical school and the social school. The technical school actively seeks the means to capture and represent the context in and through computational devices. Since the work of this school seems to be the most popular and readily available in the proceedings of any relevant journal or conference, we choose not to provide any example of this school. On the other hand, the social school acknowledges the complexity of the human variable and how it can not be reduced to a set of parameters that can be encoded and formalised. In Plans and Situated Actions [27] Suchman refuted the notion that humans can be conceived as entities acting under a master plan that defines all their actions as always meaningful and explainable. Humans are always situated, that is, constantly changing and adapting to respond to their environment, thus, a

3

preconceived plan cannot predict every action humans do at any given time [27]. Dourish supports this view arguing that the problem of context in Ubicomp is not one of representation, but one of interaction [15]. Making sense of the world, then, is an intrinsic human ability exercised continually and endlessly. This human ability to successfully negotiate a changing environment in a myriad ways is what Dourish argues is impossible to reduce and encode in a machine. Suchman and Dourish arguments suggest that Ubicomp and its disappearance into the fabric of everyday life cannot be achieved only through mechanisms to capture, encode and represent the context, but by acknowledging the role it may play in people’s lives. This idea is further advanced by Tolmie et al. [28] who observe the invisibility of Ubicomp may not be referring to making the computer vanish from sight, but rather to “invisibility in use.” This suggests a change of focus from trying to produce perceptually invisible computers that “understand” their context towards a reevaluation of what it is that renders our actions as ordinary and non-rationalised, in a sense, “invisible.” According to this view, the mere electronic enhancement of items in the environment through some form of computation cannot produce Ubicomp. Along with Chalmers we argue that is actions or practices that eventually render computational objects “invisible.” Thus, ubiquitous computing is not about the “disappearing computer,” but about providing the means through which humans can make sense of computers at any moment to integrate them into their lives. Chalmers sustains this view by reinterpreting what he considers the original aims of Weiser’s Ubiquitous Computing [11, 12] aligning them to Heidegger’s phenomenology which, in Chalmers’ view, provides a better interpretation of the nature of Ubicomp. Heidegger suggested that non-rationalised use (i.e. ready-to-hand) and rational, objectified, abstract use (i.e. present-at-hand) are two necessary elements to interpret and interact with the world. In other words, human beings necessarily need to make sense of the environment where they exist (rationalization) in order to achieve a tacit level of proficiency at interacting with it [11]. For Chalmers, then, Ubicomp is achieved through a cycle of rationalization and use that eventually renders computing technology as part of the landscape of everyday life. In arguing that Ubicomp is not achieved only through computational devices in isolation, but through a cycle of rationalisation and use, we would like to suggest the

4

importance of an element we believe shapes the context where this cycle takes place. We refer to the role of socio-cultural forces in shaping the use (and abuse) of computing technologies and in determining their relevance, utility and adequacy for a particular milieu. We believe this is an important topic of research because it is both timeless (i.e., the interplay between society, culture, and technology is necessarily a dynamic process) and timely (i.e., computing technologies, including those “beyond the desktop,” continue reaching non-traditional users in non-traditional settings continually challenging the status quo of HCI). We believe the socio-cultural context is a relevant area of research which, in exploring the vision of Ubicomp, will allow us to open the discussion to other interpretations away from the North American milieu and from the models of work and leisure around which computing technologies have traditionally been built. Having described the two prevailing views of Ubicomp, that is, the technical and the social schools, we would like to suggest that in order to achieve Ubicomp it is necessary to harmonise the preoccupations of both camps, thus, our definition of Ubicomp requires the presence of two circumstances: First, Ubicomp requires the presence of computational devices which we will call non-conventional computational devices. We decide to make this differentiation for two reasons. Firstly, to highlight the fact that Ubiquitous Computing necessarily refers to some form of electronic computation, thus, other technologies of everyday life like paper-based technologies, mechanical-based technologies, and other such technologies, no matter how pervasive, can not be classified as “Ubiquitous Computing.” Secondly, we call them non-conventional computational devices to differentiate Ubiquitous Computing technologies from regular computing technologies; the main difference being that Ubicomp technologies have the added capacity of being automatically tracked and being able to react automatically to its environment. This condition declares that the mere fact of being a computational device is not reason enough to be considered a Ubiquitous Computing device or item. The computational device thus enhanced can be a static device (see the concept of pervasive computing above), an ambulatory device (see the concept of mobile computing above), or both. In other words, it has the ability to react to a changing environment and to affect (and be affected by) other similar devices that might be

