Dario Rodighiero | Publications

Archives, Museums, and Installations: Aldo Rossi’s Legacy in Digital Transition

2026-01-01T23:59:02+00:00

This article offers a theoretical reflection grounded in curatorial analysis and supported by a case study. It examines how Aldo Rossi’s legacy is redefined through archives, exhibitions, and installations, with a focus on the transformations of his Analogous City. Drawing on scholarly concepts from Paul Ricœur, Bruno Latour, and Umberto Eco, the article considers how memory, mediation, and interpretation work together as a process of meaning-making in cultural heritage. It shows how this process shapes the ways Rossi’s work is preserved and reimagined across different forms of display. Rather than treating cultural heritage as static, the article presents it as a process formed through editorial choices, curatorial practices, and participatory engagement. Rossi’s case study shows how architectural memory can remain open and responsive, especially when activated through digital and spatial forms of display.

Introduction

Laurajane Smith describes cultural heritage as a process shaped by remembering, storytelling, and meaning making. It unfolds when people engage with the past to understand the present, relying on memory to make sense of society. “The real sense of heritage […] is […] in the act of passing on and receiving memories and knowledge” (Smith 2006, 2). This view resonates with historian Pierre Nora’s observation that modern societies have lost the communal settings in which memory once circulated – such as villages, religious rituals, or oral traditions (Nora 1989). In response, communities create new symbolic sites of memory that attempt to preserve what can no longer be experienced directly, such as monuments and commemorative rituals. While Smith focuses on the cultural work of remembering, Nora emphasizes the rupture that makes such work necessary. Both underscore that heritage emerges from efforts to retain meaning when memory becomes unstable. From this shared perspective, institutions play a vital role in preserving and transmitting memory. They offer frameworks where fragmented histories can be collected, reinterpreted, and kept active. “The role of preserving symbolic sites of memory is formalized through international instruments, such as UNESCO’s Convention on the Protection of the World Cultural and Natural Heritage” (UNESCO 1972). These frameworks task states and cultural bodies with identifying, protecting, and promoting heritage as a living process – one that supports identity, diversity, and knowledge over time.

The process of heritage-making is increasingly shaped by digital technologies. In 2003, UNESCO introduced the concept of “digital heritage” in its Charter on the Preservation of Digital Heritage. The document, later released in a revised online version, defines digital heritage as “resources of information and creative expression” that are “produced, distributed, accessed and maintained in digital form” (UNESCO 2009). This marks a broader transformation in cultural institutions, where digitization affects not only access but also the ways heritage is produced and consumed. As Burdick et al. (2012) argue in Digital_Humanities, digital practices challenge more widely established disciplines and roles, calling for more reflective and design-driven approaches to knowledge. Cameron and Kenderdine (2007) note that digital technologies do not simply enhance cultural capital – they reshape the internal cultures and practices of the institutions that care for it. These changes are both technological and institutional, driving shifts within organizations such as museums, libraries, and archives. As Paul (2015) observes, technological development often outpaces the critical vocabulary needed to assess it, requiring institutions to invent new ways of speaking about the social, economic, and aesthetic implications of digital work.

The shift toward digital heritage intersects meaningfully with the legacy of architect Aldo Rossi (1931–1997). Born in Milan and primarily active in Italy, Rossi shaped how memory is understood within the built environment. After receiving the Pritzker Prize in 1990, his influence expanded internationally through projects such as the Quartier Schützenstraße in Berlin and the Bonnefanten Museum in Maastricht. Rossi did not view the city as a neutral setting for modern life but as a layered structure where buildings and forms retain the memory of past meanings (Rossi 1996). His notion of the “urban artifact” highlights how certain structures – by virtue of their form, history, or symbolism – anchor collective identity. As Peter Eisenman writes in his introduction to The Architecture of the City, Rossi’s ideas continue to inform how cultural memory is revisited and reinterpreted over time (Eisenman 1996). Although this article does not follow Rossi’s theoretical model, his legacy serves as a valuable entry point for examining how heritage is shaped today. The ways his work has been archived, displayed, and transformed into installations offer a concrete perspective on the changing role of institutions in the digital age.

Rossi’s legacy provides a useful lens for examining the digital transition in cultural heritage. This article approaches the topic through three interrelated practices: archiving, curating, and displaying. Archives are where heritage begins to take form, as curators and researchers select, organize, and preserve materials, turning them into collections that support memory and interpretation. Museums build on this foundation by creating narratives that connect these materials to broader historical and cultural contexts. Installations present these narratives in spatial form, shaping how audiences encounter heritage – often through hybrid formats that combine physical display with digital interaction. Taken together, these practices show how heritage is preserved, interpreted, and reimagined in the digital age. A clear example is the work of MAXXI in Rome, which houses Aldo Rossi’s archive (Zhara Buda et al. 2020) and has recently presented both a physical exhibition (MAXXI 2021) and a digital one (MAXXI 2022) dedicated to his work. These initiatives demonstrate how a cultural institution engages archival material through conservation, exhibition, and digital reinterpretation, making Rossi’s legacy accessible and relevant to contemporary audiences (Rodighiero 2022).

The article is structured around three main sections, each corresponding to a key site where Rossi’s legacy is shaped and reinterpreted: archives, exhibitions, and installations. These are interwoven with theoretical insights from Paul Ricœur (1977), Latour (2005), and Eco (1989), whose perspectives illuminate different moments in the process of meaning-making. Ricœur helps frame how memory is assembled, Latour clarifies how mediation transforms materials as they circulate, and Eco shows how interpretation keeps the work open to new readings. Together, they support the central argument of the article. Rather than following a purely chronological order, the article moves between empirical cases and conceptual reflections, culminating in a discussion of heritage as a dynamic and mediated process.

Memory, Mediation, and Interpretation

Understanding how heritage takes shape requires attention to how memory is formed, how meanings are mediated, and how interpretation unfolds. These three dimensions – memory, mediation, and interpretation – provide the conceptual framework for exploring Aldo Rossi’s digital legacy. 1) Memory is not given in full but constructed through acts of selection and omission, especially in the collection and preservation of traces from the past. Paul Ricœur (1977) describes memory as a process of reconstruction, shaped by the needs of the present and the expectations of the future. 2) Mediation is often treated as neutral, yet it actively transforms what it conveys. It links subjects, objects, and meanings, shaping how knowledge is structured and circulated. Latour (2005) reframes the social as a network of associations, where mediation works by assembling and translating these relations. 3) Interpretation extends beyond the curator’s choices. It continues through the viewer, who brings their own position to the encounter. Eco (1989) defines the artwork as an “open work,” inviting multiple readings rather than a fixed message. Together, these perspectives allow us to understand cultural heritage as a dynamic process – curated, mediated, and interpreted.

Memory plays a central role in the construction of heritage, yet it is always partial and shaped by the conditions of its transmission. For Paul Ricœur, memory is not a retrieval of the past but a reconfiguration shaped by present concerns. In Memory, History, Forgetting, he writes that “memory remains the guardian of the ultimate dialectic constitutive of the pastness of the past, namely, the relation between the ‘no longer,’ which marks its character of being elapsed, abolished, superseded, and the ‘having-been,’ which designates its original and, in this sense, indestructible character” (Ricœur 1977, 498). This tension is not only personal but institutional, embedded in archival practices that influence what is remembered and what is left out. Ricœur underscores that memory is vulnerable – not only to loss but to distortion and manipulation – making its ethical handling essential. Archives, in this view, are not neutral repositories but sites of reconstruction, where traces are selected to serve contemporary needs. This perspective aligns with the evolving role of heritage institutions, which no longer merely preserve objects but actively shape how memory is assembled. Remembering, then, is never neutral: it demands critical awareness of how forgetting works and how narratives emerge from incomplete materials.

Mediation is often seen as a background process, a technical means of transmitting content. Bruno Latour challenges this assumption by placing mediation at the core of social life. In Reassembling the Social, he writes: “Mediators transform, translate, distort, and modify the meaning or the elements they are supposed to carry” (Latour 2005, 39). Every act of mediation alters what is being transmitted; nothing remains unchanged. For Latour, the social is not a fixed domain but a process of assembling associations among human and nonhuman actors. This view reframes heritage institutions as active agents, shaping what becomes culturally meaningful. Cataloging, digitizing, and exhibiting are all forms of mediation that alter how materials are interpreted. This perspective shifts attention away from content alone and toward the networks through which it gains form and value. Mediation is not a neutral layer between origin and reception – it is the space where meaning takes shape. Recognizing its role allows us to follow how heritage is constructed and reframed in the digital environments where memory now circulates.

Interpretation plays a central role in how heritage is experienced and understood. As Umberto Eco writes in The Open Work, “the ‘openness’ and dynamism of an artistic work consist in factors which make it susceptible to a whole range of integrations […] structural vitality is still seen as a positive property of the work, even though it admits of all kinds of different conclusions and solutions for it” (Eco 1989, 20). This openness is not arbitrary; it is guided by the internal structure and form of the work itself. In the context of exhibitions and installations, this perspective is particularly relevant. Visitors interpret meaning rather than simply receive it, bringing their own knowledge, expectations, and context to the encounter. This participatory dimension is integral to how cultural artifacts acquire significance. Interpretation happens within frameworks that allow for variation while preserving coherence, shaping how meaning unfolds across audiences and platforms.

Together, Ricœur, Latour, and Eco offer a way to understand heritage as a living process shaped by how memory is handled, how meanings are mediated, and how interpretations unfold. These dimensions are closely connected: memory depends on forms of mediation, and each mediation creates space for interpretation. Applied to the legacy of Aldo Rossi, this framework highlights how his work is continually reshaped as it moves through archives, exhibitions, and installations. In the following sections, we explore these three settings to examine how Rossi’s legacy is assembled and reinterpreted across both physical and digital environments.

Exhibitions

Exhibitions have played an important role in shaping the public memory of Aldo Rossi at MAXXI. While archives preserve material traces, exhibitions bring them into view – translating drawings, models, and documents into curated narratives that engage different audiences. Since its founding, MAXXI Architettura has approached exhibitions as interpretive practices that connect historical material with contemporary reflection. Rossi’s work, both visual and conceptual, fits well within this approach. From early displays like Rossi. L’archivio personale (MAXXI 2004) to the major retrospective Aldo Rossi: L’architetto e le città (MAXXI 2021), MAXXI has returned to his work to explore new curatorial strategies, including digital formats. These exhibitions offer a way to mediate archival content and bring it into public conversation. They reflect the museum’s broader aim to preserve architectural memory while fostering new readings of its relevance. Rossi’s exhibitions, in this context, become spaces where interpretation takes shape and heritage is continually redefined.

Two major exhibitions at MAXXI have recently shaped the museum’s curatorial engagement with Aldo Rossi’s legacy. The first, Aldo Rossi: L’architetto e le città (MAXXI 2021), brought together more than 800 works from his archive – drawings, models, photographs, notebooks, and letters – offering a layered view of his career. Curated by Alberto Ferlenga, the exhibition was organized thematically and chronologically, guiding visitors through key phases of Rossi’s work while allowing space for visual associations and digressions. Alongside these archival materials, the curators also included Rossi’s own words, presented in a dedicated section of the display: these fragments of text underscored how the archive brings to light ideas and reflections, foregrounding the intellectual and poetic dimensions of his work. The display echoed Rossi’s own poetics, presenting cities as accumulations of memory and form. The exhibition design emphasized continuity across time, blending architectural imagination with personal traces. Rather than presenting Rossi’s work as a static legacy, the exhibition invited viewers into an active encounter, situating his archive within a living context of interpretation.

The second initiative, Aldo Rossi Digital (MAXXI 2022), reimagined this work as an interactive online exhibition. Developed by the curatorial team that worked on both the exhibition and the inventory, with the creative agency Dinamica Studio – and then landed on MAXXI’s Know-How – the platform reorganized Rossi’s archive into thematic clusters, such as The Analogous City and Typologies, and enabled non-linear navigation across various media. Instead of replicating the structure of the physical show, the digital version made use of the web’s affordances: fluid associations, multiple pathways, and open-ended exploration. Visitors could follow their own trajectories through sketches, photographs, and texts, engaging directly with Rossi’s imagery and thought. This format emphasized accessibility and participation, opening the archive to a broader audience beyond the museum. It demonstrates how digital exhibitions can support reinterpretation by enabling multiple modes of engagement and reading.

For Bruno Latour, mediation is never neutral – it transforms what it transmits. As objects and knowledge circulate through networks, they are translated, redirected, and reshaped (Latour 2005). In Rossi’s exhibitions, mediation takes place on multiple levels. The archive is not simply moved into the gallery or online; it is reassembled through curatorial choices, exhibition design, and digital interface. Each step affects how Rossi’s work is perceived. These exhibitions act as mediating spaces, producing new associations and inviting different forms of engagement. For example, in Aldo Rossi: L’architetto e le città (MAXXI 2021), the exhibition layout turned the archive into a spatial narrative that invited active exploration. The digital version extended this further, highlighting mediation through tools like filters, thematic tags, and open navigation. In Latour’s terms, these are not secondary processes – they are central to how heritage is shaped and circulated. Rossi’s exhibitions, then, are not passive displays but sites where his legacy is continuously adapted and reinterpreted.

Taken together, these exhibitions show that Aldo Rossi’s legacy is not shaped by simply presenting archival content to the public but by the interaction of curatorial direction, exhibition design, and institutional context. The physical exhibition at MAXXI used spatial sequencing and thematic groupings to guide engagement, while the digital platform relied on modular navigation and visual links. Both created settings where heritage is experienced as an evolving process of interpretation. This combined strategy reflects changing patterns of audience engagement, offering continuity across media while adapting to different formats and rhythms. It also illustrates the role of institutions like MAXXI in shaping how memory is made accessible and meaningful. Rather than presenting a definitive image, these exhibitions treat Rossi’s work as an open collection of forms and ideas. In doing so, they frame exhibition-making as an active part of heritage work – one that contributes to the ongoing reinterpretation of the archive.

Figure 2. Installation view of the exhibition Aldo Rossi: The architect and the cities (MAXXI, Rome, 2021), curated by Alberto Ferlenga. The retrospective brought together more than 800 works from Rossi's archive and international collections, including drawings, notebooks, photographs, letters, and a spectacular series of architectural models. Displayed thematically and chronologically, the exhibition emphasized Rossi's idea of the city as an accumulation of memory and form. Photo: Courtesy of MAXXI.

Installations

Installations provide a way to engage with Aldo Rossi’s Analogous City as an open and evolving space of interpretation. First conceived for the 1976 Venice Biennale with Eraldo Consolascio, Bruno Reichlin, and Fabio Reinhart (Ghirardo 2019, 16–23), the large collage combines drawings, references, and imagined architectural forms into a single, poetic composition. It is neither a plan nor a project in the traditional sense but rather a cartographic exploration shaped by resonance and association. Over time, the Analogous City has been studied and reproduced as a key expression of Rossi’s architectural imagination. However, when adapted as an installation, it reveals additional layers. Moving beyond its two-dimensional form, the work becomes an environment – something visitors can navigate physically and conceptually. These spatial versions invite engagement with its analogies, layers, and juxtapositions through movement in space. Instead of simply reproducing the collage, the installation opens new paths of interpretation, making the act of viewing itself part of the work’s meaning.

The Map of the Analogous City (Rodighiero 2015) was developed at EPFL as part of an exhibition at the Bonnefantenmuseum in Maastricht (Celant and Huijts 2015). It revisits Rossi’s collage through a curatorial and editorial approach focused on spatial reasoning. The project began with a close study of Rossi’s visual and textual references, including fragments from his notebooks, sketches, photographs, and publications. These elements were identified, collected, and assembled into a large-format printed map that expands and reframes the original panel. In this version, the Analogous City becomes an open archive – a network of references and associations. The editorial design emphasized the collage’s associative structure, enabling new relationships to emerge across its parts. Through its layout and visual clarity, the map invites viewers to explore the city’s logic as something fluid rather than fixed – encouraging reflection with Rossi’s ideas rather than about them. As Fabio Reinhart noted, the map is “an interpretation of the picture panel, a sort of re-invention within the framework of a new social and cultural reality” (Reinhart 2015). In this way, the Map of the Analogous City acts as a curatorial tool that renders architectural memory accessible through spatial exploration, while retaining the openness of the original work.

The Map of the Analogous City was also developed in digital form, enabling it to be presented in multiple exhibition contexts (Rodighiero 2022). At MAXXI, it appeared in the exhibition Aldo Rossi. L’architetto e le città (MAXXI 2021), where an interactive version allowed visitors to explore its internal network of references. Using touch-based navigation, viewers could follow visual links and recurring motifs drawn from Rossi’s notebooks and drawings. A year later, the digital installation was adapted for Aldo Rossi Digital, the museum’s online exhibition (MAXXI 2022), offering remote access and encouraging a slower, more reflective form of engagement. In both settings, the digital map transformed the collage into an exploratory environment, where viewers could navigate fragments, connections, and layered meanings. Instead of substituting for the physical installation, the digital version expanded its potential – inviting new forms of interaction while preserving the complexity and ambiguity at the core of the Analogous City.

Rossi’s installations invite interpretation rather than prescribe a single meaning. As Umberto Eco argues in The Open Work, some artworks derive their value from their capacity to support multiple readings, with meaning unfolding through the viewer’s active engagement (Eco 1989). John Dewey offers a complementary view: in Art as Experience, he describes how art becomes alive when it forms a unified event that reorganizes perception and draws the viewer into an active encounter (Dewey 1980). The Map of the Analogous City reflects these approaches. Its composition of drawn fragments – both real and imagined – does not offer a fixed narrative. Instead, it invites viewers into a process shaped by their own associations, memories, and spatial understanding. Each experience becomes a form of reconstruction, where meaning emerges through navigation and reflection. This openness aligns with Rossi’s view of architecture as a poetic and interpretive discipline. As Eco writes, open works “are brought to their conclusion by the performer at the same time as he experiences them on an aesthetic plane” (Eco 1989, 3). In this context, the performer is the visitor, who reads the map not as a finished object but as a space for ongoing interpretation.

The continuing reinterpretation of the Analogous City – across exhibitions, formats, and audiences – shows how installations can activate cultural heritage by reshaping how it is experienced. Whether printed, constructed, or digital, the work continues to evolve through each new presentation. These variations do not diminish Rossi’s legacy; they expand its relevance. Instead of fixing the past in place, the Analogous City keeps it in motion, creating a space where historical imagination, curatorial practice, and public engagement intersect. In this way, the installation functions not only as a site of preservation but as a space for reactivation. Rossi’s work thus contributes to a broader perspective on heritage – as something dynamic, negotiated, and performed over time.

Figure 3. Digital installation of Aldo Rossi's La Città Analoga (the Analogous City) at MAXXI, Rome. The original 1976 collage – created with Eraldo Consolascio, Bruno Reichlin, and Fabio Reinhart – is displayed here alongside a tablet device that allowed visitors to explore the work's references and fragments through digital navigation. This hybrid presentation illustrates how Rossi's emblematic collage has been reinterpreted in physical-digital form, inviting new modes of exploration and participation. Photo: Courtesy of MAXXI.

