Keywords

1 Design Contribution in the Context of Data Literacy

Similar to the history of other literacy initiatives, data literacy requires behavioural change [4]. The invention of the printing press created the need for universal text literacy; likewise, the need to manipulate considerable amounts of large numbers created the need for mathematical literacy; the ubiquity and relevance of photography, film and digital drawing tools posed the need for visual literacy [7]. Similarly, with the increasing availability of large data sets, it is possible to draw a parallel with the compelling need for universal data literacy that goes from the understanding of data to the use of visualisations and other enabling systems for the transformation of information into knowledge. Like other literacies, data literacy aims to promote better communication and collaboration, empower users to understand their world, establish individual self-efficacy, and improve decision-making in many contexts. Data literacy also allows laymen not to be reduced to the mere role of passive data producers. Corrall [12] shows how the definition of data literacy constitutes a “wicked problem”: the author combines the multiple definitions of data literacy formulated by scholars to demonstrate the multifaceted and interconnected nature of this definition problem. It may be helpful to note that Buchanan [9] has previously framed the nature of design in a similar manner. In the same vein, Norman [26], echoing the “wicked problem”, refers to a social or cultural problem that is difficult or impossible to solve due to its complex socio-technical nature. In the variety of interdisciplinary contributions that experts can make to help in the development of tools and processes to support this new knowledge, design is understood in its “sense-maker” characteristic [22, 25], its multifaceted naivety and its multiverse applications [3, 15]. As such, design can play a crucial role in facilitating the process of data literacy as a qualitative driver of knowledge communication, where the focus lies not in the final communicative artefact – technical or aesthetic – but in the underlying process of transforming input data into a communication product or experience. On the one hand, design can help solve the formalisation and visual expression of masses and data flows. On the other, it can operate in order to detect new directions and scenarios for the use of data, exploring ways to translate their variable trends, and using a plurality of communication solutions on a macro and micro scale, supporting and facilitating innovations in terms of meaning and value production.

Beyond the traditional models of visual representation of data that come from statistics, which emphasise the “presentation of data”, design can give coherence and structure to the discourse originated by alphanumerical sequences – which are neither accessible nor immediate to understand. Bihanic states that “design provides real spaces of re-presentation” [5], thus facilitating a more sensitive interpretation of the dynamics of relationship between data, and providing valuable devices for detecting meaningful forms of relationship. By experimenting with emerging interactive technologies, materials and innovative processes, design produces a plurality of communication solutions and languages capable of supporting and facilitating innovations in meaning and value production. It is interesting to observe how design intervenes in favour of sense-making from data accumulation.

2 Expanding User Experience in Data Design Beyond Visualisation: The Lens of HCI

Design methodologies incorporate human-centred approaches, encompassing the dynamic and appropriate context involved in data processing, synthesis and communication. If design from the accumulation of data supports sense-making processes, the humanistic focus of the intervention facilitates the transmission of the qualitative value of data [11]. Such an approach promotes data literacy, which Bhargava understands as “the desire and ability to engage constructively in society through and with data” [4]. As already noted, design methodologies and tools contribute to the configuration and arrangement of a space appropriate to host different experience forms and relationships between the user and data.

At the root of understanding data literacy and designing for inclusion is an urgent need to rethink approaches for the design, creation, and support of data-driven systems that are more human-centred and based on inclusion, empathy, and responsiveness. Contextual, human-centred approaches are arguably critical – and often absent – elements in the design and development of data-related activities [4].