5

found permanently or transiently in the environment. From this perspective, mobile phones, PCs, PDAs, MP3 players, and other similar computational devices already populating the environment are not Ubiquitous Computing because they neither have the ability of being tracked nor they react to its environment. Second, a Ubicomp device is an item or a collection of items that, having satisfied the previous condition, have achieved wide and deep penetration into the lives of at least a particular group of people. This characteristic suggests that even when the previous condition has been satisfied it is not of itself reason enough to achieve Ubiquitous Computing. To achieve the status of U0biquitous Computing, a community of people, regardless of its size or location, has to have been able to integrate the previously described computing technologies into “the fabric of their everyday lives.” For such a group then, the status of Ubiquitous Computing has been achieved. From our perspective, Ubicomp is neither a technology, nor a collection of gadgets, but a “status of things,” an environment characterised by the integration of the above mentioned conditions. As we see it, this definition of Ubicomp maintains the original ideals of Mark Weiser’s while keeping the door open for the development of the necessary nonconventional computing technologies that satisfy the first condition mentioned above. Accordingly, our view of Ubicomp takes into account the human variable and recognises that without its intervention Ubicomp could never come to pass. Our view of Ubicomp neither attempts to dictate how the technical aspects of Ubicomp are to be achieved, nor attempts to claim to know the social process whereby technology acceptance occurs; it simply recognises that both processes should co-exist to achieve Ubicomp. As could be expected, our definition of Ubicomp leaves room to accommodate our own research agenda. Consequently, we position our research efforts to pursue a better understanding of the second condition to achieve Ubicomp. We argue that the socio-cultural context influences the way a non-conventional computational device is used, adopted and appropriated, and that a better understanding of this process will help us facilitate or expedite the achievement of multiple Ubiquitous Computing environments. We now turn our attention to consider issues particular to the socio-cultural environment.

6

3. What in the World is Culture? “Each man calls barbarism what is not his own practice for indeed it seems we have no other test of truth and reason that the example and pattern of the opinions and customs of the country we live in.” Michel de Montaigne In this section we will not attempt to provide an exhaustive survey of ideas defining culture. We will only focus on some concepts that, after a survey of the literature dealing with culture in HCI, seem to be relevant to frame our work; we are only drawing from those ideas that can provide some basis to our claim that Ubicomp is socio-culturally defined. The literature is full with definitions of what culture is. According to Kamppuri and Tukiainen [21] culture is integrated by both a subjective and an objective dimension; in other words, culture is integrated by elements that are both found inside the head and in the world around us. Subjective elements are represented by values and norms. Values are conceived as “ideas about what in life seems important” [33]; Norms are the rules of behaviour that a society enforces upon its members to secure the permanence of what a particular group considers important to preserve, its values. Norms officially enforced achieve the status of laws [33]. The objective dimension of culture is thought to be expressed through institutions and artefacts. Institutions are the structures whereby “values and norms are transmitted” [33], while artefacts are the tangible, physical elements built by a particular group to express its values and norms. For Bourges-Waldegg and Scrivener a culture can only be found among beings that have developed a representational system. Representational differences are, therefore, the main reason why different cultures exist [10]. In its simplest conception, culture is thought of as a fixed, monolithic entity. This idea is nowhere clearer than in Hofstede’s definition of culture as “the collective programming of the mind which distinguishes the members of one group of people from another” [19]. However, even Hofstede, seems to deny the homogenising implications of his definition by conceding that cultures cannot be attached to a particular geographical setting since cultures can be found both as integrated social systems and as smaller parts of those integrated social systems. National and ethnic cultures are, under this view, examples of integrated social systems because they seem

7

to cover all individuals (usually) within a common geographical area. On the other hand, gender, generational, work and organizational cultures are not seen at the same level since they only cover a limited number of individuals who belong to a larger integrated social system [19]. The second characteristic of culture usually found in the literature is its fixedness, that is, its unchanging nature. Although the symbols (e.g., words, images, gestures, and objects) of a culture may change (its objective dimension), its values remain static (its subjective dimension). For Hofstede [19], the slow pace of change in values accounts for the emergence, over hundreds of years, of what he has called cultural dimensions. This rather static nature of culture ensures the ever-lasting “relevance” of Hofstede’s model. Furthermore, a rigid and simplistic definition of culture like the one embodied in Hofstede’s model, might account for its popularity within the HCI community, since it provides a “tested,” ready-made-model that can be used as a theoretical platform to launch a research enquiry – Some would argue that the uses given to Hofstede’s model seem to reflect more of a concern with finding a suitable marketing model that can be attached to a geographical area [1], than a real attempt to understand people in their regular settings (see for instance [14]). The view of culture as a static entity has also been challenged. For instance, BourgesWaldegg and Scrivener [10] argue that one of the results of humans’ unique ability to a have a representational system is their capacity to continually assign and reassign meanings to a changing environment. Furthermore, from a progressive perspective, an immovable culture would prevent a society from adapting to a changing environment, for instance, by preventing it from learning new ways of carrying about tasks and solving problems [21] (Of course, this progressive perspective assumes the inevitability of a changing environment which, incidentally, seems to be, to a large degree, a by-product of human activity). Under this view cultures are indeed constantly changing and developing – This dynamic view of nature, however, makes us wonder if the world of tomorrow, because of its constant change in cultures’ values, norms, institutions and artefacts, would appear incomprehensible if we were to suddenly appear in a distant future. While we refrain from proposing a formal definition of culture, it seems to be clear that what we need in HCI is a view of culture that goes beyond a collection of facts

8

and figures about a particular group [1], and one that recognises that the “problem of culture” in HCI is perhaps a problem of diversity [2]. We argue that we seem to be in need of a definition of culture that allows us to understand first, the relationship established between the subjective and objective elements of diverse backgrounds (i.e., culture); second, the influence of this relationship in people’s ability to mediate a changing context through a dynamic assignment of meanings to spaces and places. Furthermore, our view of culture should take into account that those spaces and places are populated by other human agents who, having formed their own relationships between their own subjective and objective elements, also mediate a changing context through “different lenses” than ours. An analysis of current approaches in HCI to culture will be useful in illustrating the need for an approach to culture such as the one we just described.