Discussion

Understanding Aldo Rossi’s heritage requires attention to how it is continually shaped – through archival processing, curatorial design, and installation practice. These mediations do not simply preserve; they reconfigure. Rossi’s work has remained culturally resonant not because it was stabilized but because it was continually opened up to new readings and audiences. In what follows, we reflect on three modalities of mediation – archival, curatorial, and performative – drawing on theoretical perspectives that help us grasp the interpretive nature of heritage-making. Rather than isolating these modes, we consider how they interact and contribute to a dynamic process of cultural transmission.

The archive of Aldo Rossi, particularly the work done by MAXXI in structuring and cataloguing his materials, exemplifies how institutional memory is constructed. Drawing on Paul Ricoeur’s notion of refiguration, we can see this archival process not as passive preservation but as an active recomposition of meaning. The inventory of Rossi’s papers and drawings is not merely a finding aid – it is a narrative structure. Through choices about inclusion, categorization, and description, the curators at MAXXI shaped how Rossi’s intellectual world would be encountered. This editorial work mediates between raw traces and historical understanding, offering a curated map of Rossi’s legacy that remains open to reinterpretation.

Exhibitions, as shown in L’architetto e le città (2021) and its digital counterpart Aldo Rossi Digital (2022), move Rossi’s legacy from the archive into public view. Bruno Latour’s concept of mediation helps frame this transition. Exhibitions are not neutral stages – they transform the meanings of the materials they present. The scenography, thematic structure, and even navigation paths all contribute to how Rossi’s work is made legible to new audiences. The exhibitions at MAXXI created environments where visitors do not merely observe but engage, constructing their own paths through Rossi’s concepts and images. These curatorial acts do not simplify the work – they amplify its interpretive depth by situating it in contemporary discourse.

Installations, especially those involving the Analogous City, extend this curatorial logic into participatory space. Here, Umberto Eco’s theory of the open work becomes relevant. Rossi’s ideas are not transmitted in a closed form but made available for interaction and recomposition. Whether experienced in the museum or online, the installations invite audiences to navigate spatially and intellectually. Each engagement is different, shaped by the viewer’s background, interest, and movement. These works do not fix Rossi’s meaning; they activate it, allowing the past to resonate through acts of reading, walking, and interpretation.

Together, the archival, curatorial, and performative mediations of Aldo Rossi’s work demonstrate that heritage is not a matter of fixing meaning but of enabling its rearticulation. Each mode engages memory differently: the archive organizes and reframes, the exhibition composes and translates, and the installation activates and invites. Through these layered practices, Rossi’s legacy remains in motion – open to reconfiguration without losing its integrity. This dynamic view of heritage challenges traditional distinctions between conservation and interpretation. Instead of treating mediation as a secondary layer, we see it as constitutive of meaning itself. Rossi’s case reveals how cultural memory is shaped by those who curate, design, and experience it. Far from diminishing the past, these mediations expand its reach. They suggest a model of heritage not as a closed inheritance but as a collective, ongoing work of reactivation.

Conclusions

Aldo Rossi’s legacy endures through the ongoing work of mediation. From archival reconstruction to curatorial framing and spatial installation, each mode of engagement reshapes how his work is understood, accessed, and remembered. The archive, carefully cataloged and made public, offers more than a resource – it becomes a structure of thought. Exhibitions and installations extend this structure into embodied and digital spaces, creating encounters that are interpretive rather than didactic.

What emerges from these practices is not a singular portrait of Rossi but a living body of work open to reconfiguration. His architecture, deeply reflective and formally poetic, lends itself to this openness. As this study has argued, heritage does not reside in preservation alone. It is sustained through acts of reactivation – through choices made by archivists, curators, and viewers alike.

These reflections speak more broadly to the work of heritage institutions and digital curatorship. As collections move across physical and digital environments, the process of meaning-making becomes more visible and more distributed. Institutions are asked to design settings where memory can be assembled, mediated, and interpreted with care, while digital formats open new paths for participation and access. Rossi’s case suggests directions for future research, inviting further attention to how digital tools shape curatorial practice, how audiences engage with hybrid forms of display, and how mediation can support more open and adaptive understandings of the past.

In this light, the past is not simply kept – it is made present, again and again, through thoughtful engagement and creative interpretation.

References

Burdick, Anne, Johanna Drucker, Peter Lunenfeld, Todd Presner, and Jeffrey Schnapp. 2012. Digital_Humanities. Cambridge, MA: MIT Press.
Cameron, Fiona, and Sarah Kenderdine, eds. 2007. Theorizing Digital Cultural Heritage: A Critical Discourse. Cambridge, MA: MIT Press.
Celant, Germano, and Stijn Huijts, eds. 2015. Aldo Rossi: Opera Grafica: Etchings Lithographs Silkscreen Prints. Cinisello Balsamo: Silvana Editoriale.
Dewey, John. (1934) 1980. Art as Experience. New York: Perigee Books.
Eco, Umberto. 1989. The Open Work. Translated by Anna Cancogni with David Robey. Cambridge, MA: Harvard University Press.
Eisenman, Peter. (1966) 1982. “The Houses of Memory: The Texts of Analogy.” In The Architecture of the City. Cambridge, MA: MIT Press.
Ghirardo, Diane. 2019. Aldo Rossi and the Spirit of Architecture. New Haven: Yale University Press.
Latour, Bruno. 2005. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press.
MAXXI. 2002. “Centro Archivi di Architettura.” MAXXI.
———. 2004. “Aldo Rossi. L’Archivio Personale.” MAXXI.
———. 2021. “Aldo Rossi: L’Architetto e la città.” MAXXI.
———. 2022. “Aldo Rossi Digital.” MAXXI.
———. 2023. “Collezione di architettura.” MAXXI.
Nora, Pierre. 1989. “Between Memory and History: Les Lieux de Mémoire.” Representations 26: 7–24.
Paul, Christiane. 2015. Digital Art. 3rd ed. London: Thames & Hudson.
Reinhart, Fabio. 2015. “From Picture Panel to City-Map.” In The Analogous City, the Map. Lausanne: EPFL Archizoom.
Ricœur, Paul. 1977. Memory, History, Forgetting. Translated by Kathleen Blamey and David Pellauer. Chicago: University of Chicago Press.
Rodighiero, Dario. 2015. The Analogous City, the Map. Lausanne: EPFL Archizoom.
———. 2022. “Extending Museum beyond Physical Space: A Data-Driven Study of Aldo Rossi’s Analogous City as a Mobile Museum Object.” International Journal for Digital Art History 6: 3.34–3.47. https://doi.org/10.11588/DAH.2021.6.77681.
Rossi, Aldo. (1966) 1982. The Architecture of the City. Translated by Diane Ghirardo and Joan Ockman. Edited by Peter Eisenman. Cambridge, MA: MIT Press.
———. 1999. I Quaderni Azzurri. Edited by Francesco Dal Co. Milano: Electa Editrice.
Smith, Laurajane. 2006. Uses of Heritage. 1st ed. London: Routledge.
UNESCO. 1972. Convention Concerning the Protection of the World Cultural and Natural Heritage.
———. 2009. “Charter on the Preservation of the Digital Heritage.”
Zhara Buda, Carla, Veronica Vignoli, Angela Parente, and Caterina Torrini. 2020. L’archivio Aldo Rossi nelle collezioni del MAXXI architettura: L’Inventario. Roma: MAXXI.

Four Guiding Principles for Rethinking Organizational Charts

2026-01-01T23:59:01+00:00

This chapter explores the transformative impact of digital tools on visualizing large organizations, highlighting the shortcomings of traditional organizational charts in capturing the interconnected nature of human dynamics. Emphasizing a respectful distance from simplistic managerial logic, it discusses four guiding principles for rethinking organizational charts with careful consideration of digital traces and the valuable information employees leave during daily practice. These principles aim to reconsider the centrality of individuals and the complex dynamics of the whole organization. The first principle stresses comprehensive representation of every employee, promoting concepts like inclusivity and irreductionism. The second principle fosters a sense of community through mutual and collective self-recognition. The third challenges traditional hierarchical structures in favor of equal representation in network-based models. The fourth underscores transparency and bottom-up collaboration in design. The chapter combines theoretical insights with practical examples, including the Affinity Map case study. This approach, in the light of opportunities offered by datafication, as described by Brinton (1939, 59-67), reimagines organizational charts to be more accurate, realistic, and useful. It serves as a guide for using digital data and computational tools to better understand and design large organizations.

Introduction

A few years ago, a Zurich company adopted a business model aimed at optimizing office space. When working from home was recognized as an efficient method to lower energy costs, the company realized that rents could be reduced by optimizing occupancy in smaller office spaces. As a result, employees were no longer bound to personal desks, instead changing their positions daily, based on their activities: a quiet room was appropriate when concentration was required, while an open-plan area suited teamwork. The choice of an appropriate desk could be made at the office entrance through a tablet computer providing an overview of available spaces. In addition, the tablet computer displayed the identities of employees already in the office, fostering opportunities for collaboration and enhancing the creation of a dynamic workspace.

This example illustrates how everyday working life can be influenced by guiding employees’ behaviors (Hatch 2011, 22–50). When technology is involved, as in the case above, it must be supported by real-time data collected through sensors, in a process frequently referred to as datafication (van Dijck 2014). Datafication relies on digital infrastructures that sense human activity in societal environments, providing data to implement recommendation engines for decision-making (Schrage 2020). This digital turn has enabled the development of more accurate and sophisticated tools, which have already been integrated into the field of organizational theory. This text specifically examines how the datafication of working environments can lead to more accurate organizational charts, transforming them into digital tools better suited for understanding private companies and public institutions.

Digital Traces in Decision-Making

Looking closer at tools that shape human behaviors, the digital turn has led to an information overload, triggering new ways of creating, organizing, and analyzing information, as Rob Kitchin extensively discusses in The Data Revolution (2014). The advent of the internet and the spread of digital devices have resulted in a continuous production of digital traces that describe daily practices with greater accuracy (Severo and Romele 2015). While datafication may benefit society, it is crucial to understand how digital traces can improve workplace dynamics. This chapter focuses on how digital tools enable a deeper understanding of large organizations by utilizing information design (Lima 2011) within organizational theory (Hatch and Cunliffe 2013). In particular, these pages investigate the tension between individuals and their communities in organizational charts, focusing on how employees and their collaborative dimension are represented (Rodighiero 2021).

The business model discussed earlier illustrates how digital tools support organizational theory. The Zurich company optimized office space using a recommendation system capable of “mathematically predicting personal preference” (Schrage 2020, 3), similar to how platforms like Spotify and Netflix suggest content based on past consumption. In medium and large companies, anticipating employee preferences requires first collecting digital traces. These traces can come from objects such as identity cards, smartphones, or laptops. For example, an employee’s position can be inferred when their identity card opens a door, their smartphone accesses an email application, or their laptop connects to the local network. Similar outcomes can also be achieved using more invasive technologies like facial recognition, which is spreading quickly for security purposes (Roussi 2020). Although this chapter does not delve into the various methods of producing digital traces, it is important to note that personal behaviors can be captured in many ways. Without this collection process, there would be no data to power recommendation systems and better understand an organization’s internal dynamics.

Datafication is a process with distinct phases that provide a stream of continuous information. While data creation, organization, and use might seem linear, as famously illustrated by Nathan Shedroff (1999), the process is actually circular: data describes an iterative life cycle that continually accumulates information, represented as a ring (Leonelli 2019). This perpetual flow of information presents organizations as dynamic entities shaped by data, able to react when internal conditions change (Hatch 2011, 1–21). Digital traces from employees act as signals that help adjust the organization in real time, functioning like a living organism—much as the body begins digestion after eating or the immune system responds to a virus.

According to psychologist Kurt Lewin’s field theory (1951), employees function as forces that maintain organizational equilibrium, which can be disrupted by events that lead to a renewed balance (Hatch and Cunliffe 2013, 290). In this context, real-time data serve as agents, accelerating the process of reorganization in terms of speed and optimization. The flow of employee data creates a symbiotic relationship between individuals and the organization through decision-making systems.

This mutual relationship between individuals and the organization paints a more sophisticated picture than Shedroff and Leonelli’s diagrams. Drawing on Boris Groys’ (2008) theory of self-design, which describes how social networks shape personal identity, the datafication diagram is complete when the flow of information and the end users overlap as two sides of the same coin. In organizational settings, this means that recommendation systems shape employees, who in turn shape these systems with their digital traces. This interpretation turns the circular life cycle of data into a lemniscate, connecting two rings: one for the organization and one for the employees within it. As illustrated in Figure 1, although these two processes share the same flow, they remain distinct (Rodighiero 2021, 106; Rodighiero and Romele 2020, 367).

Figure 1. The data life cycle comprises two iterative loops concerning the device and user. The user's behaviors are captured in data, successively computed, and presented on the device. The lemniscate is completed when the users modify behavior by interpreting the device's recommendation. Source: Rodighiero 2021, 106; Rodighiero and Romele 2020, 367.

Organizational Decisions Through Digital Traces

The metaphor of adaptive organisms is often used to describe resilience to market volatility, where organizations must navigate rapid changes, but it also highlights the competition for authority that shapes internal power dynamics (Hatch 2011, 1–21). Authority creates disparities in control and salaries among employees, significantly influencing how organizations are governed. For this reason, sensitive matters such as promotions are of great interest to both employees and the organization. Promotions confer authority, which translates into more influence over organizational management. As a result, promotions are tightly regulated by bureaucratic mechanisms designed to ensure fairness, as noted by sociologist Max Weber (2019, pp. 335–447), who argued that bureaucracy favors more impartial management.

If promotions rely on bureaucratic documents summarizing candidates’ profiles, it is worth considering the future role of digital traces in this decision-making process. Should digital traces influence organizational decisions, new empirical ways of integrating algorithmic thinking into decision-making could emerge, particularly through digital tools. For instance, examining the impact of digital traces on academic careers provides an illustrative scenario. Citation analysis, frequently used by academic institutions and research centers, not only serves as an evaluation tool but also influences recruitment—especially in technical fields like computer science, where citing is the primary way to demonstrate professional recognition. Recognizing that operational rules vary across organizations, the next section introduces an academic case study to lay the groundwork for broader generalizations applicable to medium and large organizations.

Rethinking Organizational Charts

One of the opportunities offered by datafication is revisiting organizational charts, diagrams used by management to visualize hierarchies and plan internal activities (Brinton 1939, 59–67). Organizational charts have been part of organizational studies since the early twentieth century, when consulting engineer Willard C. Brinton published Graphic Methods for Presenting Facts (1919), reflecting industry interest in visual planning. However, since Daniel McCallum’s detailed mid-nineteenth-century visualization, visible in Figure 2 (Rosenthal 2013), organizational charts have been underestimated and drastically simplified (Rodighiero 2021, 28). This simplification has diminished their ability to represent the sociotechnical dimensions of human complexity (Latour 2005) and the diversity that makes each individual unique (Elias 1991). Rethinking organizational charts requires moving away from simplistic managerial logic and carefully considering the use of digital traces. Left by employees during daily practices, these traces are vital for refocusing on individuals and capturing the complex dynamics of the entire organization.

Moreover, digital traces might finally resolve the tension between individual elements and the whole, precisely reflecting the relationship between employees and the organization. A thoughtful organigram must carefully account for both an organization’s complexity and its members’ identities, shaping a comprehensive form constituted by individuals. Philosophically speaking, this question dates back to ancient times, initially examined by Aristotle (Cohen and Reeve 2020) and more recently reformulated by Gestalt theory (King and Wertheimer 2005). When Christian von Ehrenfels (1937, 523) famously wrote that the whole is more than the sum of its parts, he could never have imagined that his philosophical insight would inspire designers representing complexity (Bürdek 2015). Reframing this tension between the elements and the whole in organizational studies provides a novel perspective for rethinking organizational charts. Just as typographers carefully balance the weights of text, images, and negative spaces in a page layout, organigrams must be accurately designed to represent the social dynamics regulating organizations rather than static hierarchies. Indeed, organizational charts today are primarily used to depict hierarchy—a limitation that hinders understanding of organizational mechanisms and dynamics. This reflection calls for abandoning pyramidal or tree-shaped diagrams in favor of visual models that capture the richness of each individual and the complexity of human relationships. In this context, digital traces offer raw material to mold such complexity in unprecedented ways.

Figure 2. In the mid-nineteenth century, Daniel McCallum drew an incredibly detailed organizational chart to reorganize the company. Representing all the employees, the diagram looks like a tree where the board of directors is the root and the rail tracks are the branches. Source: Rosenthal 2013.

Today, network visualizations are among the most common methods for decoding societal complexity using digital traces. Although their scientific popularity is relatively recent (Scott 2000), networks originated in the eighteenth century when mathematician Leonhard Euler (1953) developed graph theory. Euler’s work, addressing the Königsberg bridges problem—whether all city bridges could be crossed without repeating any—created a mathematical object called the graph. Over time, this abstraction evolved into network visualizations, beginning with psychiatrist Jacob L. Moreno in the twentieth century. Moreno, using tabular questionnaires to investigate friendships among schoolchildren, devised sociograms: nodes represented children, and directed lines indicated friendships (Moreno 1934, 32). This innovation translated Euler’s mathematical abstraction into an intuitive visual grammar, presenting complex data more clearly to readers.

Decades after sociograms, the computational turn significantly expanded Moreno’s diagrams, enabling the analysis of large social structures in a research field coined “computational social science” by American scholars (Lazer et al. 2009). Beyond sociologists, graphic designers were also drawn to the visual complexity of large networks, experimenting with elaborate visualizations, as exemplified by Manuel Lima’s archival work (2011). This shared interest in network visualizations culminated in two groundbreaking laboratories: Albert-László Barabási’s lab at Northeastern University (Barabási et al. 2020) and Bruno Latour’s médialab at Sciences Po Paris (Venturini et al. 2017). These research initiatives established network visualizations (see Figure 3) as essential tools for analyzing human behavior by merging social sciences, computer science, and digital design (Romele and Rodighiero 2020, 111–119).

Figure 3. This network visualization created for the Digital Humanities conference 2014 embodies the effort to represent a scientific community's human dynamics. Each node is an author, and relations are based on shared keywords. Source: Rigal and Rodighiero 2015.

Organizational Charts After Datafication

The redesign of organizational charts using data is possible due to the convergence of four key factors. First, the widespread production of digital traces addresses the loss of complexity often found in traditional organizational charts (Rodighiero 2021, 28). Second, advancements in digital design allow experts to create more sophisticated, interactive visualizations (Löwgren and Stolterman 2004). Third, information design has emerged as a recognized field of research, uniting specialists from diverse disciplines (Meirelles 2013; Tufte 1997). Fourth, significant improvements in network algorithms enable the handling of larger datasets while offering flexible, web-based solutions to engage broader audiences (Bostock, Ogievetsky, and Heer 2011).