The reference to the experiential dimension to which interactive artefacts give rise represents a focus of the most recent philosophical, scientific and medialogical discourse. In this case, the concept of experience is to be understood in the Deweyan meaning, that is, as that which allows for factual and reciprocal interaction with the environment, oriented toward a sense of responsibility and participation toward solving society’s problems. Such attention resulted in the “experiential turn”, which has led to a progressive shift towards the centrality of the user’s experience, in lieu of supports, techniques, and technologies. There has been a shift in the design field from techno-centric approaches to design to developing methodologies and viewpoints centred on the user experience. In the field of user-experience theory, Hassenzahl [18] proposes research methods that lean towards an approach called “experience design”. These methods can identify individuals’ and groups’ complex and nuanced needs, challenges, and aspirations within a data ecosystem. Discovery and learning, related to experience, are central to human-centred approaches. Empathy is one of the necessary tools for understanding the experiences of others. Furthermore, it is important to take into account the increased attention given to the deepening of the notion of immersiveness. The primary interaction strategy for the use of data is measured by the possibilities offered by the most recently developed platforms, that are not limited to visual representations but also include mobile devices and platforms for the experience of “immersive environments.” This has been fostered in part by the possibilities granted by technological development and the advantages of more excellent distribution and greater accessibility of systems and devices to enjoy immersive environments. In these cases, physical interaction emerges from the two-dimensional mise-en-scène of the monitor and pervades, overlapping like a layer of knowledge, the actual space. This brings new challenges to the idea of witnessing, which can be juxtaposed with the ideas of presence and personalisation. Such reassessment could be interpreted in fact as a natural broadening of the field by the data life-cycle concerning the translation stage. With mindful attention to the explicit, implicit, and unconscious needs of different individuals and groups, appropriate activities, tools, supports, and communications for data and data-informed actions can be designed and supported [4].

With technological innovations, new communication languages for HCI have been designed, developed and spread. Throughout history, as a result of this process, different types of interactive experiences have emerged. This led, to a greater or lesser extent, to innovations in meaning and behavioural changes. Since the objects of data design are the user experience and the environmental dimension in which it takes place, and considering the above considerations, it seems appropriate to explore the topic of human-data interaction through the lens of HCI. The evolution of HCI has been traced by considering the technological innovation in terms of hardware and software, the development of the language that made HCI possible and, simultaneously, the innovations – in terms of meaning and behaviour – that have been introduced [2]. The idea is to use the same criteria to analyse a selection of case studies in which different types of Human-Data Interaction can be identified and, thus, different experiences.

3 Framework Analysis

Data visualisation is traditionally regarded as a tool for exploring data and making conjectures. Historically, its roots lie in the domain of scientific disciplines, where they are created by and for the experts. Data visualisation thus represents the result of an analytical process. Taking “data as raw material” [26] implies knowing its specific properties and uses. The specific characteristics underlying data design can be used to support the production of different forms of value.

The following framework of analysis proposes a reading of the state of the art of Human-Data Interaction through four dimensions: what technological innovations enable the development of new data communication languages? What technology innovations in hardware and software allow us to interact with data in a different way? Which spatial scale of experience does this offer? What value, what innovations of meaning does it bring? The suggested framework attempts to “establish alternative cultural decisions as engines of social transformation through design” [15]. It aims to empower the data design process by prefiguring a user experience that appeals to the data illiterate. The data illiterate is an individual who understands, explains and documents the usefulness and limitations of data by becoming a critical consumer of data, controlling their data journey, finding meaning in data and acting upon it. The data-literate individual can identify, collect, evaluate, analyse, interpret, present and protect data. Similar considerations and analyses have already been made from the visual interaction required in the case of data visualisation [14, 33]. However, several projects demonstrate that the possibilities offered can expand the interaction with data beyond mere visualisation.

Data are considered immanent presences, ductile, malleable and endowed with significant plasticity. Such concrete data design practices take on visual or physical form and static or dynamic behaviour to the aggregations, fluctuations and circulation. In this sense, data can be considered a raw material for designing experiences. In the case studies considered below, it is possible to observe how data design intervenes by organising physical or digital spaces capable of embodying immaterial data.

4 Data as an Experiential Interface to Innovations in Meaning and Change of Behaviour

Based on the above premises, a selection of case studies proves how designers are expanding the significant scope of data visualisation [34] and developing targeted design interventions.

Stating “Space junk has increased in recent decades and collisions could increase if the problem is not kept under control”, the Space Junk web app, designed by Federica Fragapane for the BCC [16], represents space debris in orbit, classified according to the type of space object and organised by its average distance from Earth. Each type of debris is also quantified by its mass in tonnes. Scholars have stumbled upon the concept of “data narrative” [23] as, through a humanistic approach, the intent to ‘give human life to data’ combines traditional codes of data visualisation with cognitive studies on perception. The result is a renewed visual syntax. In this case, human-data interaction brings about an epistemic value; the role of design facilitates the interpretative, critical and expository use derived from data in order to foster the production of new knowledge.