3.1 Dealing with the Socio-Cultural Context in HCI It seems to be that many issues are still ignored within the field of HCI regarding the interplay between society, culture and ICTs. According to a survey conducted by Kamppuri et al. [20] including five of the most important venues of publication within HCI (Journal of Human-Computer Interaction, International Journal of HumanComputer Interaction, International Journal of Human-Computer Studies, Interacting with Computers, and the Proceedings of ACM CHI ) from the years 1990 to 2005, the study of culture represents, at most, a marginal area of research that has been slowly gaining some awareness within the field for the past nine years. According to their survey, the amount of full papers published dealing with culture has barely reached 28; with the large majority of them (24) published in the period between 1998 and 2005. This number only represents 0.9% of the 3,286 published papers in these journals during the time frame considered. More interestingly, their survey indicates that among the published papers dealing with culture a full 40% does not even try to provide a working definition of culture to frame the research being presented. When a definition of culture is in fact presented, the HCI community seems to be resorting to the same sources of cultural theory including cross-cultural psychology, marketing and management practices, and the

9

cultural dimensions of Geert Hofstede [19] and Edward Hall [18]. Notably, Hofstede’s cultural dimensions have been used by seven studies in this survey, which would be of no surprise if it were not for the fact that Hofstede’s work has been criticized on the grounds of its ecological validity, and thus is applicability [25]. Anthropological sources in the study of culture in HCI are, at this time, still rare. For Kamppuri and Tukiainen [21] this circumstance has an explanation. It seems to be that the very epistemological tradition upon which HCI was established, that is, cognitive science (which in turn was based in psychology, neuroscience, linguistics, and AI), precluded the field from taking issues of culture into consideration in the development of its body of knowledge. According to these authors, the present condition of HCI would have been different if one of the additional disciplines considered in the development of cognitive science, anthropology, would have had the opportunity to claim a stake in the formation of HCI. However, it seems to be that the lack of a formal, quantifiable body of knowledge prevented anthropology, and other social disciplines, from contributing to the early development of HCI. In the face of such a situation, we in the HCI community appear to be ill-suited to address issues of culture. Such a circumstance is illustrated by the standard approaches of the community when trying to research the influence of culture in the design, use and evaluation of ICTs. In general, four approaches can be distinguished in these studies [21]: 3.1.1 The Cognitive Approach.

As the name implies, this approach draws heavily from the cognitive foundation of HCI, thus, it focuses on the cognitive traits of users and the usability of interfaces. Traditional questions of usability like efficiency, learnability, number of errors, etc. are the major concern of this approach to culture in HCI. As expected, the methods of this approach are empirical, drawing mainly from cross-cultural psychology in the search for those differences that, under certain conditions, appear to be the consequence of different socio-cultural backgrounds [21]. While this approach may be useful to increase the usability of ICTs in specific circumstances, we should recognize that its emphasis in testing, measuring, typifying, quantifying, and cataloguing culture and its traits [1] seems to render it unsuitable to

10

answer larger questions like what is the process whereby a technology is appropriated, what are the circumstances leading to the modification of a technology, or what is the role of a piece of technology in people’s life. These types of questions are precisely the type of questions we are interested in exploring. Thus, this approach seems to be unsuitable for our endeavour. 3.1.2 The Cultural Dimensions Approach

Three might be the reasons why this approach, usually in combination with the previous one (see for instance [13, 22, 24]), is one of the most popular in culture and HCI. First, it provides a reductionist and ready-made model of culture; second, being, a model, it is based in so-called “quantifiable” traits of culture (power distance, collectivism vs. individualism, uncertainty avoidance, masculinity vs. feminity, and long term orientation); and finally, this model was cited in one of the earliest books dealing with culture and HCI, “International User Interfaces” edited by del Galdo and Nielsen. Many are the criticisms of this approach including a claim for cultural homogeneity among the members of a culture; disregard for behaviour and values that are not necessarily determined by a particular cultural setting; variations in the expression of cultural traits because of changing situations; and finally, differences in the importance assigned by different cultures to a particular cultural trait [21]. An extended analysis of the shortcomings of cultural dimensions can be found in [25]. Furthermore, the static nature of cultural dimensions seems to prevent it from providing an intelligible account of, for instance, the fast-changing pace of computermediated communication. Such a circumstance can be observed in a study analysing McDonalds websites around the world [34] in 2003 with their counterparts in July 2007. Clear-cut issues of low- vs. high-contextuality [18] and invidualism vs. collectivism, like those found in Würtz study [34], are difficult to distinguish in the current design of those websites. As seen, it would be erroneous to try to apply the cultural dimensions model to understanding social interactions mediated by computing technologies when the model itself seems to be flawed and, therefore, lacks the ecological validity necessary to explain interactions unbounded to a desktop, the province of Ubicomp. This