These four factors form a foundation for rethinking organizational charts, culminating in the development of four design principles. These principles stem from insights gained during my doctoral and postdoctoral research, particularly the collaborative creation of the Affinity Map (Rodighiero 2021). Revisiting and refining personal research is a gradual process, as demonstrated in this chapter, where earlier findings have been restructured into a practical framework for use by private companies and public organizations. Refined over the postdoctoral years, this list of observations was formulated and presented for the first time at Université de Liège during a seminar with Frederik Stjernfelt titled Images Today: Archives, Identities, and Algorithms and organized by Enzo D’Armenio and Maria Giulia Dondero (Rodighiero 2022).

1. Represent Every Employee

As mentioned above, organizational charts have been overly simplified in recent decades. In the mid-nineteenth century, Daniel McCallum realized an incredibly detailed organigram (see Figure 2) to reorganize approximately 800 kilometers of New York and Erie Railroads (Rosenthal 2013). The fascinating capacity of McCallum’s diagram is that it entirely represents each individual working for the railway company, from the directors on the board to the workers distributed along the rails. Although the format is not large enough for displaying full names, everyone is equally represented in what can be defined as a more egalitarian organigram when compared to more recent diagrams representing only the management. In addition, McCallum’s diagram is oriented like a tree whose roots are represented by the board and leaves by workers, while today’s organizational charts present the president at the top of a pyramid, like a pharaoh or a divinity in a panoptic position. This common representation adds an unhealthy sense of superiority into the organization’s dynamics.

Digital traces allow us to draw a more democratic way to represent employees as if datafication is returning an objective, honest signal (Pentland 2008). For example, digital traces do not lie when mapping workers’ presence in the office space: they will show who is in the office and who is not by using identity cards or Wi-Fi connections. Also, imagine the richness that email conversations can leave behind: today’s privacy rules do not let us use this kind of information, but they would reveal connectivity different from hierarchical relations.

2. Convey a Sense of Community

One issue related to the presence of all employees in a diagram concerns the feeling of being part of the same community. Often, private organizations publish websites where only high-ranked employees have a dedicated web page. Why does this happen? Creating a web page for each employee is costless and does not affect the company’s environment. When some employees—also a tiny part—are excluded from organigrams, their identity is not part of something bigger and is automatically excluded.

It is essential to realize that organizational charts, like group pictures of student classes, are not mere visual representations but powerful instruments for mutual recognition (Rodighiero and Cellard 2019, 10). If we are not represented, we cannot be recognized; consequently, the sentiment of belonging cannot be developed. The mirror metaphor cannot reduce organigrams, allowing one only to recognize the self. Mutual recognition requires reciprocal connectivity between individuals, establishing peer-to-peer connections between employees to be part of a community. The overall connectivity can be described only through these personal connections. The whole and its elements are based on elementary relationships, embodying the whole community’s glue.

Inclusiveness in organizational charts, as much as in working environments, can be achieved through a democratic use of technology. When datafication is put in place, the process has to interest all the employees. For example, when an employee scans a badge to enter the office, this is true for the office worker as well as for the manager, and this is something that unites individuals. Another good example is given by the internal dynamics that happen equally at all levels of the organization, showing that connectivity is a characteristic that should be used in organizational charts to give a sense of community. If all the employees have networked with each other, it is simple to imagine an organigram where all the elements are part of the same picture.

3. Avoid Evaluation and Comparison

One of the most challenging issues in designing an organizational chart is the risk of comparison that characterizes hierarchical structures, and network visualizations offer a more democratic solution. For example, when describing the Actor-Network Theory, Bruno Latour always pointed out the flatness of networks, capable of situating all human and non-human actors on the same level (Latour 2005, 171–172). As a result, network visualizations are more egalitarian representations than trees because of the absence of orientation or predominant positions—pharaohs would be represented not at the apex of the pyramid but rather at the center of the network.

Individuals are not comparable when represented with the same symbol. However, representing individuals with symbols of different sizes according to the employees’ salaries, for example, stresses quantitative differences that reduce individual complexity. Try to imagine an organizational chart in which the size of individuals is represented by their salary or their seniority; even though the information is correct, these values do not represent the overall importance of individuals, which can be demonstrated by other dimensions like connectivity or recognition. This example proves why financial data are often unavailable and sets a threshold to understand what has to be kept in the private sphere (Rancière 2000). However, other information, like the collaborations we experience daily in the working environment, should be visible as an activity taking place in public spaces.

4. Support Transparency and Inclusion

When creating an organizational chart, it is essential to disclose the methodology by making visible the process of decision-making that is brought to finalize the map. For example, employee network visualization for representing employees also means disclosing the metric employed in the visual. Transparency is essential to reveal the thought behind creating a representation, as it typically happens in politics today.

When the Affinity Map was presented to one thousand individuals represented (see the case study below), it was necessary to be as inclusive as possible from datafication to the presentation throughout the design process. At the beginning of the project, employees were asked if they wanted to be part of the organizational chart, and the design process was modified after employees expressed interest in being part of it.

The agreement with the individuals represented is part of the disclosure of an organigram. In the Affinity Map, for example, the agreement was probably one of the most challenging steps of the design process. Some individuals expressed discomfort in displaying their names to the public at a certain point but were okay with sharing information internally. This bifurcation caused the creation of two maps: one for the employees with their identities and one for the public with just the names of laboratory members.

Digitalization in Academia

Like most organizations, universities must cope with digitalization. The daily practice of scholars and employees has been increasingly translated into data over the years, radically changing how academia works. An excellent example of how datafication affected academia is represented by the considerable attention that citation impact received in the last decades. Initially conceived to provide innovative tools to help librarians with journal subscriptions (Garfield 1970), the citation metric suddenly became a decision-making tool to assess scholars through scientific publications (Garfield 2006). Although the value of such a measure to find the most relevant scientific literature is not questioned, its widespread use has heavily distorted the job market of universities and research centers (Gingras 2014). This distortion is visible, for example, in the unappealing prospects of minor topics whose coverage is bound to receive fewer citations than popular research, such as the more attractive artificial intelligence (Dick 2019). When the citation index became crucial in academic recruitment, scholars became interested in increasing the number of citations to be more attractive to employers.

This citation-based evaluation undoubtedly relies on a reductive model that ignores many dimensions of academic practice, focusing on literature. “Scientific and academic practices are more complex than the sole scientific literature. They are heterogeneous, multidimensional, and as rich as human nature” (Rodighiero 2021, 20). This contradictory example shows how digitalization can affect employees positively and negatively.

Recent initiatives from educational institutions opposed this unequal simplification by establishing impartial rules, as in the case of the career diversity proposed in the Netherlands (VSNU et al. 2019). This initiative draws attention to how academic practice is much more diversified than what is currently revealed by digitalization. This variety is immediately evident when observing scientists through the eyes of an ethnographer (Latour and Woolgar 1986) or when looking at organizations as complex relational systems where individuals are connected by collaboration at different levels (Rodighiero 2021). In the past few decades, the academic world has been harmed by excessive attention to citations in recruitment, triggered by a fragmentary translation of everyday practice into data. Although technological distortions also occurred in the private sector where, for example, managers monitored employees during the lockdown (Satariano 2020), digitalization brought enormous benefits to companies (Parviainen et al. 2022). While severe issues concerning the right to privacy are tackled by recent regulations (UNESCO 2022, 21), it is crucial to make the most of digitalization by understanding its incomplete and progressive translation from society. As stated by José van Dijck, “Datafication as a legitimate means to access, understand and monitor people’s behavior is becoming a leading principle, not just amongst techno-adepts, but also amongst scholars who see datafication as a revolutionary research opportunity to investigate human conduct” (van Dijck 2014, 198).

The Case Study of Affinity Map

The Affinity Map (see Figure 4) is a project focusing on mapping academic institutions through the metric of collaboration. Initially conceived as an instrument for decision-making for managers and directors of the ENAC school, a public assembly unanimously decided to transform it into an instrument for all employees. Starting with an investigation regarding available digital traces and network visualization, the Affinity Map was developed to show the identity of all the employees to make the representation as inclusive as possible. Additionally, the metric based on collaborations in publications, courses, and supervisions—collected by institutional information systems—was offered as an alternative to citation indexes not only to foster interdisciplinarity among scholars but also to raise awareness about the dynamics within the faculty.

The double connectivity level is one of the Affinity Map’s most distinctive aspects. When networks typically connect elements on a unique sole surface, the Affinity Map introduces nodes containing a network. The result is a network of laboratories in which each node is a network of laboratory members. This graphic solution simplifies the overall connectivity, avoiding the noisy overlapping of thousands of lines and protecting individuals. Academic newcomers represent the most visible example of this latter consequence. When, for example, doctoral students join a laboratory, they often do not have collaborations because they have just started. The result in mapping is a detached node at the edges of the network visualization that creates a sense of not being part of the community. Laboratory nodes, in this sense, work like cocoons that protect the identity of individuals under the responsibility of the laboratory director.

Figure 4. The Affinity Map shows the ENAC Faculty as a constellation of laboratories, connected by a multidimensional metric inspired by the idea of actual and potential collaborations. Employees are equally represented within the laboratories and visible after zooming. Source: Rodighiero 2021.

Another dimension of inclusiveness was achieved by unveiling the design process to the employees represented on the map. As in the academic environment, evaluation is a serious topic of discussion, and the Affinity Map would not be another academic metric for assessing the work of scholars. Instead, the map was disclosed during a public event to demonstrate the willingness to include all the employees in the design process. For example, a large version of it, measuring 15 by 15 meters, was printed on tarpaulin and placed on the floor of a large hall in the ENAC schools (Rodighiero 2018). During the event, employees were invited to walk on the Affinity Map and speak with the team collaborating to design it. This knowledge transmission was beneficial in unveiling the design process and listening to the employees’ reactions and suggestions. The design process became collective, and all the individuals represented on the map were able to contribute. For example, a member concerned with the map’s output to the generic public asked for a higher threshold of privacy of identities. Finally, the design team considered this comment so seriously that when the map went online, all users’ identities were revealed only through logging into the institution, and, consequently, not accessible outside.

Along with these concerns that affected the design process, we finally realized that the Affinity Map was an instrument and a symbol of representation. From the beginning, all the school members had to be part of the organigram. In addition, the lack of orientation in network visualizations removed the pyramidal view from the director’s point of view but introduced the centrality problem. At a certain point, we realized that individuals at the edges of the networks were unhappy with such a positioning. After the doctoral studies, this issue was analyzed more thoroughly, envisioning mapping on continuous spherical surfaces.

One of the most promising outcomes of the Affinity Map was the linguistic distance between individuals. Such a digital trace was probably the most promising because language is a dimension that impacts every individual, but only the latest advancements in artificial intelligence allowed for the implementation of solutions from the computation of large corpora that were not available before (Moretti 2005). For example, if one cannot be connected to other colleagues because he is working individually, he still belongs to the community. Language and the words scholars use in publications and presentations are essential metrics to describe organigrams where no one is disconnected from the community. In addition, the linguistic dimension can reveal more hidden dimensions in daily working practices; for example, uncovering potential collaborations between individuals, which in a large organization, may have never had the opportunity to meet.

Conclusion

Scholars of organizational studies have to look at digital traces and network design to advance their studies at a moment when technology pervades organizations. Digital traces can extend the perception of the discipline by providing a more detailed level of information and revealing some unexpected patterns in human behavior. However, as demonstrated by the case study of the Affinity Map, some criticalities still have to be resolved, starting from the inaccuracy of datafication. The window of opportunity that computational means provide offers new solutions and challenging problems. The question of representing the dynamics of medium and large organizations is fascinating and helps to reflect on the current limits and future perspectives as the process of datafication is under development.

If you want to read more, you can discover the further scope of the Affinity Map by reading Mapping Affinities, published by Métis Presses (Rodighiero 2021). Further studies about the democratization of organizational charts can be found in publications by the MIT Press (Rodighiero et al. 2022), IEEE (Rodighiero 2020), and De Gruyter (Rodighiero et al. 2024). A more recent review of the last ten years of research in mapping scientific communities has been published by Sage in open access (Rodighiero 2024).

References

Barabási, A.-L., M. Bello, J. Kluge-Fabényi, K. Forde, J. Készman, I. Meirelles, C. G. Ratti, M. Ritchie, A. Szántó, P. Weibel, and A. Stang. 2020. Hidden Patterns: Visualizing Networks at BarabásiLab. Berlin: Hatje Cantz Verlag.
Bostock, M., V. Ogievetsky, and J. Heer. 2011. “D3: Data-Driven Documents.” IEEE Transactions on Visualization and Computer Graphics 17 (12): 2301–2309. https://doi.org/10.1109/TVCG.2011.185
Brinton, W. C. 1919. Graphic Methods for Presenting Facts. New York: The Engineering Magazine Company.
———. 1939. Graphic Presentation. New York: Brinton Associates.
Bürdek, B. E. 2015. Design: History, Theory and Practice of Product Design. Berlin: Birkhäuser.
Cohen, S. M., and C. D. C. Reeve. 2020. “Aristotle’s Metaphysics.” In The Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. https://plato.stanford.edu/archives/win2020/entries/aristotle-metaphysics/
Dick, S. 2019. “Artificial Intelligence.” Harvard Data Science Review, July 1. https://doi.org/10.1162/99608f92.92fe150c
Elias, N. 1991. The Society of Individuals. Edited by M. Schröter. Oxford: Basil Blackwell.
Euler, L. 1953. “Leonhard Euler and the Koenigsberg Bridges.” Scientific American 189 (1): 66–72.
Garfield, E. 1970. “Citation Indexing for Studying Science.” Nature 227 (5259): 669–671. https://doi.org/10/b9962z
———. 2006. “The History and Meaning of the Journal Impact Factor.” JAMA 295 (1): 90–93. https://doi.org/10/bgvhsf
Gingras, Y. 2014. Les dérives de l’évaluation de la recherche: du bon usage de la bibliométrie. Paris: Raisons d’agir éditions.
Groys, B. 2008. “The Obligation to Self-Design.” E-Flux, November 1. http://www.e-flux.com/journal/00/68457/the-obligation-to-self-design/
Hatch, M. J. 2011. Organizations: A Very Short Introduction. Vol. 264. New York: Oxford University Press.
Hatch, M. J., and A. L. Cunliffe. 2013. Organization Theory: Modern, Symbolic, and Postmodern Perspectives. 3rd ed. Oxford: Oxford University Press.
King, D. B., and M. Wertheimer. 2005. Max Wertheimer & Gestalt Theory. New Brunswick: Transaction Publishers.
Kitchin, R. 2014. The Data Revolution: Big Data, Open Data, Data Infrastructures and Their Consequences. London: SAGE Publications.
Latour, B. 2005. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press.
Latour, B., and S. Woolgar. 1986. Laboratory Life: The Construction of Scientific Facts. Princeton, NJ: Princeton University Press.
Lazer, D. M. J., A. Pentland, L. Adamic, S. Aral, A.-L. Barabási, D. Brewer, N. Christakis, N. Contractor, J. Fowler, M. Gutmann, T. Jebara, G. King, M. Macy, D. Roy, and M. V. Alstyne. 2009. “Computational Social Science.” Science 323 (5915): 721–723. https://doi.org/10.1126/science.1167742
Leonelli, S. 2019. “Data Governance Is Key to Interpretation: Reconceptualizing Data in Data Science.” Harvard Data Science Review, July 1. https://doi.org/10.1162/99608f92.17405bb6
Lewin, K. 1951. Field Theory in Social Science: Selected Theoretical Papers. Oxford: Halpers.
Lima, M. 2011. Visual Complexity: Mapping Patterns of Information. New York: Princeton Architectural Press.
Löwgren, J., and E. Stolterman. 2004. Thoughtful Interaction Design: A Design Perspective on Information Technology. Cambridge, MA: MIT Press.
Mayer-Schonberger, V., and K. Cukier. 2013. Big Data: A Revolution That Will Transform How We Live, Work, and Think. London: John Murray.
Meirelles, I. 2013. Design for Information: An Introduction to the Histories, Theories, and Best Practices behind Effective Information Visualizations. Beverly, MA: Rockport.
Moreno, J. L. 1934. Who Shall Survive? Washington: Nervous and Mental Disease Publishing Co.
Moretti, F. 2005. Graphs, Maps, Trees: Abstract Models for a Literary History. London: Verso.
Parviainen, P., M. Tihinen, J. Kääriäinen, and S. Teppola. 2022. “Tackling the Digitalization Challenge: How to Benefit from Digitalization in Practice.” International Journal of Information Systems and Project Management 5 (1): 63–77. https://doi.org/10.12821/ijispm050104
Pentland, A. 2008. Honest Signals: How They Shape Our World. Cambridge, MA: MIT Press.
Rancière, J. 2000. Le partage du sensible: esthétique et politique. Paris: La Fabrique-éditions.
Rigal, A., and D. Rodighiero. 2015. “Trajectoire d’une représentation cartographique en réseau.” Cartes & Géomatique 225: 33–41.
Rodighiero, D. 2018. “Printing Walkable Visualizations.” In Proceedings of the 5th Biennial Research Transdisciplinary Imaging Conference, 58–73. https://doi.org/10.6084/m9.figshare.6104693.v2
———. 2020. “Drawing Network Visualizations on a Continuous, Spherical Surface.” Proceedings of the 24th International Conference Information Visualisation (IV), 573–580. https://doi.org/10.1109/IV51561.2020.00097
———. 2021. Mapping Affinities: Democratizing Data Visualization. Open-access English edition. Geneva: Métis Presses. https://doi.org/10.37866/0563-99-9
———. 2022. “The Lemniscate of Data Visualization: Production and Consumption of Data from an Actor-Network Perspective.” Paper presented at Images Today: Archives, Identities, and Algorithms, Université de Liège, March 29.
———. 2024. “In Praise of Visual Representation: An Inquiry into Text Analysis and Network Visualization for Charting Scientific Communities.” Visual Communication 23 (3): 513–526. https://doi.org/10.1177/14703572241248408
Rodighiero, D., and L. Cellard. 2019. “Self-Recognition in Data Visualization: How Individuals See Themselves in Visual Representations.” EspacesTemps.Net, August. https://doi.org/10.26151/espacestemps.net-wztp-cc46
Rodighiero, D., and A. Romele. 2020. “The Hermeneutic Circle of Data Visualization: The Case Study of the Affinity Map.” Techné: Research in Philosophy and Technology 24 (3): 357–375. https://doi.org/10/gg6w5p
Rodighiero, D., P. Rivière, and S. Kenderdine. 2024. “Super-Vision: Tracing EPFL History through 8,000 Doctoral Theses.” Journal of Digital History 3 (1).
Rodighiero, D., E. Wandl-Vogt, and E. Carsenat. 2022. “A Visual Translation of the Pandemic.” Leonardo 55 (3): 297–303. https://doi.org/10.1162/leon_a_02203
Romele, A., and D. Rodighiero. 2020. “Digital Habitus or Personalization without Personality.” HUMANA.MENTE Journal of Philosophical Studies 13 (37): 98–126. https://doi.org/10/ghm6cn
Rosenthal, C. 2013. “Big Data in the Age of the Telegraph.” McKinsey Quarterly, March 1. http://www.mckinsey.com/business-functions/organization/our-insights/big-data-in-the-age-of-the-telegraph
Roussi, A. 2020. “Resisting the Rise of Facial Recognition.” Nature 587 (7834): 350–353. https://doi.org/10.1038/d41586-020-03188-2
Satariano, A. 2020. “How My Boss Monitors Me While I Work from Home.” The New York Times, May 6. https://www.nytimes.com/2020/05/06/technology/employee-monitoring-work-from-home-virus.html
Schrage, M. 2020. Recommendation Engines. Cambridge, MA: MIT Press.
Scott, J. 2000. Social Network Analysis: A Handbook. 2nd ed. London: SAGE Publications.
Severo, M., and A. Romele, eds. 2015. Traces numériques et territoires. Paris: Presses de Mines. https://doi.org/10.4000/books.pressesmines.1984
Shedroff, N. 1999. “Information Interaction Design: A Unified Field Theory of Design.” In Information Design, edited by R. Jacobson. Cambridge, MA: MIT Press.
Tufte, E. R. 1997. Visual Explanations: Images and Quantities, Evidence and Narrative. Cheshire, CT: Graphics Press.
UNESCO. 2022. Recommendation on the Ethics of Artificial Intelligence. Programme and Meeting Document SHS/BIO/PI/2021/1. Paris. https://unesdoc.unesco.org/ark:/48223/pf0000381137.locale=en
van Dijck, J. 2014. “Datafication, Dataism and Dataveillance: Big Data between Scientific Paradigm and Ideology.” Surveillance & Society 12 (2): 197–208. https://doi.org/10.24908/ss.v12i2.4776
Venturini, T., M. Jacomy, A. Meunier, and B. Latour. 2017. “An Unexpected Journey: A Few Lessons from Sciences Po Médialab’s Experience.” Big Data & Society 4 (2). https://doi.org/10/gf7tk2
Von Ehrenfels, C. 1937. “On Gestalt-Qualities.” Psychological Review 44 (6): 521–524. https://doi.org/10/bd8dpw
VSNU, NFU, KNAW, NWO, and ZonMw. 2019. Room for Everyone’s Talent: Towards a New Balance in the Recognition and Rewards of Academics. https://www.nwo.nl/en/position-paper-room-for-everyones-talent
Weber, M. 2019. Economy and Society: A New Translation. Edited by K. Tribe. Cambridge, MA: Harvard University Press.