Similarly, in Plastic Air [24], Giorgia Lupi designs an interactive data experience produced in collaboration with the Google Arts and Culture project. The experience provides a lens through which it is possible to visualise and explore – on the web – the invisible plastic particles always present in the atmosphere surrounding us. It further considers how they impact the environment and our health. In this case, the value proposition is poietic, as data as raw material relies on a more exploratory and heuristic experiential logic.

Another linguistic development fostered by technological innovation has been called “data physicalisation” [20]. Data physicalisation can be experienced through physical and material forms, conveying information through unusual physical paradigms. Digital technologies have undoubtedly enhanced the possibilities of the visual representation of data. In contrast, humans have historically used physical representations of data – consider hourglasses, ancient notational systems, or mercury thermometers. The online archive Dataphys [13] collects examples of this type of visualisation and traces its history. There seems to be a resurgence of “data physicalisation” due to the use of technologies, new materials and processes. For instance, digital fabrication techniques such as 3D printing and digital milling may produce physical forms from databases. It is the case of Emoto by Moritz Stefaner [31]. Emoto is an installation designed during the 2012 London Olympics. It started with a web application that monitored public engagement via Twitter and then returned the results by creating an offline material data sculpture. The value proposed here is also poietic, in that using data as raw material is based on an exploratory and heuristic logic.

Environmental projects reveal data in the context of the natural environments around us, often exploring the use of natural processes. The following case introduces another valuable language: “data sensification” [19]. In this case, data acquire environmental dimensions in which the emphasis shifts to how it interacts with various inputs provided by users or sensors. The public site-specific installation Orbacles designed by TenxTen, Minn Lab [32], consist of three spherical environments that connect the citizens of Minneapolis to the phenomena of climate change through the behaviour of birdlife in the surrounding area and the language of the senses. The data design proves helpful as a device for documenting a natural phenomenon and speculating on the future. Orbacles facilitates the communication of a species’ decline and migration related to the effects of climate change. Each of the 147 bird species found in Minnesota can be accommodated because the covers are of a size proportional to the length and wingspan typical of the species – a nesting box, feeder or rainwater reservoir.

Another case of “data sensification” [19] is where data acquire environmental dimensions, in which the emphasis shifts to the interaction modes with the inputs provided by the users. Dustmark/Staubmarke [29] is an installation for the public space of Stuttgart, a city particularly affected by air pollution. The project displays air pollution by drawing attention to the patina on the surfaces of the city. The dust marks are made in reverse graffiti – a sensitive material – making the accumulated pollution visible by partially removing it. The process draws attention to dust as a concrete material rather than abstract numerical data. In the following months, the dust marks will vanish, as new dust will accumulate in the areas of the sign that were cleaned.

Applied data automation introduces physical platforms that use automation and robotics to encode data dynamically and interactively. The case of Surfacing Women in Smithsonian History [17] experimented with the development of tools for machine learning to explore the history and contributions of women in science. Automation applied to data promotes the production of new knowledge and thus epistemic value, but also praxeological value, as it encourages the discovery of new methods for organising museum archives. This represents just one of many cases of methods applied to archive storytelling [8]. We can say it is a participatory language that shows how designers invite users to create new data configurations, allowing them to encode or reveal data through their interactions with a piece, material or other people. In the same way, Surprise Machines [29] is a visual investigation of the archives of the Harvard Art Museum that takes the form of an installation. The project aims to organise the collections of the museum, with the goal of opening up previously unexplored sections of the multitude of objects that comprise them. This process makes use of algorithms capable of configuring visualisations from the public’s gestural input on the interface. The body of the visitor becomes a kind of “choreographic interface” [29] for interacting with the collection. This system allows visitors to move through the visualisation through total body interaction.

5 Humane-Centred Data Design?

Each of these approaches involve precise design solutions that affect the design and production processes, but also the tools and skills needed to create the connection between the designer, the user, and the data. One could argue that humane-centred data design is about more than just the individual, because it extends to group dynamics in the social context. Data designed based on their entire lifecycle could redeem the position of human beings from their position as passive producers of data, giving people the capacity to act within these data systems; the purpose of expanding data literacy is to be understood in this sense.