11

approach seems to fall short at satisfying the conditions set forward above (see page 8) of a definition of culture that might help us advance our understanding on this matter. 3.1.3 The Semiotic Approach

Two methodologies have been developed within the semiotic approach to culture in HCI, these are known as Meaning in Mediated Action (MIMA) [9] and Rapid Semiotic Modelling [26]. Both methodologies rely in the study of the acquisition of meanings within a group of people because of shared contexts. These methodologies seem to differ only in their particular emphasis to context in the case of MIMA and to language in the case of Rapid Semiotic Modelling [21]. The semiotic approach suggests the possibility of particular meanings assigned to particular circumstances regardless of the multiplicity of individuals interacting in multiple contexts. The semiotic approach, as described here, departs from rigid classifications of culture bounded to a geographical area and deemphasizes reliance in usability metrics as the sole predictor of cultural-appropriateness. The semiotic approach seems to be useful in our endeavour to understand the process whereby a computing technology becomes ingrained in people’s lives and thus, it might be useful in our future analysis of the meanings associated to particular technologies in particular contexts. 3.1.4 The Activity Theory Approach

Activity Theory seeks to understand the surrounding conditions where an activity takes place, thus, it aims at understanding relationships between subjects, objects (or goals), artefacts, actions, and operations. Activity theory emphasises the unconscious nature of operations that, when executed in a certain sequence depending on the contextual conditions, lead to the performance of an action or a conscious act. Given the influence of the context in the sequence of operations need to perform an action and, subsequently achieve a goal, the activity theory approach is open to consider the role of the different aspects of context including the physical, the social and the cultural [1]. The weakness of activity theory seems to be its descriptive nature, that is, activity theory is not a framework or a predictive theory that might be used to understand

12

“culture, values, motivation, emotions or human personality” (Kaptelinin (1996) cited in [21]). The subject-artefact-object model of activity theory, however, opens up the space for a first approach to try to understand what sort of relationships can be established between those elements, and between those elements and a changing socio-cultural context. This model could be valuable in our ongoing research as we are also interested in the influence of a changing context in people’s use, adoption and appropriation of ICTs. 3.1.5 The Ethnographic Approach

The ethnographic approach is one of the newest additions to the arsenal of HCI in dealing with culture. Its popularity within the HCI community seems to be on the rise. This fact might be due to the continuous push by the community to move the computer away from the desktop as envisioned by Ubicomp. The ethnographic approach seems to be especially useful when dealing with questions transcending a limited physical space and individual users. Ethnography provides a description and an interpretation of phenomena observed in living communities; through its field studies, that is, by reason of being there, ethnography tries to provide an account of actions and of the influences of context upon those actions [1]. The work of Bell [3-7] seems to be one of the most representatives of this approach in HCI. Thanks to generous resources from Intel, Bell has been able to conduct ethnographic work around Europe and Southeast Asia to understand how home technologies acquire their significance in everyday living outside the North American milieu. Bell’s work seems to be useful in a number of ways. First, it illustrates the use of computing technologies under different constraints (e.g., power supply, network availability, social customs, etc) [4]. Second, it illustrates the appropriation of computing technologies in a manner different to that originally intended. For instance, through what Bell calls the “secret life” of ICTs Bell shows how these technologies are appropriated to satisfy other areas of people’s lives like love, sexuality, and the expression of spirituality [3]. Finally, it highlights the mismatch between the knowledge gained by the HCI community in studies of work settings and its applicability beyond the desktop – a circumstance especially true of the “home” where Bell’s work has been focused [5, 6] and naturally relevant to our aim of understanding the process whereby ICTs become part of the landscape of everyday

13

life. Bell’s ethnographic research has shown that the vision of Ubicomp might be already happening both far from the North American influence and in a manner different to that envisioned by Weiser himself [7]. We must indicate however, that we disagree with Bell and Dourish’s view of Ubicomp since it does not comply with one of the required conditions to achieve an Ubicomp environment (see page 5). Bell and Dourish’s work, however, supports the underlying thesis of our research, that is, that because of different socio-cultural circumstances (i.e., needs, belief systems, social customs, infrastructures, etc.) the form that Ubicomp takes is relative to a given sociocultural milieu and, thus, instead of a single, “tidy” and uniform vision of Ubicomp, we should expect to see a multiplicity of Ubicomp environments. The HCI community also seems to favour the same methods in researching the role of the socio-cultural context in the use of ICTs. These methods include questionnaires, formal experiments, interviews, contextual methods, and the above mentioned cultural models introduced by Hofstede. These methods have been used to research several areas of concern within HCI including general design principles, the Web, groupware, embedded systems, and interface agents. While we should recognize that the field of Ubicomp has, at this point, its own venues of publication (which were not covered in this survey by Kamppuri et al.), it is interesting to note the lack of research addressing specifically the interplay between the socio-cultural context and Ubicomp. This circumstance is, however, beneficial for us as it will give us some latitude to experiment with several research methods or interventions. Beyond identifying some shortcomings in current research regarding the role of culture within HCI, Kamppuri et al.’s survey [20] highlights the fact that most of the research being conducted in this area is comparative in nature, that is, one culture will be used as the bar against which other groups will be measured. A comparative study necessarily places the researcher, usually an individual from a different country/culture, in the observer’s perspective precluding him from being able to understand (interpret) how others understand themselves but through the lenses of his own cultural perspective and his own limited knowledge of the culture surveyed. It is precisely this circumstance we would like to avoid in our own research. As indicated, we are interested in understanding how our proposed view of culture (see page 8) may give rise to what we call multiple Ubicomp environments (see page 5). To achieve