Grounding AI: Exploring the Role of Algorithms in Science

2026-01-01T23:59:00+00:00

You are seeing a map — but not of streets or cities. This is a map of two million scientific papers about artificial intelligence, machine learning, and algorithms, where each dot is a paper and each cluster is a conversation happening in science about how algorithms are being put to work in the world. Like any city map, it invites you to explore: zoom in on a neighborhood, follow a path, or simply wander, and the further you walk the more the landscape changes — from medicine to education, from language to vision, from prediction to control. Take your time with it, asking yourself which of these applications you had heard of, which ones surprised you, and whether more of them should be part of the public conversation we are all having about AI.

Unfolding AI, or Learning to Read the Sky

There is a version of AI that lives in headlines, in social media posts, in the quiet shared glance when someone mentions it at dinner. It feels enormous and yet hard to grasp, always there, almost inevitable by now. It presents itself as both promise and threat. You probably know this version, as most of us do.

But if you stop to think about it, what are we actually talking about when we talk about Artificial Intelligence? What do we have in mind, exactly? Is it a song our music platform recommended to us? The chat we ask for help when replying to emails? A hospital’s triage system? An autonomous flying drone?

When we talk about AI this way, we are building a figure — singular, capitalized, implicit — that doesn’t need to explain where it is or what it’s doing, because it’s everywhere and doing everything, or about to be. Even critique tends to accept this framing. Whether you think AI will save us or end us, whether you’re excited or exhausted by the conversation entirely, the grammar stays the same: “AI will, AI can, AI threatens.” A single actor with a single story.

This construction, this representation of AI, is real. We are all living through it, whether we work with algorithms daily or only encounter them as invisible infrastructures operating around us. The unease is real. The excitement is real. The feeling that something large is shifting, or has already shifted, is real. But all of this is a specific view on the matter. A view from somewhere. Maybe a view from nowhere. Definitely a view from a distance. And when we look at things from a distance, we tend to reduce, simplify. Our minds turn details into outlines, making the multiple singular. We see a shape, but only once we stop seeing everything that shape is made of.

You are holding something that was made in response to that distance. Not to dismiss what you can see from there, but to ask: what happens when you step closer? What is actually there?

Look at the map you are holding right now. A folded square, about the size of a book. Unfold it all the way. What you’re looking at, once it’s open, is what we called the Grounding AI Map: a visualization of around two million scientific papers about artificial intelligence, algorithms, and machine learning, retrieved from the Scopus database and published between 1985 and 2024.

In the map, each document becomes a dot, and these are arranged by semantic similarity: papers that share vocabulary, problems, and technical approaches sit close to one another, and the dense regions can be seen as clusters of research that grew up around common questions. The colors trace time, from red for older work to blue for newer. The labels you’ll find scattered across the map were generated through an AI-enabled pipeline, where we trained a large language model to work with us, reading each cluster from the inside and creating a title and a short summary for each of the 4062 clusters that we found. Just remember that ours is an invitation to explore, not an attempt to give answers.

From up close, you can see what scientists actually work on. There is research on smart wearables, on the chemical composition of soil, on how to identify trees from drones, on detecting fraud in financial transactions, on optimizing energy grids. Each cluster is a community of researchers working on something specific, with their own methods, their own debates, their own slow accumulation of findings.

People often say it looks like a galaxy — densities, gradients, something that resembles deep space, or weather systems seen from very far above. The shapes suggest nebulae, or the terrain of some other planet, something your eye keeps trying to resolve into a recognizable figure.

“Why is it called Grounding AI?” you might ask. This version of the map, the one you can hold, fold, and carry with you, is a pocket-size version of the same visualization that we printed as a floor mat large enough to walk on. We literally place AI research on the ground. The map becomes a stage where you move through research areas, questions, and ideas. Stand in the middle of a cluster about medical diagnostics, or flood prediction, or materials science, and you’ll be able to read only what’s immediately around you: the labels at your feet, the neighboring regions within a few steps. The rest of the map becomes blurry. You can see where it ends, but you can’t read all the labels at the same time. It is not a view from nowhere. It forces you to place yourself somewhere specific, reading from a position you had to choose — even if you chose it by accident. Your body becomes part of the experience, of the reading.

This foldable map does something different. Think of it as a wayfinder, an astronomer’s chart: you can see the whole sky at once, but you’re seeing it reduced, compressed, without the granular detail of the full installation. Of the 4062 clusters, only 72 are labelled here, not because they are more important than the others, they were just selected as a sample to be distributed evenly across the map. The trade-off is legibility. This version of the map helps you orient yourself, find the region you want to explore before you step onto it, or remember where you were standing after you’ve left. Folded, it contains everything the unfolded version does — the same two million papers, the same clusters, the same semantic distances — but in a form you can carry. The act of unfolding becomes more than a practical gesture. Unfolding is the gesture the whole project is asking you to make.

A galaxy has a name. The Milky Way. Andromeda. Names that make the unknown vastness just a bit more familiar, so that we can point up and say “that one”. We give names to things to make sense of them, to categorize the world. But once we give a name to something so immense like a galaxy, we turn its multiplicity into singularity. We make a figure out of something that was never trying to be one.

Constellations work the same way. We looked up at the sky and found hunters, bears, wagons — shapes that felt inevitable once someone pointed them out. But they’re not inevitable. They are stories we told from exactly this distance, because we needed the sky to make sense. The Greeks saw a hunter where the Japanese saw a drum. The stars were the same. They just got caught in different stories, held inside different sets of human needs and hopes and fears.

The stars themselves are real. The constellation doesn’t lie about them — it simplifies. It collapses something intricate and particular into a silhouette you can remember, a shape you can find again on a different night. We made up constellations to talk about the stars, and then, slowly, we started talking about the constellations instead. The hunter acquired a mythology. The stars disappeared behind it.

AI, as one thing, works like this.

A name given from a distance, one that turns something multiple, specific, and deeply interconnected into a single figure with a single story. Sometimes the figure is a threat, sometimes a promise, sometimes both at once, depending on who’s speaking and what they need it to mean. But it remains one thing, with one trajectory, moving through the world as if it had intentions and moods and a coherent will.

When you only have the constellation, you lose the specific distances between the stars, the difference in their temperatures, the fact that some of them burned out centuries ago and you’re seeing old light. You lose the granularity. You’re left with a story about them — a story that can be useful in its way, navigable, orienting, but that cannot substitute for seeing what’s actually there: the specific, burning, aging, dying particularity of each one.

What the Grounding AI Map does is let you step close enough that the figure starts to dissolve. Not into nothing — into its actual parts. Into the two million conversations researchers have been having about what algorithms can do, where they fail, what problems they were built to solve, and what problems they created instead. The stars don’t disappear. They just stop telling you a story about a hunter.

The view from far away, the one that lets you see AI as one. One solution, one controversy, one story, remains real and sometimes necessary. But it remains a specific view. And there are other views to look from. This map is one of them.

Should we call it deflation? Not because AI gets smaller — it doesn’t — but because the inflated story finally meets the ground.

Consider the constellation as the inflated version. When we talk about Orion, we’re talking about a hunter with a belt and a bow, a figure with mythological weight and narrative built all around him. We forget that there’s no hunter up there. There are stars doing what stars do: burning, dying, drifting slowly through space on their own trajectories, indifferent to the stories we’ve organized around them. The hunter exists because we looked up from exactly here and told a story that made sense from this angle, at this distance, in this moment of human history — to make the sky more readable.

Inflation serves a purpose. It’s a way of organizing complexity so we can talk about it, navigate by it, form an opinion about it. But the story develops its own gravity, its own pull. Soon we’re asking questions about the hunter — what is he doing, what does he want, should we be afraid of him — and we’ve stopped asking about the stars.

Deflation is what happens when you step close enough that the story can’t hold. The hunter dissolves — not into nothing, but into the actual, interconnected parts that were always there. This map deflates AI: not by claiming it isn’t real or doesn’t matter, but by refusing to let it stay singular. Every cluster you see — Smart Wearables, Underwater Sensor Networks, Student Performance Prediction, Secure e-voting — contains papers that mention algorithms, describe specific techniques, and belong, fully, to what we mean when we say AI. Deflation recontextualizes the matter rather than removing it. AI doesn’t vanish. It lands somewhere specific: inside a problem, a field, a community of researchers who were mostly trying to solve something that had nothing to do with AI in the first place, and who found that certain algorithmic approaches turned out to be useful.

The political and social weight doesn’t disappear either. Some of these clusters carry real stakes, real consequences for real people: who gets diagnosed, whose neighborhood floods, what counts as quality, who gets to decide. Deflation doesn’t dismiss urgency or critique. It doesn’t decrease controversiality. If anything, it refines it. It makes critique possible in a different register, because now we can ask: which AI, where, doing what, for whom, solving what problem, and at whose expense?

When AI is no longer one thing with one story, the questions get better. Not what will AI do to us, as if we’re standing still and it’s approaching from outside, but: what is being done here, in this context, through these decisions, with which consequences, and for whom? Paradoxically, deflation makes the thing bigger. The singular story compresses everything into one trajectory. The deflated view opens into multitudes. Two million papers. Hundreds of fields. Thousands of specific, local, deeply situated problems where algorithms turned out to matter.

Walk onto the floor map, and you might find yourself standing on Bean Authentication. There is something funny in realizing that the vast intelligence you’ve been reading about is also, for some reason, concerned with coffee beans. Sit with that reaction for a moment. What is it, exactly? Perhaps a kind of relief. Or recognition. The algorithms are still there — the papers in that cluster describe machine learning techniques, optimization methods, classification models — but they describe them in relation to something concrete, something grown and harvested and tasted: authenticating bean quality, predicting yields, modeling roasting temperatures. You suddenly face the fact that AI, this almost magical thing, is also grounded in something as simple as a cup of coffee.

Seen from inside the clusters, AI means something different than it does from the satellite view. More specific. More entangled with things that smell and taste and grow and flood and break and need to be authenticated or predicted or optimized or understood. More ours, in some sense, because it turns out to live inside the same material world we do.

You’ll fold this back up soon. Crease it along the lines it already knows, tuck it into a bag or a pocket. A condensed galaxy to carry around for a while.

This map won’t answer the big questions people keep asking at dinner tables and on social media, the ones that make AI feel enormous and inevitable. Perhaps that’s not what it’s for. What it offers instead is a different kind of question — the kind that can only be asked from somewhere specific. Not what AI will do to us, but what is happening here, in this cluster, in this field, with these people, working on this problem?

The next time someone says AI is going to change everything, you might remember what it felt like to look at it from close up. To stand on a cluster about flood prediction or bean authentication and see the situated, entangled, deeply material ways that algorithms are already changing things — some of them worth celebrating, some of them worth resisting, most of them worth understanding better than we currently do. The map doesn’t tell you which is which. It helps you find a way to look, and reminds you that the one story dissolves into many once you step close enough.

What becomes possible from here is yours to find.

Matilde Ficozzi, April 2026

AI-Generated Images for Representing Individuals: Navigating the Thin Line Between Care and Bias

2025-01-01T23:59:08+00:00

This research discusses the figurative tensions that arise when using portraits to represent individuals behind a dataset. In the broader effort to communicate European data related to depression, the Kiel Science Communication Network (KielSCN) team attempted to engage a wider audience by combining interactive data graphics with AI-generated images of people. This article examines the project’s decisions and results, reflecting on the reaction from the audience when information design incorporates figurative representations of individuals within the data.

Choreography, Design, and Technology: An Interview with Lins Derry from the metaLAB (at) Harvard

2025-01-01T23:59:07+00:00

In this enlightening interview with Lins Derry, a pioneering researcher at the intersection of dance, design, and technology, we explore the evolution of her groundbreaking work from its conception to its implementation in academia. Derry discusses her journey from professional dancer to leading figure at metaLAB (at) Harvard, emphasizing the integration of choreographic principles into interaction design. Through projects like the choreographic interface and data embodiment, she illustrates the potential of movement as a medium for interpreting and interacting with abstract data. Her work challenges traditional disciplinary boundaries, advocating for a more integrated approach to teaching and research that leverages the expressive power of the body in digital environments. This interview not only highlights Derry’s innovative contributions but also reflects on the broader implications of her work for enhancing sensory and expressive experiences with technology.

Collective Care: VISAP ‘25 Catalog

2025-01-01T23:59:06+00:00

Care is easy to recognize on a personal level, especially when it appears in the small, attentive gestures woven into daily life. We see it when someone nurses a sick friend, tends a garden, or stitches a quilt by hand. Each act is marked by presence, patience, and a quiet commitment expressed through attention. It takes form through deliberate actions that often go unnoticed yet carry enduring meaning. But what does care look like when it scales up across complex systems where the risks are greater, the people more dispersed, and the consequences harder to trace?

This year at VISAP, we ask what it means to approach visualization not only as an academic and professional practice, but as an act of collective care.¹ How can we design visualizations that not only represent but protect, nurture, and respect the environments and communities woven into datasets? What practices emerge when we visualize data with empathy, thoughtfulness, and intention? In a time when data shapes public perception, national policy, and personal identity, centering care in our visual methods is not merely desirable but essential.

In the world of data visualization, “care” is not a term we use often. We tend to speak of clarity, insight, and impact. We use words that suggest objectivity and utility. Yet as datasets expand to encompass our communities and algorithms increasingly shape our societies, visualization itself becomes a cultural and ethical act. It determines how we see one another and the systems surrounding us. In this light, the role of the designer reaches beyond aesthetics or communication, demanding a deeper engagement with social consequence and ethical responsibility. Acknowledging this role means accepting that visual choices can influence narratives, reinforce or challenge bias, and shape collective understanding in enduring ways.

In contemporary digital culture, data no longer stands as a static artifact but as a living archive that holds memory, identity, and collective history. Biometric scans, environmental sensors, and geotagged images, nearly every facet of human life is now captured and transformed into data. Giorgia Lupi reminds us that working with data can reveal more than patterns; it can expose the human connections that shape them. Her approach asks us to see data not as detached or abstract, but as something deeply rooted in the stories, emotions, and lived experiences of individuals and communities.

Within this context, data visualization is not merely a cosmetic tool but a critical process of reinterpretation, contextualization, and communication. It becomes a way to narrate our datafied collective histories and to shape how communities become legible. Professionals working with data act as communicators and storytellers. Through visual, sonic, spatial, or even olfactory forms, they translate abstraction into experience, something we can sense, question, and connect with. In doing so, they turn datasets into living archives and visualizations into spaces for reflection, empathy, and care.

Maria Puig de la Bellacasa describes this orientation as “matters of care,” a call to move beyond surface concern into the dense, affective work of maintenance, repair, and relationality. It invites us to care for our practices as we care for one another, not efficiently, but attentively and critically. Within this framework, care is not sentimental; it is relational, collective, and ethical. It asks us to take responsibility for the data we engage with and to honor the lives, communities, and ecosystems it represents. To visualize with care is to visualize with empathy, to expose environmental harm, surface silenced narratives, reveal shared experiences, and confront the structural biases that persist unseen.

Building on this understanding of visualization as a relational and embodied practice, we imagine a future where data visualization serves as a process of restoration, connection, and long-term social resilience. This vision invites us to treat data as a space for healing, resistance, and belonging. It calls for visual practices that sustain the well-being of both environments and the communities most affected by them. These same values must also shape how we collaborate with emerging technologies, especially as we co-create meaning with algorithmic systems and AI, shaping how data is interpreted, narratives are generated, and decisions are informed by machine learning tools. Such collaborations raise urgent questions about authorship, agency, and ethics: Whose data is used? Whose voices are amplified or silenced? A care-centered approach to AI foregrounds transparency, accountability, and relational design, prioritizing systems that are socially responsible and culturally attuned.

The selected papers presented at VISAP ‘25 embody Collective Care through distinct yet interwoven approaches to visualization. Psychomare translates psychoanalytic theory into an immersive experience, turning nightmares into symbolic data that invite emotional reflection and empathy. In Tides of Memory, digital mourning becomes a spatial monument, preserving collective grief drawn from ephemeral traces on social media. The Knowledge Cosmos reimagines scientific literature as a navigable universe, fostering curiosity and cross-disciplinary discovery as acts of epistemic care. Simulacra Naturae entwines biological computation with generative ecosystems, decentering human agency and emphasizing care for nonhuman intelligences. You Only Have Seven Seconds reframes participatory data as an intimate archive, transforming whispered recollections into shared memory through AI-driven imagery. Finally, Living Library of Trees maps the Arnold Arboretum as a living archive of environmental and curatorial labor, positioning visualization as both analytical method and ecological ethics. Together, these works illuminate care as a methodological, emotional, and ecological principle, reshaping how visualization connects people, data, and the worlds they inhabit.

The selected pictorials at VISAP ‘25 expand the idea of Collective Care through sensory, material, and participatory design. Winds Through Time turns paleoclimate data into a tactile landscape, merging physicalization and visualization to make the ancient dynamics of wind and ice perceptible through touch and play. Rejecting Colonial Practices in Data Storytelling challenges dominant epistemologies in data journalism, advancing decolonial visualization practices grounded in local vernaculars and collective authorship. Kaleidoscope of Thoughts translates cognitive turbulence into an immersive audiovisual environment, weaving multilingual voices and mirrored projections into a shared meditation on emotion, resilience, and empathy. AI-Generated Images for Representing Individuals explores how portraiture in data communication exposes the ethical tensions between empathy and representation, prompting reflection on the affective power of visual design. Balaton Borders reimagines ecological data as ceramic tableware, transforming environmental monitoring into the ritual of shared meals where care becomes an embodied practice. Finally, The Fire We Share reconfigures wildfire data as a living archive, layering ecological, political, and emotional narratives into plant-inspired data forms that evoke repair and remembrance.

The selected artworks extend Collective Care into lived, participatory spaces where data becomes touch, ritual, and relation. Tides of Memory materializes online mourning as walkable monuments that weave personal authorship with collective remembrance. Damaged Leaf Herbarium and Dataset reframes pest-scarred foliage as a cared-for dataset, turning acts of preservation, scanning, and algorithmic labeling into gestures of shared custodianship. Knowledge Cosmos stages discovery as embodied navigation, inviting participants to traverse a universe of seventeen million papers. Simulacra Naturae entwines brain organoid signals with agent ecologies, composing a hybrid environment that shifts attention from human control to multispecies attunement. The Secret Life of Collective Plastic Microfibre Traces transforms microfibre evidence into collective sensing and mapping, linking garments to planetary residue. Empathic Growth connects human GSR data with plant biosignals to prototype cross-species empathy. A Walled City turns personal images into a dense, evolving architecture of memory and relation, while You Only Have Seven Seconds gathers whispered recollections into a cinematic archive of shared remembrance. Polyurethane, cellulose… urea-formaldehyde performs a chemopoetics of fabrication residue, sonifying SEM micrographs to reveal invisible exposures. Distance Unknown weaves currency into a physical visualization of migration costs, linking policy discourse with individual testimony. FeltSight reorients perception from vision to touch through haptic, mixed-reality sensing, while Weaving Water, Interleaving Silence uses water to explore the ethics and limits of affect recognition. Finally, Balaton Borders Tableware brings ecological data to the table, where disruptions in use become cues for reflection and care.

This year’s selection approaches Collective Care as an evolving act of holding memory, tending to the more-than-human world, and creating spaces for empathy through sensory and affective experience. They question how we represent others, cultivate relational practices, and sustain accountability within complex systems. Across these varied approaches, care becomes a shared lens, one that sustains connection between data, beings, and the worlds they represent.

This catalog reissues the curatorial statement of VISAP ‘25 as it accompanies the exhibition during the IEEE VIS Conference in Vienna. In revisiting the text, we recognize how Collective Care continues to resonate, not only through the selected papers, pictorials, and artworks, but also through the shared effort that sustains them. From the labor of students and organizers to the generosity of institutions, this exhibition embodies care as collaboration, attention, and presence. Our gratitude extends to the IEEE VIS conference and organizing committee, to the University of Applied Arts Vienna for hosting, to the Moholy-Nagy University of Art and Design for their support, and to the many contributors whose dedication made VISAP ‘25 possible.

As we gather in Vienna, this year’s theme, Collective Care, reminds us that data visualization is not only a tool for analysis but a gesture of attention, empathy, and shared responsibility. We invite you to engage with this exhibition as an act of care: attentive, relational, and deeply human.

General Chairs
Damla Çay, Moholy-Nagy University of Art and Design
Dario Rodighiero, University of Groningen
Weidi Zhang, Arizona State University

Exhibition Chairs
Martina R. Fröschl, University of Applied Arts Vienna
Peter Mindek, Nanographics GmbH

Design Chair
Beatrice Gobbo, Politecnico di Milano

This is an extended edition of a text that appeared in Nightingale in June 2025: Çay, Damla, Dario Rodighiero, and Weidi Zhang. 2025. “Visualizing as a Form of Collective Care.” Nightingale, June. https://nightingaledvs.com/visualizing-as-a-form-of-collective-care/. ↩

Experiments of Network Literacy for Urban Designers: Bridging Information Design and Spatial Morphology

2025-01-01T23:59:05+00:00

Urban morphology has long been studied through typologies, spatial configurations, and historical change, yet cities are not static artifacts but dynamic environments continually reshaped by people, infrastructures, and politics. This article brings Actor–Network Theory (ANT) into dialogue with Aldo Rossi’s notion of the locus to rethink urban design as both enduring form and relational process. Building on Manuel Lima’s taxonomy, the study develops a methodological workflow that translates street networks into visualizations, pairing embeddings with topographic maps to highlight structural patterns. Applied to a comparative set of cities, the analysis distinguishes three broad morphological tendencies—archetypal, geometrical, and relational—each reflecting different logics of urban organization. The results show how scale and connectivity condition the interpretability of embeddings, revealing both alignments and divergences between cartographic and topological representations. Beyond empirical findings, the article frames network literacy as a meeting ground for design theory, science and technology studies, and information visualization. It concludes by proposing that advancing urban morphology today requires not only new computational tools but also sustained interdisciplinary collaboration across design, urban studies, and data science.

Grounding AI Map: The Consequences of Living with the Trouble of an Irreductionist Map

2025-01-01T23:59:04+00:00

Data visualizations are often seen as tools to make expert knowledge more accessible. Yet, without careful design, they risk simplifying or misrepresenting complex ideas. This case study examines the Grounding AI Map, a 100 m² floor installation at the Danish Technical Museum that visualizes millions of scientific articles on artificial intelligence. Visitors navigate this mapped knowledge physically, aided by automated bots that summarize and interpret content using large language models. These bots serve not just as guides but as mediators, prompting users to question and reinterpret the mapped knowledge through their own perspectives. By inviting interaction and critique, the map becomes a participatory space where knowledge is co-constructed. A quali-quantitative framework assesses this mediation by analyzing app data, observing visitor behavior, and conducting interviews. The study highlights how visualization, when combined with reflexive AI tools, can create more inclusive and situated ways of engaging with scientific knowledge.

« Aucune chose n’est par elle-même, réductible ou irréductible à aucune autre. » — Bruno Latour, Irréductions (1984)

Abstract

Data visualization is emerging as a mechanism to make expert knowledge accessible, bridging the gap between research institutions and the broader public. However, scientists must be cautious, as visualizations can oversimplify or misrepresent information, inadvertently widening this gap (Drucker 2021). While visual abstraction can enhance inclusion, it alone is insufficient; effective mediation of knowledge requires designing encounters that enable audiences to inhabit and explore semantic spaces. In this process, data visualizations function as cognitive prosthetics for viewers, redistributing the epistemic labor of making sense of the world.

The proposed case study exemplifies this approach by mapping AI scientific literature into a tangible and experiential format. Such scientific literature is physicalized at the Danish Technical Museum as a 100 m² floor mat, inviting audiences to explore through walking. The “Grounding AI Map” analyzes millions of scientific articles, transforming them into annotations summarizing the most important topics around artificial intelligence (Munk et al. 2024). Those annotations are then LLM-summarized and embedded in automated bots, which mediate between the representation and the viewers by assessing the nature of knowledge abstraction. Inspired by the concept of counter-AI by Chateauraynaud (2019), in the case study AI methods interrogate their own epistemic structures rather than uncritically amplifying algorithmic outputs. By positioning AI as an object of reflexive critique, this approach aligns with the broader sociological effort to reintroduce multiple perspectives into digital knowledge infrastructures. Extending this logic of critique and engagement, the automated bots are also made accessible via a smartphone app, enabling audiences to interact with the LLM-generated insights, question their framing, and actively reinterpret the mapped knowledge through their perspectives (Jensen et al. 2021).

By fostering exchanges, the automated bots transform the map into a participatory space, where knowledge is co-constructed between visitors (Rodighiero et al. 2022). A quali-quantitative framework evaluates the mediation quality in the case study. Quantitative analysis of the app reveals the audience’s interests, while the qualitative observation of social interactions captures how visitors make sense of the space. Additionally, open-ended interviews provide insight into how visitors interpret, reprocess, and re-appropriate the exhibited knowledge on their terms. The findings highlight the situated nature of visualizing and abstracting knowledge, as the exhibition renders visible the negotiation between the visualization and the public’s perspective. Understood as a hybrid infrastructure that naturalizes scientific knowledge into an undramatic yet multifocal artifact, this experiment suggests how LLMs’ epistemic flexibility could reshape the ways we construct and interpret a shared understanding of the world.

Introduction

Data visualisations often communicate a clear message, condensing complex information into a readable and understandable form. However, in some cases, instead of presenting a conclusion, they invite a different kind of engagement, one where meaning is not given but negotiated. This assumption raises a question: how can audiences engage with a visualisation that has no message to convey? This text adopts a dialogic perspective to investigate the apparatus that turns the visualisation viewer into an active participant in the transfer of knowledge. This approach is explored through the case of the Grounding AI Map, a large-scale data visualisation that transforms scientific literature into an exhibition format, inviting the audience to shape knowledge as they explore it.

The Grounding AI Map is a network visualisation printed on a 100-square-meter floor mat for the Danish Technical Museum in Helsingør. It visualises a dataset extracted from Scopus, comprising two million scientific articles on algorithms, artificial intelligence, and machine learning, published between 1985 and 2024. The map represents this vast corpus using clusters spatially organised by semantic similarity. To support exploration, the clusters are annotated using large language models, which act as research companions of the audience who explores the map through a dedicated app with automatically generated slanted opinions. In this way, AI plays a three-folded role in the project: as a tangible object of observation, a computer-generated mapping algorithm, and a prompt for critical reflection. Rather than conveying a fixed message, the Grounding AI Map encourages interpretation, fostering engagement with AI in its many forms.

This text focuses on how researchers can make their work accessible to the generic public. The Grounding AI Map project addresses this challenge by engaging the audience through visual exploration. Rather than leading viewers toward a single interpretation, it invites them to walk, look at, and reflect. This deeply explorative approach brings to light broader issues that arise when researchers attempt to design visualisations that open up questions rather than close them. Engagement is not only shaped by the visualisation itself but also by its situatedness — in our case, through annotations, an interactive app, physical presence, and the context of a museum space.

Grounded in Actor-Network Theory, this article is structured in four parts. Each section focuses on one element of the project and links it to a key concept from the ANT framework. The first part explores abstraction, describing how patterns emerge from data through spatial organisation. The second part turns to irreduction, examining how annotations preserve complexity rather than simplify. The third discusses mediation, focusing on the app and its role in guiding interpretation. Finally, the fourth part reflects on assemblage, looking at the physical exhibition and its interaction with the museum collection and curatorial choices. Together, these sections show how the Grounding AI Map functions as a layered apparatus, inviting audiences to co-construct meaning rather than receive it.

Map (Abstraction)

Abstraction, etymologically meaning “to drag away, detach, or divert,” is central to the field of information design. In the context of data visualisation, it refers to the process of selecting and representing chosen subjects of study or specific aspects of data, while leaving others aside. The designer has to decide what to show, what to leave out, and how to structure what is left. These choices determine what information becomes visible and what information stays hidden. In this sense, abstraction is not a neutral operation at all: it influences how knowledge is presented and interpreted — it can be called knowledge design (Schnapp 2013). By identifying specific elements and arranging them meaningfully, visualisations reduce the initial data to give the viewers the opportunity to understand something that wouldn’t be accessible otherwise.

This process of abstraction is visible in the field of science mapping, a discipline that aims to describe the science in the making (Latour and Woolgar 1986). Typically in the field, scientific articles are analysed and transformed into visual representations to reveal the structural alliances between scholars and institutions, which are usually not visible in publications themselves. From its origin, to design explanatory visuals, science mapping has relied on established metrics such as mapping co-authorships or bibliographic coupling (Börner et al. 2005; Chen 2017), but more recent approaches use natural language processing to detect text similarities, offering a different way to analyse corpora (González-Márquez et al. 2024; Noichl 2023).

As a result, any science map represents a form of abstraction. This is true not only for geographical maps but also for non-geographical ones — such as network maps, self-organizing maps, or embeddings created through dimensionality reduction — which all share this abstracting function. Unlike more conventional charts, these maps reveal structures through spatial cues like density and distance, making visible features such as clusters, structural gaps, or centre–periphery relations. The Grounding AI Map is one such example, offering a spatial perspective on the organisation of scientific knowledge about artificial intelligence.

Non-geographical maps also have the distinctive ability to prompt questions rather than provide definitive answers. Unlike explanatory visualisations that convey a fixed message, these maps are created by researchers to actively explore complex data — a capability not typically aligned with hypothetico-deductive methods that rely on pre-formulated hypotheses. As Tukey describes (1977, 131–160), exploratory maps expose viewers to new insights, often taking the form of further questions rather than concrete evidence — this interpretative openness depends heavily on the viewer’s subjectivity. Echoing Dewey (1938), Munk et al. (2019, 112) suggest that data-driven inquiries often “question the initial frame,” emphasizing that exploratory visualisations frequently generate more questions than they answer.

The Grounding AI Map is an example of a non-geographical, exploratory visualisation. Created through an embedding model (Cohan et al. 2020) combined with a dimensionality reduction technique using a k-nearest neighbours network (Cover and Hart 1967), the map organises scientific articles on AI based on their semantic similarity. Although the visualisation is technically a network, it does not explicitly depict connections as lines; rather, relationships between documents are indirectly visible through their spatial positioning (Rodighiero and Romele 2022), determined by a force-directed layout algorithm (Jacomy et al. 2014). The resulting clusters are semantic abstractions, groups of documents sharing similar content. Crucially, as all the non-geographical maps, the Grounding AI Map does not aim to communicate a specific message or conclusion. Instead, its design deliberately preserves openness, prompting audiences to explore, question, and engage in dialogue.

Creating the Grounding AI Map involves five clearly identifiable steps. First, a corpus of approximately two million scientific articles is collected from the Scopus database. The query targets documents that include the terms algorithm*, AI, “artificial intelligence”, or “machine learning” in either the title or the abstract. This initial selection criterion aims at capturing “what algorithms are doing in science.” The limitations of this query are acknowledged. Previous iterations of the map, presented to researchers across disciplines, have drawn attention to the partial nature of the query. Despite this, the decision is made to retain a general scope. Expanding the query to include field-specific vocabulary risks over-representing the research team’s existing knowledge while still missing less familiar or emerging uses. As a result, the query is deliberately limited, with the recognition that it yields an incomplete yet broadly representative view of algorithmic discourse in scientific literature. Second, a relational system is established to reveal connections between articles using text analysis and natural language processing techniques. Each article is represented as a node organised by semantic similarity, enabling readers to access individual documents while visualising the overall relational structure of the corpus. Third, an embedding technique is applied, mapping documents spatially according to semantic proximity. Visually, this approach generates clusters that reflect distinct scientific areas, with varying sizes and proximities, thereby supporting audience interpretation of artificial intelligence research.

In the fourth step, the design of the Grounding AI Map focuses on graphic choices that enhance clarity and visual appeal, which is essential for navigating such a large-scale map. Each document appears as a coloured dot, with the colour indicating its year of publication, forming a gradient from red (older) to blue (newer). Hill-shading techniques are employed to make clusters visually prominent, imparting a familiar geographical aesthetic while conveying relative sizes and densities. Additionally, clusters are highlighted with labels generated through an annotation protocol based on large language models. These annotations provide titles and summaries that guide audience interpretation. Finally, the map is physically printed as a large-scale artifact, giving tangible form to the data. This decision introduces challenges related to image resolution, printing materials, format, and transportation. However, transforming the map into a physical, explorable landscape enables the audience to interact with the data directly — walking across it, observing distances, and zooming in or out physically — thus influencing engagement through scale and movement (Figure 1).

Figure 1. Visitors walking on the Grounding AI Map.

To conclude, the concept of abstraction comes back into focus: the Grounding AI Map functions as a mechanism of abstraction for an extensive corpus of scientific literature. However, abstraction inevitably raises questions of intelligibility: how can the complexity of scientific knowledge remain understandable without succumbing to oversimplification? The next section addresses this critical challenge through the concept of irreduction, examining how annotations can preserve complexity while enhancing accessibility.

Annotation (Irreduction)

Creating a map from scientific literature inevitably involves processes of abstraction, which implies some choices of reduction. However, in the case of the Grounding AI Map, the original data points are not simply reduced into clusters. Instead, they become central elements in constructing annotations, which have the role of significantly limiting information loss and amplifying the visualisation. As a result, annotations are added as an interpretative layer to the walking visualisation, without affecting the benefits of the visual abstraction. This reversibility allows the audience to move between the annotation and the visualisation, choosing which level of detail they wish to engage with. By preserving the complexity of the individual data points, the annotations remain open to scrutiny, encouraging the audience to question their appropriateness. In this sense, annotations serve not as definitive answers but as points of entry, keeping the underlying questions alive.

Annotation, broadly understood, is a dynamic practice connecting readers, texts, and contexts through meaningful interaction. Kalir and Garcia (2021) describe annotation as not merely explanatory notes but as a form of active dialogue — a way readers “talk with their texts, to their texts, about and beyond texts.” Rather than existing in isolation, annotation generates intertextual relationships, establishing connections across multiple layers of meaning. It acts as a communicative gesture that provides context, illuminates hidden perspectives, and initiates conversations among readers, authors, and texts. Thus, annotations become integral to shaping knowledge, not by manipulating original content but by enriching it through the inclusion of new voices, interpretations, and engagements.

The annotation process for the Grounding AI Map addresses a critical challenge: the map itself, derived from highly specialised scientific texts, is not immediately relatable or accessible to general audiences. To overcome this, a specific annotation protocol was developed, leveraging artificial intelligence to generate contextual labels that maintain the complexity and richness of the underlying data. Unlike conventional mapping techniques, which often oversimplify information into uniform clusters, this approach strikes a careful balance — it simplifies content sufficiently to support readability without reducing nuanced research insights into generic narratives of technological advancement. Annotations thus serve as points of entry, providing clarity while preserving the map’s capacity to provoke exploration and interpretation (Figure 2).

Figure 2. Detail of annotations on the Grounding AI Map.

Involving AI at the annotation level aligns closely with Francis Chateauraynaud’s (2019) concept of “counter artificial intelligence.” Simply put, this approach represents a counter-program to the predominant vision of AI, as promoted by influential proponents such as OpenAI, who aim at automating and replacing humans in as many contexts as possible. Chateauraynaud states: “Rather than pursuing autonomous machine development, these AI methodologies aim to test hypotheses and inquiry logic. … The conceptual reference point lies in … counter-expertise or counter-power. The objective is to re-situate artificial intelligence within a critical framework, thereby avoiding the risk of contributing to a fully realised form of digital nonsense — disconnected from real-world contexts — by remaining captivated by algorithmic capability” (Chateauraynaud 2024, our translation). In this project, rather than allowing AI to independently simplify the complex landscape of artificial intelligence, AI is used specifically to facilitate — not replace — human interpretation. The AI-generated summaries supplement rather than substitute the original scientific papers. By annotating instead of reducing, human interpretation remains central, retaining the detailed information necessary for critically engaging with or even disagreeing with AI-generated outputs.