14

this, and to bypass the bias of an external observer, we believe our research should: a) study current and ongoing social practices – the kind that concerns us here being those social practices mediated by computing technologies in people from different countries; and b) approach those phenomena to make them “accountable” from the perspective of those under study. We believe both conditions can be satisfied using an ethnographic approach in our research.

4 Current Work In accordance to our views of culture and Ubicomp, and inspired by the work of Bell, we designed an ethnographic study consisting of four interventions aimed at understanding the process whereby the socio-cultural context influences the use and appropriation of ICTs. We believe the outcome of this study will advance our understanding of the influence of the socio-cultural context in shaping multiple Ubicomp environments; a circumstance we call the Socio-cultural Construction of Ubiquitous Computing. Several research questions have guided our approach. What is the process whereby a computing technology becomes part of the “landscape of everyday life”? What factors have a role in this process of appropriation? Are these factors common across different countries or groups? We believe a greater understanding of these issues might lead us to characterize different Ubicomp environments that are better aligned to the circumstances of different groups and cultural milieus. Overall, we aim at informing the process to achieve different Ubicomp environments from the perspective of the second condition in our definition of Ubicomp (see page 8). 4.1 Intervention 1 – Semi-structured interviews

Our first intervention consisted of in-depth, semi-structured interviews with fifteen male and female master students from fields other than computing science at the University of Glasgow. Our sample included five subjects each from China, India and Greece between the ages of 22 to 28 years old who have been in the United Kingdom for no more than six months at the time of the interview. We chose this method of ethnographic research because it would allow us to produce a rich narrative around

15

the reasons why a subject would exhibit a particular behaviour towards ICTs. Having a rather retrospective character, the interview sought to understand participants’ previous experience with ICTs, and how this behaviour might have been altered since their arrival to the UK. Although recollections are of themselves inaccurate, this method would, notwithstanding, allow us to have a better understanding of what a person considers valuable by revealing lasting impressions of their contact with ICTs [17]. Thus, questions ranged from landmarks such as the time and circumstances of first contact with computers and the Internet, to uses of Internet, IM, mobile phones, and online social networks while in their home countries, to current uses of these technologies while in the UK. We have chosen a sample of international students for the following reasons. Strictly speaking, a proper ethnographic research demands the researcher’s immersion in the community that is to be studied for a reasonable length of time, usually from 6 months to one year or more [17], thus, the presence of a large pool of international students, like this writer, at the University of Glasgow makes feasible the exploration of some cultural traits relevant to our research. Several other reasons account for our choosing of this sample. Firstly, master students from China, India and Greece represent the largest international student communities at the University of Glasgow, thus, a large pool of subjects was guaranteed. Secondly, the requisite of having been in the UK for no more than six months would secure both enough exposure to current British conditions of Internet connectivity and some degree of familiarity with the offerings of national mobile network providers. At the same time, the relative short length of time living in the UK would likely ensure that our subjects have not forgotten the prevalent conditions of Internet connectivity and access to mobile networks back in their home countries. Thirdly, we assumed the requisite of participants living in student accommodation would “normalize” previous dissimilar socio-cultural contexts securing similar living arrangements (e.g., individual bedrooms, a personal laptop computer, and permanent connection to the Internet at home). Conversely, we also assumed that our participants would be aware of these differences between their current living arrangements and their previous conditions back in their home countries. Finally, while we recognise that ICTs (e.g., Internet, IM, mobile phones, and online social networks) do not satisfy the first condition of our definition of a Ubicomp environment (see page 5), these technologies seem to have reached a level

16

of “pervasiveness” in the lives of our sample. For our subjects, ICTs seem to be so natural that they have indeed weaved into the fabric of their everyday lives [7]. As in the case of the “home” for Bell [4, 5], we believe ICTs are a good vector of research to compare and contrast practices of use and appropriation across different countries. 4.1.1 Intervention 1 - Limitations

We recognise the following limitations in our approach. First, the classic sample of university students may provide information that, while culturally-diverse, will be restricted to a very particular segment of the population from each country surveyed. With the exception of those holding a scholarship, just a particular segment of the population from the countries surveyed would be able to afford studying in the UK. Second, conducting interviews in a second language like English seems to prevent participants from fully expressing their inner thoughts even when questioning revolves around such trivial matters as the use of mobile phones. Third, our limited exposure to the countries surveyed most likely indicates that valuable information might be “lost in translation,” that is, lost because of the lack of a common sociocultural background that might be used to interpret past and present experience with ICTs – however, the emic-etic tension of any ethnographic project is a constant element of this approach. Finally, while the findings of such a small sample might be reliable, we are in no position to generalise them to an entire country, or even an entire group (for instance, all the master students from a specific country). Despite these disadvantages, we trust this intervention will shed some light on the issues under research. We intend to address the weaknesses of this first study in our following interventions describe below. 4.1.2 Intervention 1 – Ongoing analysis

We are analysing the transcripts of the interviews using the Grounded Theory approach. This iterative method seeks to identify themes across the data gathered which, after a process of comparison, might be used to formulate theoretical models that can be supported by the very data from where they arose [8].