While annotations enrich the layer of the map, the experience of interpretation becomes one of the central elements through mediated interactions. This brings us to the role of the app as a tool for elicitation and critical dialogue.

App (Mediation)

At first glance, the vastness of the map contrasts with one of the main goals of data visualisation: offering a representation that allows viewers to access specific knowledge easily. Instead, the Grounding AI Map presents a scale that resists immediate understanding, calling for tools that support navigation rather than deliver answers. This approach reflects a fundamental commitment: to foreground complexity rather than to obscure it through oversimplification.

The exhibition space is designed so that visitors, standing directly on the map, can read the cluster titles near their feet and grasp those within their immediate movement. As the gaze moves outward, however, the readability fades. Titles blend into the broader visual field, merging into the geography shaped by the data. This partial visibility is not a flaw but a deliberate choice: the map’s vastness conveys the scale of the literature it represents and reminds viewers of the impossibility of grasping such complexity in a single glance.

The app (Figure 3) was created as a tool for mediation, offering support for navigating the map and engaging in dialogue with it. The map itself acts as a mediator, not a message. The goal is not comprehension through simplification but exploration through interaction. Once the app is in use, it changes the experience: the user becomes part of an assemblage with the app and the map. This assemblage is where mediation happens.

Figure 3. The “Explore the map” section of the web-app.

One of the central features of the app is the slanted bots (Figure 4), animated textual agents that the users meet when they open a cluster summary. Each bot presents a distinct personality: one reflects an optimistic, progress-oriented reading of the scientific domain, while the other assumes a more sceptical, critical posture. Rather than resolving ambiguity, these bots heighten it, presenting interpretations that are intentionally partial and situated. Their synthetic opinions prompt visitors to confront the instability of knowledge claims, and invite them to form an opinion themselves. In this way, the bots act not as guides but as provocateurs, opening a space for critical engagement rather than offering closure.

Figure 4. When opening a summary, the users are met with the slanted bots.

This added layer encourages negotiation of meaning. The bots offer differing perspectives, making it possible for the viewer to challenge what is presented and reflect on their own stance. Once access to knowledge is facilitated, a process of reinterpretation and re-appropriation begins, where understanding is shaped by dialogue, not by closure.

Yet mediation does not occur in isolation. It is deeply shaped by the physical and institutional context in which the map is encountered. The museum space itself becomes an active participant in the process of meaning-making, framing the interaction between visitor, map, and app. In the final section, we turn to how the broader curatorial conditions and the museum itself modulate this experience, becoming an active ally in shaping the knowledge encounter.

Physicalisation (Alliance)

Data visualisation is often an object that is published on the web, leaving the accessibility open to a vast public. However, when viewers are interacting with a personal computer in their houses or offices, it’s impossible to say how they behave, or even more interestingly, we cannot guide them in using data visualisations, especially when they are complex objects or interfaces. Yet what we can draw from Latour’s legacy is that we can rely on alliances that can help us in fully interacting with the audience.

In that sense, institutions like the Danish Technical Museum can be seen as allied. Museums, as many other institutions, can connect the data visualisation to a specific public by offering a public space in which the interaction takes place. This type of alliance is fundamental when we talk about data physicalisation, as such a controlled environment offers us a two-folded goal: on one hand, it allows us to speak with the public and guide them in the consumption of knowledge; on the other hand, it allows the designers to observe how the audience relates with the data visualisation: the spectator itself covers an active role, being not a passive actor in the knowledge transfer but representing an actor taking part in the whole mechanism of learning publicly (Rodighiero 2018).

At the Danish Technical Museum, visitors are invited to explore the scientific landscape of artificial intelligence in different ways. The Grounding AI Map occupies 100 square meters in one of the museum rooms, inviting visitors to walk on it. A series of posters working as a system of wayfinding (Figure 5) helps visitors to understand the process of design and the ways in which they can interact with the map. One of these techniques relies on the app that visitors can access using their telephone, letting them discuss specific topics represented in the visualisation. The map, the viewers, the system of wayfinding, and the app compose a mechanism of learning through discussion in a very Freire way to transfer knowledge by engaging with dialogue (Freire [1970] 2000).

Figure 5. Posters from the exhibition that explain the design and methodology behind the Grounding AI Map.

Differently from previous experiments (Rigal and Rodighiero 2015; Rodighiero 2021; Rodighiero et al. 2022), the Danish Technical Museum also offered the opportunity to contextualise the map within their collection, developing a further level of connectivity in the exhibition space. Here, the exhibition is built around the map, both conceptually and physically. The map precedes the exhibition, as a self-contained visualisation of scientific discourse around AI and algorithms. It doesn’t require additional framing to be meaningful. Instead, the exhibition develops in response to it, using the map as a foundation to create a narrative that resonates with the museum context. This relationship is made tangible in the space: the map covers the entire floor, occupying 100 square meters, inviting visitors to walk across it. Their movement becomes a form of engagement, positioning them inside the knowledge landscape that the map represents.

Around this area, the exhibition grows through displays that present artifacts from the museum’s collection (Figure 6). These objects trace how specific practices and technologies have evolved from analog to algorithmic forms and applications. These thematic pairings were developed collaboratively with the museum’s curators. Drawing on their deep understanding of the collection, and our familiarity with the map’s structure, we selected regions of the map and anchored them with physical objects that would be representative of the research field and relatable for the audience. The curatorial strategy here is not simply illustrative, it is translational. This translation helps bridge the distance between abstract terminology and everyday understanding. Many cluster labels on the map use academic or technical language, terms that may not be intuitive to a general audience. When placed alongside a physical object that embodies the function or context of those terms, the map becomes easier to navigate. Objects provide a point of entry, offering a way for visitors to recognise, relate, and begin to interpret a field they might otherwise find opaque.

Figure 6. One of the displays placed around the Grounding AI Map, showing artifacts related to finance.

This translation is not unidirectional. It emerges from an ongoing process of negotiation between curators and researchers. The collaboration with the museum was shaped by a shared interest in experimentation. There is an early recognition that our project might not naturally align with the museum’s typical audience, yet both sides engage in a process of negotiation and adaptation. Early conversations acknowledged the potential disconnect between the map’s abstract, data-intensive content and the expectations of the museum’s general audience. What followed was a process of mutual adaptation: we brought a novel way of visualising and materialising algorithmic culture, while the curators contributed their deep expertise in public engagement. This negotiation hinged on an understanding of the museum’s visitors. The curators emphasised clarity, accessibility, and emotional resonance as guiding principles. One persona repeatedly surfaced in our conversations: the grandparent and grandchild pair. This intergenerational duo embodies the museum’s aim to provoke dialogue across age groups, often through nostalgia-driven storytelling. Exhibitions are crafted to let grandparents share memories and contexts while sparking curiosity in younger visitors. This ethos shapes many of the curatorial decisions in the museum and became central in shaping how the Grounding AI Map was presented. Even as our project introduces abstract and data-intensive content, we work to anchor it in recognizable material culture. Objects from the museum’s collection were selected to anchor abstract terms in tangible artefacts, making unfamiliar concepts accessible without diluting their complexity.

Conclusion

What emerges from the Grounding AI project is not just an exhibition, but an assemblage. The map, the artifacts, the app, the posters, and the viewers form a network of actors, each shaping how knowledge is encountered. The museum, in this configuration, is not a neutral host, it is an active ally. It provides the conditions for public learning to occur: a space of slowed-down attention, guided discovery, and embodied dialogue. This is not a dissemination of knowledge from expert to layperson, but a situated encounter with the complexity of algorithmic culture, made possible through a carefully designed infrastructure of alliance.

Crucially, this infrastructure does not aim to resolve the complexity it presents. The dialogue it enables is not a problem to be solved or a message to be decoded. It is something to be nurtured, kept alive and in motion. The inclusion of museum artifacts plays a key role here: rather than closing the gap between abstraction and experience, they hold it open. Each object adds texture and context, not to simplify the discourse, but to keep it porous, multiple, and accessible from many vantage points. This refusal to land on a single takeaway is not a lack, but rather a deliberate commitment to epistemic generosity. The exhibition invites visitors not to conclude, but to continue.

References

Börner, Katy, Luca Dall’Asta, Weimao Ke, and Alessandro Vespignani. 2005. “Studying the Emerging Global Brain: Analyzing and Visualizing the Impact of Co-Authorship Teams.” Complexity 10 (4): 57–67. https://doi.org/10.1002/cplx.20078
Chateauraynaud, Francis. 2019. “Petit traité de contre-intelligence artificielle. Retour sociologique sur des expérimentations numériques.” Zilsel 5 (1): 174–195. https://doi.org/10.3917/zil.005.0174
———. 2024. “La sociologie, le pragmatisme et la contre-intelligence artificielle. Un regard critique sur les enjeux numériques après le tournant des IA génératives.” Horizon.
Chen, Chaomei. 2017. “Science Mapping: A Systematic Review of the Literature.” Journal of Data and Information Science 2 (2): 1–40. https://doi.org/10.1515/jdis-2017-0006
Cohan, Arman, Sergey Feldman, Iz Beltagy, Doug Downey, and Daniel S. Weld. 2020. “SPECTER: Document-Level Representation Learning Using Citation-Informed Transformers.” arXiv:2004.07180 [cs]. https://arxiv.org/abs/2004.07180
Cover, Thomas, and Peter Hart. 1967. “Nearest Neighbor Pattern Classification.” IEEE Transactions on Information Theory 13 (1): 21–27. https://doi.org/10.1109/tit.1967.1053964
Dewey, John. 1938. Logic: The Theory of Inquiry. New York: Holt.
Drucker, Johanna. 2021. “Visualization.” In Uncertain Archives, edited by Nanna Bonde Thylstrup, Daniela Agostinho, Annie Ring, Catherine D’Ignazio, and Kristin Veel, 561–568. Cambridge, MA: MIT Press. https://doi.org/10.7551/mitpress/12236.003.006
Freire, Paulo. (1970) 2000. Pedagogy of the Oppressed. Translated by Myra Bergman Ramos. 30th anniversary ed. New York: Continuum.
González-Márquez, Rita, Luca Schmidt, Benjamin M. Schmidt, Philipp Berens, and Dmitry Kobak. 2024. “The Landscape of Biomedical Research.” Patterns 5 (6): 100968. https://doi.org/10.1016/j.patter.2024.100968
Jacomy, Mathieu, Tommaso Venturini, Sébastien Heymann, and Mathieu Bastian. 2014. “ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software.” PLoS ONE 9 (6): e98679. https://doi.org/10.1371/journal.pone.0098679
Jensen, Torben Elgaard, Andreas Birkbak, Anders Koed Madsen, and Anders Kristian Munk. 2021. “Participatory Data Design: Acting in a Digital World.” In Making & Doing: Activating STS through Knowledge Expression and Travel, edited by Gary Downey and Teun Zuiderent-Jerak, 117–136. Cambridge, MA: MIT Press.
Kalir, Remi, and Antero Garcia. 2021. Annotation. Cambridge, MA: MIT Press.
Latour, Bruno. 1984. Les Microbes: Guerre et paix, suivi de Irréductions. Paris: Métailié.
Latour, Bruno, and Steve Woolgar. 1986. Laboratory Life: The Construction of Scientific Facts. Princeton: Princeton University Press.
Munk, Anders Kristian, Anders Koed Madsen, and Mathieu Jacomy. 2019. “Thinking Through the Databody: Sprints as Experimental Situations.” In Designs for Experimentation and Inquiry: Approaching Learning and Knowing in Digital Transformation, edited by Åsa Mäkitalo, Timothy Nicewonger, and Mark Elam, 110–128. London: Routledge. https://doi.org/10.4324/9780429489839
Munk, Anders Kristian, Mathieu Jacomy, Matilde Ficozzi, and Torben Elgaard Jensen. 2024. “Beyond Artificial Intelligence Controversies: What Are Algorithms Doing in the Scientific Literature?” Big Data & Society 11 (3): 20539517241255107. https://doi.org/10.1177/20539517241255107
Noichl, Mario. 2023. “How Localized Are Computational Templates? A Machine Learning Approach.” Synthese 201 (3): 107. https://doi.org/10.1007/s11229-023-04057-x
Rigal, Alexandre, and Dario Rodighiero. 2015. “Trajectoire d’une représentation cartographique en réseau.” Cartes & Géomatique 225: 33–41.
Rodighiero, Dario. 2018. “Printing Walkable Visualizations.” In Proceedings of the 5th Biennial Research Transdisciplinary Imaging Conference, 58–73. https://doi.org/10.6084/m9.figshare.6104693.v2
———. 2021. Mapping Affinities: Democratizing Data Visualization. With Jeffrey T. Schnapp. Geneva: Métis Presses. https://doi.org/10.37866/0563-99-9
Rodighiero, Dario, and Alberto Romele. 2022. “Reading Network Diagrams by Using Contour Lines and Word Clouds.” In Graphs and Networks in the Humanities. https://doi.org/10.31235/osf.io/3g7bt
Rodighiero, Dario, Eveline Wandl-Vogt, Elian Carsenat, Jules Döring, Oliver Elias, Michaela Fragner, and Stepha Farkashazy. 2022. “Immersive Architectures for Visual Data Literacy.” Information Design Journal 27 (3): 295–308. https://doi.org/10.1075/idj.22016.rod
Schnapp, Jeffrey T. 2013. “Knowledge Design: Incubating New Knowledge Forms/Genres/Spaces in the Laboratory of the Digital Humanities.” Herrenhausen Conference, December 5. https://d-nb.info/1104703475/34
Tukey, John W. 1977. Exploratory Data Analysis. Reading, MA: Addison-Wesley.

Living Library of Trees: Mapping Knowledge Ecology in Arnold Arboretum

2025-01-01T23:59:03+00:00

As biodiversity loss and climate change accelerate, botanical gardens serve as vital infrastructures for research, education, and conservation. This project focuses on the Arnold Arboretum of Harvard University, a 281-acre living museum founded in 1872 in Boston. Drawing on more than a century of curatorial data, the research combines historical analysis with computational methods to visualize the biographies of plants and people. The resulting platform reveals patterns of care and scientific observations, along with the collective dimensions embedded in botanical data. Using techniques from artificial intelligence, geospatial mapping, and information design, the project frames the arboretum as a system of shared agency—an active archive of more-than-human affinities that records the layered memory of curatorial labor, the situated nature of knowledge production, and the potential of design to bridge archival record and future care.

Moving Pictures of Thought: Extracting Visual Knowledge in Charles S. Peirce’s Manuscripts with Vision-Language Models

2025-01-01T23:59:02+00:00

Diagrams are crucial yet underexplored tools in many disciplines, demonstrating the close connection between visual representation and scholarly reasoning. However, their iconic form poses obstacles to visual studies, intermedial analysis, and text-based digital workflows. In particular, Charles S. Peirce consistently advocated the use of diagrams as essential for reasoning and explanation. His manuscripts, often combining textual content with complex visual artifacts, provide a challenging case for studying documents involving heterogeneous materials. In this preliminary study, we investigate whether Visual Language Models (VLMs) can effectively help us identify and interpret such hybrid pages in context. First, we propose a workflow that (i) segments manuscript page layouts, (ii) reconnects each segment to IIIF-compliant annotations, and (iii) submits fragments containing diagrams to a VLM. In addition, by adopting Peirce’s semiotic framework, we designed prompts to extract key knowledge about diagrams and produce concise captions. Finally, we integrated these captions into knowledge graphs, enabling structured representations of diagrammatic content within composite sources.

Network Literacy: How to Understand, Design, and Read a Visual Relational Model

2025-01-01T23:59:01+00:00

Models help us navigate the complexity of social life, offering simplified structures that make invisible dynamics legible. Networks stand out for their ability to represent relations directly: nodes and links reduce society to actors and their connections, exposing patterns that often remain hidden in linear accounts. Since the eighteenth century, networks have evolved from mathematical curiosities to essential tools across disciplines. Early sociograms revealed classroom friendships, sociological diagrams exposed social reproduction and inequality, and computational studies now map everything from recipes to scientific collaborations. With their visual grammar, networks invite comparison, clustering, and interpretation across diverse domains. Yet their ubiquity also introduces risks: layouts may be mistaken for objective spaces, central nodes assumed to be more important, and dense graphs admired more for aesthetics than insight. To address these challenges, a new form of literacy is required. Network literacy can be defined as the ability to understand, design, and read visual relational models, combining conceptual knowledge of complex systems with practical skills of visualization and critical interpretation. This paper develops the notion of network literacy as a civic and professional competency, bridging traditions of data literacy and visual literacy. It traces the history of networks from their mathematical and sociological origins to their integration into digital media and design, showing how they reconfigure the codex into a relational mode of reading. It then explores three dimensions: design choices that shape meaning, spatial thinking that guides interpretation, and experimental projects that turn visualization into performative practice. By situating networks at the intersection of information design, critical inquiry, and cultural practice, the paper argues that cultivating network literacy is essential for engaging responsibly with the relational fabric of contemporary knowledge.

Toward Network Literacy

The notion of network literacy arises from the recognition that networks have become one of the most pervasive ways of modeling and visualizing relational complexity in contemporary culture (Lima 2011). To speak of literacy is to frame networks not only as analytical tools but also as cultural artifacts—forms of writing that require both technical competence and interpretive awareness. Just as visual literacy equips readers to decode images and data literacy provides the skills to analyze statistics, network literacy is the ability to understand, design, and read relational models. It involves learning how networks are constructed, recognizing the role of design in shaping their meaning, and developing critical sensibilities for what they reveal and conceal. At its core, network literacy can be defined as the capacity to navigate networks as both designers and interpreters. It is practical, because it demands fluency with the techniques and tools that translate societal dimensions into nodes and links, but it is also cultural, because it acknowledges that networks are not neutral mirrors of reality. They are constructed artifacts that encode decisions, priorities, and assumptions. A network map may chart friendships in a school, citations in a scholarly field, or images in a museum collection, yet in each case the network expresses a worldview: what is considered a node, what counts as a link, and which properties are emphasized. To be literate in networks means being able to decode these decisions, to ask how the model was built, and to interpret its visual conventions with a critical eye.

The notion of network literacy can be better understood if we situate it between data literacy and visual literacy. Catherine D’Ignazio has argued for creative data literacy as a bridge between technical skills and critical reflection, showing how learners must not only acquire the ability to read and analyze data but also use it to question power relations and work toward emancipation (D’Ignazio 2017). Jonathan Gray, Carolin Gerlitz, and Liliana Bounegru have extended this perspective with the idea of data infrastructure literacy, highlighting how infrastructures for producing and circulating data must themselves be interrogated and reimagined (Gray, Gerlitz, and Bounegru 2018). On another plane, Donis A. Dondis framed visual literacy as the universal capacity to compose, interpret, and share meaning through visual forms, a competency as fundamental as verbal literacy (Dondis [1973] 1975). Network literacy stands at the intersection of these two traditions. Like data literacy, it calls for technical and critical awareness of how information is structured; like visual literacy, it emphasizes interpretation, communication, and the social power of images. Yet it is distinct in focusing on the relational model itself: the way nodes and links encode meaning, how design decisions shape interpretation, and how audiences navigate networks as cultural artifacts. In this sense, network literacy extends both literacies, proposing that in an age where knowledge is increasingly organized relationally, we need the skills not only to read data and images, but also to engage with the visual grammars of connection.

The call for literacy becomes urgent when we observe how networks permeate contemporary life. Journalists and artists such as Mark Lombardi have drawn complex diagrams to trace political alliances and financial flows, making visible the hidden ties that shape global power (Hobbs 2004). Museums and archives increasingly adopt network visualizations to reveal the breadth of their collections, moving beyond catalogues to synoptic views that expose unexpected relationships (Windhager et al. 2020). Scientific research too relies on networks, where citation maps and co-authorship graphs communicate relations between papers, authors, and fields, enabling researchers to chart the evolution of knowledge (Noichl 2019). These proliferations demonstrate that networks have moved from specialist instruments to everyday cultural forms. Yet their ubiquity also creates risks. Many viewers read networks as if they were transparent representations, mistaking layout proximity for inherent importance or assuming that central nodes are objectively more influential. Without a literacy that combines technical fluency and critical interpretation, audiences risk being misled by the very tools meant to clarify complexity. Network literacy, in this sense, is an essential civic skill, enabling designers and publics alike to make sense of the increasingly relational fabric of information.

The theme of this journal issue invites us to rethink the codex form. Traditionally, the codex organizes knowledge sequentially, page by page, producing an ordered and linear narrative (Eco 2009). Networks challenge this paradigm. They propose instead a relational codex, where meaning is not found by turning the next page but by following a link, traversing clusters, and exploring proximities. In a network, knowledge is structured spatially and associatively: it is less a path to be walked than a landscape to be navigated. This shift is more than metaphorical. Just as the codex once transformed cultural memory by enabling indexing, cross-referencing, and systematic reading, networks now reconfigure how we encounter information. They allow multiple narratives to unfold simultaneously, depending on where the reader chooses to enter or what connections they follow. In this sense, networks do not abolish the codex but extend it into a form attuned to complexity, multiplicity, and non-linearity. A striking example is Bruno Latour’s AIME Project (Latour 2013), where paragraphs, references, and keywords were woven into digital interfaces that turned a book into a relational environment, foregrounding the interconnected elements of inquiry (Ricci et al. 2017). Cultivating network literacy thus means learning to read this new codex: to treat nodes and links as pages and paragraphs, to interpret clusters as chapters, and to understand that the act of reading is an act of navigation.

Design plays a decisive role in this new literacy. Layout algorithms, color palettes, interactive features—all these elements guide how networks are perceived and interpreted. A force-directed layout may bring similar nodes into visible clusters, while a geographic projection anchors nodes to physical space. Colors may signify communities or categories, while interactivity allows the viewer to filter or zoom. None of these choices are neutral. Each frames the data, emphasizes some patterns, and suppresses others. This is why network literacy requires critical awareness alongside design competence. Designers must acknowledge that visual decisions are epistemic acts: they shape the knowledge that the network communicates. Audiences must learn to ask what assumptions underpin the visualization: why certain data are included or excluded, why centrality appears to be of equal importance, or why a cluster looks more coherent than it really is. A literate approach recognizes that networks can seduce with aesthetic appeal, mislead through algorithmic bias, or oversimplify complex realities. By bringing design and critical interpretation together, network literacy cultivates a double vision: the ability to construct networks responsibly and the ability to question them insightfully. It teaches us to see networks not just as technical diagrams but as rhetorical devices, persuasive in their form as much as in their content.

Building on the historical trajectory traced in History of Networks, this framework sets the stage for the rest of the paper, which explores network literacy in practice. Network Design examines how choices in data preparation, layouts, and tools shape what networks communicate, revealing how meaning is encoded through design. Spatial Thinking turns to interpretation, considering how networks are read as spaces and how conventions such as centrality influence understanding. Surprise Machines offers a case study of experimental visualization, showing how networks can become performative, inviting exploration and critical reflection. Taken together, these sections present network literacy as a constellation of practices—designing, reading, and questioning—that make it possible to navigate complexity. Just as the codex once transformed cultural literacy, networks today call for a literacy of their own, teaching us to read the world not line by line but link by link.

History of Networks

The birth of networks as a model can be traced back to a puzzle of 18th-century Königsberg: was it possible to find a route through the city that crossed each of the seven bridges exactly once without repeating any? When the mathematician Leonhard Euler solved this problem, thinking shifted from physical geography to a new way of seeing connections, establishing what he famously named the “geometry of position” (Sachs, Stiebitz, and Wilson 1988; Euler 1953). His solution showed that what mattered was not the actual distance but how the bridges connected different parts of the city. This discovery laid the groundwork for topology and graph theory, revealing that objects and their relations could form patterns independent of their geographical locations (Shields 2012). In this sense, Euler offered a way to understand structures through their connectivity, moving beyond static maps to dynamic systems of links. His work inspired later thinkers to apply network models to social and cultural questions, from tracing school friendships to exploring power structures. It marked the start of seeing networks as powerful tools for grasping the complexity of the world around us.

If Euler was thinking mathematically, Jacob Moreno was the first to fully recognize the power of networks by giving them their modern visual form. In the 1930s, his diagrams introduced a way to see the hidden structures of social relationships, carefully detailed in Who Shall Survive? (Moreno 1934). Moreno and his collaborator Helen Hall Jennings asked students to name classmates they liked sitting next to. They then mapped these preferences as diagrams, laying the foundation for modern social network analysis (Scott 2000). These sociograms showed how social norms could be visualized, making abstract ties visible and readable. As Moreno himself put it, if we could chart a whole city or nation, we would uncover a vast “solar system of intangible structures” that powerfully influence behavior, just like gravity affects bodies in space (Venturini, Munk, and Jacomy 2019). Moreno argued that until we see these structures, we work blindly to solve social problems. These early sociograms were not yet called “networks,” but they marked the first formal analysis of social relations through diagrams, paving the way for modern network visualization.

A few decades later, Pierre Bourdieu distinguished himself by integrating computational methods into societal studies, moving beyond purely descriptive or qualitative approaches. In Distinction (Bourdieu [1979] 1984), he demonstrated how social reproduction operates through everyday practices of taste, showing that cultural preferences are structured by the distribution of economic, social, and cultural capital. To capture these dynamics empirically, he pioneered the use of correspondence analysis on interview data, enabling the visualization of social layers and the relationships between groups and their lifestyles (Blasius and Schmitz 2014). These diagrams revealed how symbolic boundaries of taste and distinction mapped onto structural inequalities, offering a visual account of stratification that complemented his broader sociological theory. He was also one of the first sociologists to interpret network visualizations, showing how visual forms could expose patterns of imitation and differentiation (Romele and Rodighiero 2020). Building on this legacy, contemporary scholars have argued that network visualization offers epistemic evidence, enabling interpretations of social phenomena that move beyond fixed categories and traditional frameworks (Manovich 2018). These developments highlight how computational networks provide empirical evidence that deepens our understanding of complex social dimensions.

From that moment on, networks developed in parallel with the rise of computational power. Scholars began tackling increasingly large and complex datasets, shifting from small-scale studies to ambitious analyses that demanded advanced computational resources. One influential group of scholars in the Boston area published a manifesto calling for the development of computational social science, setting an agenda that emphasized the integration of massive data, advanced algorithms, and network models (Lazer et al. 2009). Over the years, the physicist Albert-László Barabási, one of the authors of the Boston manifesto, gained significant visibility by popularizing the study of complex networks and making them accessible to a broad audience through his research and publications (Barabási 2002; Barabási et al. 2020). Meanwhile, in Europe, the philosopher Bruno Latour was invited to create a new laboratory at Sciences Po in Paris, the médialab, which combined sociology and computer science, using design as the connecting element between these disciplines (Boullier 2018; Venturini et al. 2017). Latour’s contribution here built on his broader intellectual project: Actor-Network Theory, which redefined society as networks of heterogeneous actants whose relations are constantly negotiated (Latour 2005). As Harman (2009) observes, Latour’s approach was not only methodological but ontological, treating networks as a way to grasp how reality itself is assembled. This orientation gave the médialab its distinctive character, positioning networks as both conceptual and practical tools for exploring social complexity through digital and visual means.

Once networks began circulating in public discourse and media, their visualization became a key mode of interpretation. Design, as Löwgren and Stolterman (2004) argue, makes complex systems legible by highlighting patterns and relations. Manuel Lima (2011) further shows how visual metaphors and aesthetics shape comprehension. What originated in mathematics and sociology now also lives as visual culture, where design plays a central role in how networks are produced and understood.

Figure 1. Jacob L. Moreno's original sociogram from Who Shall Survive?, visualizing interpersonal choices in a group to reveal attractions, repulsions, and the emergent social structure. This foundational diagram demonstrates how sociometry uncovers hidden dynamics of affiliation and exclusion within communities (Moreno 1934).

Network Design

In Visual Complexity, Manuel Lima notes that “an information-visualization designer organizes data to help users see the patterns” (Lima 2011, 12), implying that design choices define what networks communicate. Network design begins with a research question, followed by an empirical synthesis that translates a societal dimension into a relational structure. Designers represent objects as nodes and relationships as links, focusing on the parts of society they want to highlight—this means choosing what to show and what to leave out. Once objects are linked, metrics can define whether those connections carry weights expressing the intensity of relationships—stronger weights lead to stronger attractions in network layouts. Moreno’s sociograms, for example, were created through interviews asking questions like “Whom do you like to sit next to?” (Moreno 1934). While these early diagrams were drawn by hand, today data often take the form of spreadsheets, databases, or JSON files. The transition from manual sociograms to digital mapping, and how it reshapes social research, is explored in Noortje Marres’ Digital Sociology: The Reinvention of Social Research (Marres 2017), which highlights how contemporary means have transformed societal analysis.

Once data are prepared, the layout becomes the next decisive step in network design. Layouts define the visual arrangement of nodes and edges, shaping the spatial configuration. Force-directed layouts, for example, highlight specific aspects of the data by revealing clusters, hierarchies, or flows (Jacomy et al. 2014). These algorithms are central to network analysis: by processing the attractive forces between nodes, they create spatial arrangements where closer positions indicate stronger relationships, helping viewers identify clusters and connections. As Venturini, Jacomy, and Jensen (2021) emphasize, layouts introduce an inherent ambiguity: they do not map an objective space but construct one where relationships become visible through the logic of design. Yet this ambiguity is not a limitation but a feature, offering a subjective perspective to explore complex social structures visually. By embracing this subjectivity, thoughtful layout design reveals hidden patterns, clarifies complexity, and makes networks more engaging—far from being a neutral technical step, it becomes a fundamental design decision guiding how audiences interpret the data.

Layouts are often generated through programming, but to make network literacy more accessible, there are digital tools that do not require coding skills. Software like Gephi offers a friendly environment for manipulating network data, combining interactive exploration with customizable views (Bastian, Heymann, and Jacomy 2009). Gephi allows users to import external data and apply a layout by adjusting its parameters in real time, helping designers make network configurations more legible. It also offers intuitive tools to refine appearance, allowing users to adjust details like node spacing, edge opacity, and label styles for clearer and more compelling visualizations (Grandjean 2015). Fine-tuning elements such as preventing node overlap, adding curved edges, and choosing effective color palettes transforms dense graphs into more readable and engaging network maps. These digital tools not only improve clarity but also support a deeper understanding of network dynamics by allowing designers to iteratively adjust how data are presented. By combining analytical metrics with interactive visualization, tools like Gephi turn networks from static diagrams into exploratory interfaces that guide users through complex relational data.

When it comes to publishing, web frameworks like D3.js open new possibilities for creating engaging network visualizations. The D3.js library in particular enables designers to build customized visuals for the browser, binding data to document elements and rendering them dynamically (Bostock, Ogievetsky, and Heer 2011). By supporting scalable vector graphics and integrating seamlessly with JavaScript, HTML, and CSS, the library allows designers to tailor every aspect of a visualization for web environments, optimizing readability and interactivity. Michael Bostock, who introduced information design at The New York Times, created D3.js to give designers more control over web-based visualizations and later developed Observable, a platform for building and sharing interactive notebooks (Bostock 2017). Observable’s reactive programming environment automatically updates visuals as data or code change, making it easier to experiment, communicate insights, and create dynamic explanations. Together, D3.js and Observable can transform complex networks into accessible and engaging web experiences, enabling designers to turn intricate relational data into clear, explorable stories across devices and browsers.

In a more technical arena, dimensionality reduction techniques like t-SNE (van der Maaten and Hinton 2008) and UMAP (McInnes, Healy, and Melville 2018) enable designers to simplify high-dimensional data into readable, lower-dimensional layouts that reveal hidden patterns. By projecting complex datasets into two or three dimensions, these methods make it possible to visualize similarities and groupings among data points that would otherwise remain obscured in high-dimensional space. t-SNE excels at preserving local relationships, capturing subtle clusters in data, while UMAP balances local detail with a better representation of global structures, offering faster performance on large datasets. These algorithms help designers visualize much larger datasets, ensuring that visualizations remain faithful to key relationships. By connecting advanced data analysis with effective visual communication, dimensionality-reduction techniques make it possible to uncover insights in collections of data far larger than those manageable by traditional network layouts (Rodighiero, Wandl-Vogt, and Carsenat 2022; Rodighiero et al. 2022; Rodighiero and Daniélou 2023; Rodighiero, Rivière, and Kenderdine 2024).

As these design choices and methods show, every step in network visualization leads naturally to considering how space shapes meaning. Spatial arrangements guide how viewers interpret network structures, turning tabular data into visual patterns. Understanding networks as designed spaces prepares us to explore how spatial thinking influences perception, revealing not just what networks contain but how they communicate insights through their spatial organization.

Figure 2. The Grounded AI Map, a large-scale network visualization translating two million scientific articles on artificial intelligence into a walkable 100 m² installation. Clusters are annotated using large language models, enabling visitors to explore and question algorithmic knowledge through interactive bots and a dedicated app (Ficozzi et al. 2025).

Spatial Thinking

In addition to design, reading networks is the other key aspect of network literacy: it involves developing spatial thinking to understand how networks present information. As highlighted in recent cartographic theory, recognizing the interplay of topography and topology helps reveal how network maps construct spaces that resonate with contemporary spatialities—spaces defined more by relational patterns that shape meaning in the social world (Lévy, Maitre, and Romany 2016). Revisiting Euler’s Königsberg bridges through the lens of cultural topology shows that networks are not just logical structures but dynamic spaces where relationships, historical contexts, and social forces intertwine, demonstrating how spatial arrangements can carry cultural meaning and reflect shifting power and identities (Shields 2012).

Beyond theoretical perspectives, mathematical metrics provide essential tools for reading and interpreting networks, including density, degree, and modularity (Grandjean 2021). Density measures how many connections exist relative to the maximum possible, giving a sense of how tightly knit a network is. Degree, the simplest centrality measure, shows how many connections each node has, highlighting the most locally active or connected elements. Modularity detects communities by identifying regions of the network where nodes are more densely linked to each other than to the rest of the graph, uncovering clusters that may correspond to visual interpretation. Together, these metrics transform complex visual patterns into quantifiable insights, enriching understanding and supporting nuanced interpretations.

Centrality is a particularly influential spatial convention in network visualization, shaping how viewers perceive importance or influence. According to Grandjean (2021), centrality metrics such as degree, closeness, and betweenness each capture different aspects of a node’s role in the network—whether it is the most connected, the quickest to reach others, or the best positioned to bridge separate groups. Yet when nodes appear near the center of a visualization, they are often read as more important, even if their placement stems from algorithmic layouts rather than objective hierarchies. In the Affinity Map, for example, individuals displayed at the center felt more valued, as if centrality signaled prominence, even though the map reflected affinities rather than formal status (Rodighiero 2021). This shows how centrality in network maps may be interpretative, influencing how people see themselves and others within a network.

Interactivity further enhances readability, turning static visuals into dynamic tools for exploration. As Grandjean (2021) emphasizes, interactive functions such as selecting nodes, filtering edges, and zooming into dense areas allow users to manipulate the network directly, uncovering structures or relationships that might remain hidden in a static layout. This hands-on engagement supports active learning, echoing Piaget’s idea that understanding emerges through three-dimensional manipulation. By enabling viewers to adjust perspectives or highlight specific features, interactive network visualizations foster deeper, more personal comprehension of complex data, transforming networks into spaces for inquiry and discovery rather than passive observation.

Reading networks spatially is much like reading a city map: just as a street map helps us navigate neighborhoods or landmarks, spatial thinking in networks allows us to explore paths, proximities, and boundaries among nodes. As Latour (2013) argues, networks do not simply represent relationships; they create spaces of interpretation where meaning emerges through connections and distances. By zooming into clusters, we can see why certain nodes are close; by zooming out, we can ask why clusters appear together or apart—each perspective offering a different narrative. This reading is inherently subjective: though general patterns may be shared, each exploration reflects individual questions, interests, and timing. Such personal, performative engagement makes reading a network an interpretative act, but it also brings biases—our motivations shape what we see and how we understand the network’s spatial story.

To illustrate these ideas in practice, the next case study, Surprise Machines, explores how experimental network visualizations engage spatial thinking and challenge our expectations.

Figure 3. Visualization by Martin Grandjean showing the network of influence among 4,300 philosophers in the English Wikipedia, illustrating how encyclopedic knowledge structures can reveal or obscure intellectual traditions (Grandjean 2014).

Surprise Machines

Surprise Machines (Rodighiero et al. 2022) is a critical visualization project developed for the Curatorial A(i)gents exhibition at the Harvard Art Museums, designed to reimagine how visitors engage with large digital collections. It addresses the aesthetics of networks by transforming over 200,000 images into a dynamic visual interface that reveals the often-invisible scale of the museum’s archives. Rather than simply presenting data, the project turns network visualization into a performative, immersive experience, encouraging visitors to question what it means to see and understand a collection in its entirety. By mapping images into a two-dimensional space based on visual similarity, Surprise Machines blends aesthetic choices with data-driven algorithms, creating an interface that is both scientifically rigorous and artistically evocative. This hybrid approach situates the project at the intersection of design, technology, and museology, expanding what network aesthetics can mean beyond static graphs toward interactive, embodied, and surprising encounters with cultural data.

The conceptual foundation of Surprise Machines is to challenge traditional expectations of museum displays by using networks to generate ever-changing visual landscapes. Inspired by Alan Turing’s notion of machines producing surprise, the project embraces unpredictability as a core design principle. Instead of fixed exhibitions curated by humans, it offers dynamic arrangements where visitors navigate clusters of related images, encouraging exploration and serendipitous discovery. By organizing images through dimensionality-reduction algorithms like UMAP (McInnes, Healy, and Melville 2018), the system generates patterns that defy linear storytelling, shifting focus from individual artworks to relationships across the collection. This challenges the conventional view of museums as static repositories of knowledge and instead proposes them as fluid spaces for personal interpretation. By intentionally embracing ambiguity and surprise, the project invites viewers to reflect on how both algorithms and human choices shape what is seen, offering a new paradigm for museum engagement rooted in networked thinking.