17

At the time of writing we have not finished analysing the data gathered from our interviews using the grounded theory approach. However, we believe to have identified at least three emergent themes: Computer-Mediated Social Interaction While questioning on the use of online social networks we found varied responses consistent across the countries surveyed. These range from a complete aversion to the use of these technologies because of privacy concerns and unnecessary exposure to unknown people, to indifference, to a steady use. Unlike our Greek participants, Chinese and Indians seem to take online social networks as the means to remain involved with people that, though geographically scattered, actively participate and indirectly promote a continuous exchange of tokens of friendship (e.g., blog comments). For some, we are aware, this positive attitude to online social networks could be interpreted as the expression of a “collectivistic identity” translated to the digital realm. We argue, nothwithstanding, that this positive attitude to social interaction mediated by computing technologies is rather the natural integration of ICTs into a lifestyle. This circumstance is certainly the case with our Chinese and Indian participants whose first contact with ICTs took place several years earlier than the Greek participants who, in most cases, are unaware of the existence of social networks. Parental Involvement This second theme seems to be a common factor underpinning early adoption of ICTs across the countries surveyed. Parents seem to play an active role in introducing, providing, encouraging, and facilitating their children’s access to ICTs, for instance, in the provision of a personal computer and Internet access, for considering ICTs as an advantage to their educational opportunities; or as an element to foster their well being, for instance, in the provision of a mobile phone when children reach the age when they are expected to spend longer periods of time outside the family home. Again, this circumstance seems to be characteristic of the Chinese and Indian participants, but not of the Greeks. Universal Adaptation

18

We believe this theme has to do more with the sample chosen than with any other circumstance. It seems to be that despite the cultural diversity of our sample, it is still too homogeneous. The prevalent conditions of university living, at least at the University of Glasgow, seem to force all international students alike, no matter what their previous experience with ICTs might be, to adapt, in a seemingly homogeneous manner, to a different technological milieu to succeed in their studies. This situation could be interpreted as an instance of the endless ability of humans to constantly renegotiate meanings – ICTs infrastructures in this case – because of changing circumstances (see the Semiotic Approach on page 12). This circumstance might be a welcome finding to the HCI community, since it seems to imply that the sociocultural context where initial contact with ICTs took place has no definitive, unchanging influence in future use, adoption and appropriation of computing technologies. 4.2 Intervention 2 - Participant Observation Online

We will conduct an intervention that can be called an “participant observation online” of a popular online social network, Facebook. We will conduct this intervention in two stages. First, we will study the use of this system among international students who were at the University of Glasgow during the school session 2006-7, but who are now elsewhere. To set the stage for the first part of this intervention we have formed in Facebook, through a third-party entity known as “Isoc Glasgow” (International Society at the University of Glasgow), a group known as the “International Society 2006-07 - University of Glasgow” ( http://glasgowuni.facebook.com/groups.php?ref=sb&pwstdfy=b29f0608295d1b24205 52a8836e71c15). Formed in early June, this group currently has 280 members. The main purpose of the existence of this group is to be able to identify and classify members of that generation of students. The second stage of this study is similar. In late August, through the entity known as “Isoc Glasgow” we will set up a second group in Facebook called “International Society 2007-08 - University of Glasgow.” Once this second group is functional we will invite future members of the International Society at the University of Glasgow during the coming year to join Facebook and make use of this online resource to keep in touch with friends and acquaintances here in Glasgow.

19

4.2.1 Intervention 2 - Justification

We have chosen to study the use of this particular online social network not only because of its current appeal in the general student community, but because its use would represent the adoption of a novel technology for those not familiar with it, and because, having recently opened its API, Facebook could represent, from a certain perspective, a system that can be adapted, personalised and appropriated. Although we are aware that Facebook, in its current state, cannot be classified as an Ubicomp environment, we think is a better option in exploring issues that will be indicated below because at least it satisfies one of the conditions we set forward to achieve an Ubicomp environment. We feel that a technology in its prototype stage falls short in fulfilling this requirement for the following reasons: •

It presupposes use.

•

It presupposes and is restricted to a specific number of users.

•

It presupposes scenarios of use that might not come to pass.

•

It has a limited life span, usually that of the trial.

•

It lacks peer-support to encourage continuous use.

•

It has no real expectations from the user’s point of view, that is, the user is unsure of what the device does, how it does it, and what might be its value in a larger context.

4.2.2 Intervention 2 - Methodology

We intend to explore in more detail several issues identified in our first intervention. For instance, we will compare and contrast the use of Facebook among people who are geographically dispersed (those in the first group), and among people who are collocated in the same city (those in the second group). We will explore the influence of a changing environment (from their native countries to the UK) in the adoption of a novel technology (in the sense that it might not have been used before); the process whereby a novel technology is appropriated, or the reasons not to; and finally, long

20

term use of a technology, since we intend to conduct this study for at least four months beginning in September 2007. We will use a variety of ethnographic methods in this intervention including again semi-structured interviews to be analysed via grounded theory with both collocated and distant participants; and questionnaires, taking advantage of the large number of people that can be reached online through Facebook. 4.2.3 Intervention 2 - Outcome

As indicated, this intervention will allow us to further explore some themes already identified in our first intervention; themes that we feel were not properly addressed since the method used, semi-structured interviews, was unsuited to address an holistic view of everyday interaction with different computing technologies. We will also use this intervention to continue exploring issues of culture; we will return to analyse the role of the socio-cultural context in fostering or hindering adoption of a new technology, and in shaping the way those novel technologies might be used. We believe these issues will advance our knowledge of the conditions whereby Ubicomp environments might be achieved because of various socio-cultural contexts. 4.3 Intervention 3 – Use of GPS devices

Someone mentioned this writer that GPS units currently being used by drivers could be considered as true Ubicomp technologies. Upon further consideration of this piece of technology, it seems to be that this is in fact a technology that satisfies, at least to a certain degree, both conditions necessary to attain an Ubicomp environment at least according to our definition. First, by constantly and automatically modifying the information being displayed, GPS units seem to be able to adapt to a changing environment because of a changing geography. Second, for a certain group of people, not necessarily collocated in a geographical space, GPS devices seem to have attained a level of pervasiveness by reason of being integrated with a previously held practice, driving. Unlike our previous interventions, we will focus both in a single, straightforward technology such as GPS devices, and in a sample of users very different than those surveyed before. With this, we will strive to overcome weaknesses of our previous

21

approaches, including the seemingly restricted validity of our findings beyond a very particular profile of users like international students. Additionally, in this intervention we expect to find greater variations in age, previous experience with technology, different degrees of integration of a novel technology with an existing practice. Like in our previous interventions, we will continue studying the use of technologies already in the wild, without the shortcomings identified above of a prototype technology. 4.3.1 Intervention 3 - Methodology

This intervention will be conducted using participant observation to explore the use of GPS in a sample, ideally, of taxi drivers in Glasgow for two months. Unlike our previous interventions, we aim at conducting an actual field study documenting current practices among this particular community. We expect to use the standard resources of this methodology including note-taking, photography, recording and informal interviewing to identify areas of relevance for our own research. We hope to be able to find users at different stages on the process of appropriation, including those who reject such devices. 4.3.2 Intervention 3 - Outcome

With these initial observations we will design a semi-structured interview exploring in more detail those areas that might contribute to our understanding of the conditions whereby an Ubicomp environment might be achieved for a particular sample of users in a specific activity. As with the previous interventions, we expect to be able to draw some recommendations for the design of technologies unbounded to a desktop.

5 Conclusions In this report we have presented our ongoing research around what we have called the Socio-cultural Construction of Ubiquitous Computing, that is, the possibility of a multitude of Ubicomp environments because of different socio-cultural contexts. We described our ongoing exploration of this issue, the limitations in our first intervention, and some tentative themes observed. We have also described our work

22

plan for the coming year in the form of two more interventions that will contribute to enrich our research views. Overall, we expect to contribute to a better understanding of the nature of what we have called the conditions to achieve Ubicomp environments by studying the role of the socio-cultural context in shaping the use, adoption and appropriation of ICTs.

6 References 1.

2.

3.

4.

5.

6.

7. 8. 9.

10. 11. 12.

13.

Abdelnour Nocera, J.L., Culture in Human-Computer Interaction Studies. in Proceedings Cultural Attitudes Towards Communication and Technology 2002, (Université de Montréal, Canada, 2002), 505-524. Abdelnour Nocera, J.L., Global Software, Local Voices. in Proceedings Cultural Attitudes Towards Communication and Technology 2004, (Murdoch University, Australia, 2004), 29-42. Bell, G., No More SMS from Jesus: Ubicomp, Religion and Techno-spiritual Practices. in 8th International Conference, UbiComp 2006, (Orange County, CA, USA, 2006), Springer Berlin/Heidelberg, 141-158. Bell, G., Other Homes: Alternate Views of Culturally Situated Technologies for the Home. in Conference on Human Factors in Computing Systems, (Ft. Lauderdale, Florida, USA, 2003), ACM Press. Bell, G., Blythe, M., Gaver, B., Sengers, P. and Wright, P., Designing Culturally Situated Technologies for the Home. in Conference on Human Factors in Computing Systems, (Ft. Lauderdale, Florida, USA, 2003), ACM Press, 1062-1063. Bell, G., Blythe, M. and Sengers, P. Making by Making Strange: Defamiliarization and the Design of Domestic Technologies. Transactions on Computer-Human Interaction, 12 (2). 149-173. Bell, G. and Dourish, P. Yesterday's Tomorrows: Notes on Ubiquitous Computing's Dominant Vision. Personal and Ubiquitous Computing, In Press. Bernard, H.R. Research Methods in Anthropology: Qualitative and Quantitative Approaches. AltaMira Press, 2006. Bourges-Waldegg, P. and Scrivener, S.A.R. Applying and testing an approach to design for culturally diverse user groups. Interacting with Computers, 13 (2). Bourges-Waldegg, P. and Scrivener, S.A.R. Meaning, the Central Issue in Cross-cultural HCI Design. Interacting with Computers, 9 (3). 287-309. Chalmers, M. A Historical View of Context. Computer Supported Cooperative Work (CSCW), 13 (3). 223-247. Chalmers, M. and Galani, A., Seamful Interweaving: Heterogeneity In The Theory and Design of Interactive Systems. in Proceedings of the 2004 conference on Designing interactive systems: processes, practices, methods, and techniques, (Cambridge, MA, USA, 2004), ACM Press, 243-252. Choi, B., Lee, I., Kim, J. and Jeon, Y., A qualitative cross-national study of cultural influences on mobile data service design. in Proceedings of the

23

14. 15. 16. 17. 18. 19.

20.

21.

22.

23. 24. 25. 26. 27.

28.

29. 30. 31.

32. 33.

SIGCHI conference on Human factors in computing systems, (Portland, Oregon, USA, 2005), ACM Press, 661-670. de Mooij, M.K. Global marketing and advertising: understanding cultural paradoxes. Sage Publications, Thousand Oaks, Calif., 1998. Dourish, P. What We Talk About When We Talk About Context. Personal and Ubiquitous Computing, 8 (1). 19-30. Edwards, P.N. The Closed World: Computers and the Politics of Discourse in Cold War America. MIT Press, Cambridge, Mass., 1996. Fetterman, D.M. Ethnography: Step-by-Step. Sage Publications, Inc., 1998. Hall, E.T. The Silent Language. Doubleday, Garden City, N.Y., 1959. Hofstede, G.H. Culture's Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations. Sage Publications, Thousand Oaks, Calif., 2001. Kamppuri, M., Bednarik, R. and Tukiainen, M., The Expanding Focus of HCI: Case Culture. in Proceedings of the 4th Nordic conference on Humancomputer interaction: changing roles, (Oslo, Norway, 2006), ACM Press, 405-408. Kamppuri, M. and Tukiainen, M., Culture in Human-Computer Interaction Studies. in Proceedings Cultural Attitudes Towards Communication and Technology 2004, (Murdoch University, Australia, 2004), 43-57. Kayan, S., Fussell, S.R. and Setlock, L.D., Cultural Differences in the Use of Instant Messaging in Asia and North America. in Proceedings of the 2006 20th Anniversary Conference on Computer Supported Cooperative Work, (Banff, Alberta, Canada, 2006), ACM Press, 525-528. Lyytinen, K. and Yoo, Y. Introduction. Issues and Challenges in Ubiquitous Computing. Communications of the ACM, 45 (12). 62-65. Marcus, A. and Gould, E.W. Crosscurrents: Cultural Dimensions and Global Web User-interface Design interactions, 2000, 32-46. Ratner, C. and Hui, L. Theoretical and Methodological Problems in CrossCultural Psychology. Journal for the Theory of Social Behavior, 33. 67-94. Sacher, H. and Margolis, M. Business: The Culture of Interaction: About Foreign and Not-so-foreign Languages. interactions, 7 (1). 39-45. Suchman, L.A. Plans and Situated Actions: The Problem of Human-Machine Communication. Cambridge University Press, Cambridge [Cambridgeshire] ; New York, 1987. Tolmie, P., Pycock, J., Diggins, T., MacLean, A. and Karsenty, A., Unremarkable Computing. in Proceedings of the SIGCHI conference on Human factors in computing systems: Changing our world, changing ourselves, (Minneapolis, Minnesota, USA, 2002), ACM Press, 399-406. Weiser, M. The Computer for the Twenty-First Century Scientific American, 1991, 94-104. Weiser, M. Some Computer Science Issues in Ubiquitous Computing. Communications of the ACM, 36 (7). 75-84. Weiser, M. and Brown, J.S. The Coming Age of Calm Technology. in Denning, P.J. and Metcalfe, R.M. eds. Beyond Calculation: The Next Fifty Years of Computing, Springer-Verlag, New York, 1997. Weiser, M., Gold, R. and Brown, J.S. The Origins of Ubiquitous Computing Research at PARC in the late 1980s. IBM Systems Journal, 38 (4). 693-696. Wikipedia, Culture, viewed 26 July 2007, .

24

34.

Würtz, E., Intercultural Communication on Websites: An Analysis of Visual Communication in High- and Low-context Cultures. in Proceedings Cultural Attitudes Towards Communication and Technology 2004, (Murdoch University, Australia, 2004), 109-122.

25