At its core, Surprise Machines relies on a sophisticated technical process that combines layout algorithms with deliberate visual rules to create emergent patterns. Using tools like PixPlot (Duhaime [2017] 2021), the project maps images according to visual similarity, producing a network-like nebula where clusters form organically. This computational approach departs from conventional force-directed layouts by embedding images in a space defined by their aesthetic features, allowing unexpected formations to emerge. Visual rules further guide the experience: image previews are spaced to prevent overlap, clusters maintain coherence, and zooming transitions are smoothed to enhance exploration. These rules ensure that while the layout is algorithmically determined, it remains navigable and engaging. As a result, surprising relationships appear within the collection—such as unforeseen groupings of portraits or materials—transforming the dataset into a rich visual landscape that resists linear interpretation.

The project foregrounds the viewer’s interpretative agency, positioning exploration as a core element of network literacy. Unlike traditional visualizations that prescribe a narrative, it invites visitors to perform their own readings, deciding where to zoom, which clusters to investigate, and which patterns to pursue. This participatory model aligns with the idea that visual literacy involves not only recognizing patterns but also critically questioning the processes and choices behind what is made visible. By making navigation itself a form of meaning-making, Surprise Machines empowers viewers to construct personal interpretations, acknowledging that each act of seeing is shaped by individual curiosity, background, and context. This approach democratizes access to complex collections, encouraging deeper engagement and reflective encounters with both data and design. In doing so, it challenges passive consumption of information and advocates for an active, questioning stance at the heart of visual literacy.

Finally, Surprise Machines brings together spatial and design decisions in a way that highlights the playful, performative dimension of network literacy. By choreographing visitors’ gestures to interact with the visualization—through a choreographic interface—the project merges physical exploration with digital navigation. This spatial engagement transforms reading a network into a kind of dance, where each movement reshapes the visual field and opens new interpretations. The design choices, from clustering algorithms to interactive transitions, encourage experimentation and surprise, underscoring that networks are not static maps but evolving spaces that respond to viewers’ actions. This performative aspect embodies the core of network literacy: understanding that meaning in networks arises not only from data and design but also from how users move through, manipulate, and interpret these complex structures. Surprise Machines thus stands as a testament to how thoughtful spatial and interactive design can transform network visualizations into immersive, interpretative experiences.

Figure 4. Surprise Machines visualization of Harvard Art Museums' 200,000+ digitized images, mapping visual similarities to enable visitors to explore the museum's vast collection through unexpected connections and choreographic interaction (Rodighiero et al. 2022).

Conclusion

This text has explored three key dimensions of network literacy: design, spatial thinking, and critical interpretation. Design shows how choices in data preparation, layout, and visual conventions shape what networks communicate. Spatial thinking highlights how networks function as spatial imaginaries, revealing patterns, clusters, and relations that might otherwise remain hidden. Critical interpretation underscores the designer’s role in guiding how audiences navigate these structures, reminding us that every decision frames what becomes visible. Taken together, these perspectives define network literacy as a transversal skill that spans visual communication, digital humanities, and data journalism.

One of the most recent developments in this landscape is the use of artificial intelligence to extend how networks are produced and read. Language models can now label clusters automatically, translating dense relational structures into more interpretable categories (Ficozzi et al. 2025). This opens possibilities for multi-scale interpretation, where different levels of zoom reveal labels tailored to the granularity of the view. McInnes’s Toponymy project illustrates such an approach: it embeds contextual semantics into clustering, enabling hierarchical annotation that adapts dynamically as users move between global overviews and local details (McInnes 2025). Together, these techniques reconfigure the balance between human interpretation and machine annotation: while they offer new ways to scale analysis across vast corpora, they also raise questions about what kinds of meaning are introduced, obscured, or distorted through algorithmic labeling (Munk et al. 2024).

Cultivating network literacy, then, is not only about learning to design or read networks but also about working critically with AI as a collaborator in the production of meaning. For designers, this means making informed and creative choices while remaining attentive to the epistemic and ethical consequences of automated interpretation. Just as reading the codex once transformed cultural literacy, learning to read and question AI-augmented networks may become an equally transformative skill for our age. The challenge ahead is to integrate these tools without erasing human interpretation, ensuring that networks remain spaces for dialogue between computational insight and human imagination.

Bibliography

Barabási, Albert-László. 2002. Linked: The New Science of Networks. Cambridge, MA: Perseus Pub.
Barabási, Albert-László, Mónica Bello, Julia Kluge-Fabényi, et al. 2020. Hidden Patterns: Visualizing Networks at BarabásiLab. Edited by Alanna Stang and Peter Weibel. Berlin: Hatje Cantz Verlag.
Bastian, Mathieu, Sebastien Heymann, and Mathieu Jacomy. 2009. “Gephi: An Open Source Software for Exploring and Manipulating Networks.” Proceedings of the Third International ICWSM Conference.
Blasius, Jörg, and Andreas Schmitz. 2014. “Empirical Construction of Bourdieu’s Social Space.” In Visualization and Verbalization of Data, edited by Jörg Blasius and Michael Greenacre. Boca Raton: CRC Press.
Bostock, Michael. 2017. “A Better Way to Code.” Medium, April 28. medium.com/@mbostock/a-better-way-to-code-2b1d2876a3a0.
Bostock, Michael, Vadim Ogievetsky, and Jeffrey Heer. 2011. “D3: Data-Driven Documents.” IEEE Transactions on Visualization and Computer Graphics 17 (12): 2301–9. doi:10.1109/TVCG.2011.185.
Boullier, Dominique. 2018. “Médialab Stories: How to Align Actor Network Theory and Digital Methods.” Big Data & Society 5 (2). doi:10/gfpmxh.
Bourdieu, Pierre. (1979) 1984. Distinction: A Social Critique of the Judgement of Taste. Cambridge, MA: Harvard University Press.
D’Ignazio, Catherine. 2017. “Creative Data Literacy: Bridging the Gap between the Data-Haves and Data-Have Nots.” Information Design Journal 23 (1): 6–18. doi:10.1075/idj.23.1.03dig.
Dondis, Donis A. (1973) 1975. A Primer of Visual Literacy. Cambridge, MA: MIT Press.
Duhaime, Douglas. (2017) 2021. PixPlot. Yale Digital Humanities Lab. github.com/YaleDHLab/pix-plot.
Eco, Umberto. 2009. Vertigine della lista. Milan: Bompiani.
Euler, Leonhard. 1953. “Leonhard Euler and the Koenigsberg Bridges.” Scientific American 189 (1): 66–72. jstor.org/stable/24944279.
Ficozzi, Matilde, Mathieu Jacomy, Dario Rodighiero, Anne Beaulieu, and Anders Kristian Munk. 2025. “Grounding AI Map: The Consequences of Living with the Trouble of an Irreductionist Map.” Design et Abstractions, Revue Design Arts Medias, June. journal.dampress.org/issues/design-et-abstractions/grounding-ai-map.
Grandjean, Martin. 2014. “La connaissance est un réseau.” Les Cahiers du Numérique 10 (3): 37–54. doi:10.3166/lcn.10.3.37-54.
———. 2015. “GEPHI – Introduction to Network Analysis and Visualization.” martingrandjean.ch. martingrandjean.ch/gephi-introduction.
———. 2021. “Introduction to Social Network Analysis.” Paper presented at the HNR+ResHist Conference, June 30. halshs.archives-ouvertes.fr/halshs-03351755.
Gray, Jonathan, Carolin Gerlitz, and Liliana Bounegru. 2018. “Data Infrastructure Literacy.” Big Data & Society 5 (2). doi:10.1177/2053951718786316.
Harman, Graham. 2009. Prince of Networks: Bruno Latour and Metaphysics. Melbourne: Re.press.
Hobbs, Robert Carleton. 2004. Mark Lombardi: Global Networks. New York: Independent Curators International.
Jacomy, Mathieu, Tommaso Venturini, Sebastien Heymann, and Mathieu Bastian. 2014. “ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software.” PLoS ONE 9 (6). doi:10.1371/journal.pone.0098679.
Latour, Bruno. 2005. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford University Press.
———. 2013. An Inquiry into Modes of Existence: An Anthropology of the Moderns. Cambridge, MA: Harvard University Press.
Lazer, David M. J., Alex Pentland, Lada Adamic, et al. 2009. “Computational Social Science.” Science 323 (5915): 721–23. doi:10.1126/science.1167742.
Lévy, Jacques, Ogier Maitre, and Thibault Romany. 2016. “Rebattre les cartes.” Réseaux 195 (1): 17. doi:10/ghqdk7.
Lima, Manuel. 2011. Visual Complexity: Mapping Patterns of Information. New York: Princeton Architectural Press.
Löwgren, Jonas, and Erik Stolterman. 2004. Thoughtful Interaction Design. Cambridge, MA: MIT Press.
Maaten, Laurens van der, and Geoffrey Hinton. 2008. “Visualizing Data Using T-SNE.” Journal of Machine Learning Research 9 (86): 2579–605. jmlr.org/papers/v9/vandermaaten08a.html.
Manovich, Lev. 2018. “Can We Think Without Categories?” Digital Culture & Society 4 (1): 17–27. doi:10/gjt7qq.
Marres, Noortje. 2017. Digital Sociology: The Reinvention of Social Research. Cambridge: Polity.
McInnes, Leland. 2025. “Toponymy.” GitHub. github.com/TutteInstitute/toponymy.
McInnes, Leland, John Healy, and James Melville. 2018. “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.” arXiv. doi:10.48550/arXiv.1802.03426.
Moreno, Jacob L. 1934. Who Shall Survive? Washington, DC: Nervous and Mental Disease Publishing Co.
Munk, Anders Kristian, Mathieu Jacomy, Matilde Ficozzi, and Torben Elgaard Jensen. 2024. “Beyond Artificial Intelligence Controversies.” Big Data & Society 11 (3). doi:10.1177/20539517241255107.
Noichl, Maximilian. 2019. “Modeling the Structure of Recent Philosophy.” Synthese. doi:10/ghp7f4.
Ricci, Donato, Robin De Mourat, Christophe Leclercq, and Bruno Latour. 2017. “Clues. Anomalies. Understanding.” In Designing Interactive Hypermedia Systems, edited by Everardo Reyes-Garcia and Nasreddine Bouhaï. Hoboken, NJ: John Wiley & Sons. doi:10.1002/9781119388272.ch6.
Rodighiero, Dario. 2021. Mapping Affinities: Democratizing Data Visualization. Métis Presses. doi:10.37866/0563-99-9.
Rodighiero, Dario, and Jean Daniélou. 2023. “Weather Map: A Diachronic Visual Model for Controversy Mapping.” In Zoomland: Exploring Scale in Digital History and Humanities, edited by Florentina Armaselu and Andreas Fickers. Berlin: De Gruyter. doi:10.1515/9783111317779-017.
Rodighiero, Dario, Lins Derry, Douglas Duhaime, et al. 2022. “Surprise Machines: Revealing Harvard Art Museums’ Image Collection.” Information Design Journal 27 (1): 21–34. doi:10.1075/idj.22013.rod.
Rodighiero, Dario, Philippe Rivière, and Sarah Kenderdine. 2024. “Super-Vision: Tracing EPFL History through 8,000 Doctoral Theses.” Journal of Digital History 3 (1). doi:10.1515/JDH-2023-0004.
Rodighiero, Dario, Eveline Wandl-Vogt, Elian Carsenat, et al. 2022. “Immersive Architectures for Visual Data Literacy.” Information Design Journal 27 (3): 295–308. doi:10.1075/idj.22016.rod.
Romele, Alberto, and Dario Rodighiero. 2020. “Digital Habitus or Personalization without Personality.” HUMANA.MENTE Journal of Philosophical Studies 13 (37): 98–126. doi:10.5281/zenodo.3950030.
Sachs, Horst, Michael Stiebitz, and Robin J. Wilson. 1988. “An Historical Note: Euler’s Königsberg Letters.” Journal of Graph Theory 12 (1): 133–39. doi:10.1002/jgt.3190120114.
Scott, John. 2000. Social Network Analysis: A Handbook. 2nd ed. London: SAGE Publications.
Shields, Rob. 2012. “Cultural Topology: The Seven Bridges of Königsburg, 1736.” Theory, Culture & Society 29 (4–5): 43–57. doi:10.1177/0263276412451161.
Venturini, Tommaso, Mathieu Jacomy, Axel Meunier, and Bruno Latour. 2017. “An Unexpected Journey: A Few Lessons from Sciences Po Médialab’s Experience.” Big Data & Society 4 (2). doi:10/gf7tk2.
Venturini, Tommaso, Anders Kristian Munk, and Mathieu Jacomy. 2019. “Actor-Network vs Network Analysis vs Digital Networks.” In DigitalSTS: A Field Guide for Science & Technology Studies, edited by Janet Vertesi and David Ribes. Princeton, NJ: Princeton University Press.
Venturini, Tommaso, Mathieu Jacomy, and Pablo Jensen. 2021. “What Do We See When We Look at Networks.” Big Data & Society 8 (1). doi:10.1177/20539517211018488.
Windhager, Florian, Saminu Salisu, Roger A. Leite, Velitchko Filipov, and Silvia Miksch. 2020. “Many Views Are Not Enough.” IEEE Computer Graphics and Applications, 58–71.

Visualizing as a Form of Collective Care

2025-01-01T23:59:00+00:00

Care is easy to recognize on a personal level, especially when it takes the form of small, attentive gestures woven into daily life. We see it in how someone nurses a sick friend, tends a garden, or stitches a quilt by hand. Each act, marked by presence, patience, and the quiet commitment to care through touch, time, and attention. It takes shape through quiet, deliberate acts that often go unnoticed, yet carry lasting weight and meaning. But what does care look like when it scales up—across complex systems where the risks are higher, the people more dispersed, and the consequences harder to see?

This year at VISAP—a mini conference and exhibition exploring the intersection of data visualization, art, and science—we’re asking what it means to approach visualization not just as a practice of analysis and synthesis, but as a form of collective care. How do we design visualizations that not only represent, but also actively protect, nurture, and respect the environments and communities embedded within datasets? What practices emerge when we begin to visualize data with thoughtfulness, empathy, and intention? In a time when data shapes public perception, policy, and personal identity, centering care in our visual methods becomes not just desirable, but essential.

In the world of data visualization, “care” is not a term we use often. We usually talk about clarity, insight, and impact; terms that suggest objectivity and utility. But as datasets expand to reflect our bodies, beliefs, environments, and communities, and as algorithms increasingly shape our collective realities, the visual representation of data becomes an act reflecting politics, culture, and ethics. It shapes how we understand one another and the systems built around us. In this light, the role of the visualizer extends beyond aesthetics or clarity, calling for a deeper engagement with social consequence and ethical responsibility. Recognizing this role means accepting that visual choices can influence narratives, reinforce or challenge biases, and shape public understanding in lasting ways.

In contemporary digital culture, data functions not as a static artifact but as a living archive, one that holds memory, identity, and collective history. Biometric scans, environmental sensors, and geotagged images—nearly every aspect of human life today is captured and converted into data. Giorgia Lupi suggests that working with data can uncover deeper connections (Lupi 2015), revealing not just patterns in the world, but insights into what it means to be human. Her approach invites us to see data not as detached or abstract, but as deeply embedded in the stories, emotions, and lived experiences of individuals and communities.

Within this context, data visualization is not merely a cosmetic tool for representation, but a critical process of reinterpretation, contextualization, and communication. It offers a means to narrate our datafied collective histories, shaping how communities are made legible. Artists and designers working in data visualization act as communicators and storytellers. Through visual, sonic, spatial, or even olfactory forms (such as scent-based installations) they transform abstract data into something tangible, something we can feel, question, and connect with. In doing so, they turn datasets into living archives and visualizations into spaces for reflection, empathy, and care.

Maria Puig de la Bellacasa calls this orientation “matters of care” (Puig de la Bellacasa 2011). It’s a call to move beyond surface concern and into the thick, entangled, affective labor of maintenance, repair, and relationality. It’s an invitation to care for our practices the way we care for each other—not just efficiently, but attentively and critically. Within this framework, care is not about sentimentality; it is a relational and communal ethic. It urges us to take responsibility for the data we engage with and to honor the lives, communities, and ecosystems it represents. To visualize with care is to visualize with empathy: to make visible environmental harm, surface suppressed narratives, reveal shared experiences, and confront the structural biases that too often remain hidden.

Building on this understanding of visualization as a relational and embodied practice, we envision a future in which data visualization becomes a process of restoration, connection, and long-term social resilience. This vision invites us to approach data as a space for healing, resistance, and belonging. It encourages the use of data visualization to support the well-being of both environments and the communities most affected by them. These same values must also guide how we collaborate with emerging technologies, especially as we begin to co-create meaning with algorithmic systems and AI—for example, by shaping how data is interpreted, narratives are generated, and decisions are guided by machine learning tools. This collaboration raises critical questions around authorship, agency, and ethics: Whose data is used? Whose voices are amplified or erased? A care-centered approach to AI foregrounds transparency, accountability, and relational design, prioritizing systems that are socially responsible and culturally aware.

VISAP ‘25 explores the theme Collective Care, inviting bold, critical, and creative works that reflect on the role of visualization in an interconnected world. In conjunction with IEEE VIS 2025, the program welcomes papers, pictorials, and artworks engaging with care, solidarity, and ethical collaboration. VISAP will take place in person at the University of Applied Arts Vienna from November 6–15, while IEEE VIS runs at the Austria Center Vienna from November 2–7.

References

Lupi, G. 2015. “The New Aesthetic of Data Narrative.” In New Challenges for Data Design, edited by D. Bihanic. London: Springer. https://doi.org/10.1007/978-1-4471-6596-5_3
Puig de la Bellacasa, M. 2011. “Matters of Care in Technoscience: Assembling Neglected Things.” Social Studies of Science 41 (1): 85–106. https://doi.org/10.1177/0306312710380301

Ethical and Aesthetical Questions on Stock Images: The Case of AI’s Depictions

2024-01-01T23:59:03+00:00

In this article, the authors deal with stock images depicting AI as a face or a body that undergoes a process of fragmentation into particles, pixels, or voxels. These images, they contend, are the symptoms of a datafied worldview. In the first section, the authors discuss stock images of AI and account for their qualitative-quantitative analyses of about 7,500 images from the online catalog of Shutterstock. These analyses have brought out datafied faces and bodies as one of the main themes among stock images of AI. In the second part, the authors elaborate on the notion of datafication of the worldview and offer some examples from architecture and design. This second section includes a methodological detour, in which the authors propose articulating Panofsky’s iconology and Didi Huberman’s “symptomatic” perspective. In conclusion, the authors reflect on an apparently marginal aspect of stock images of AI: the abundant use of blue.

In Praise of Visual Representation: An Inquiry into Text Analysis and Network Visualization for Charting Scientific Communities

2024-01-01T23:59:02+00:00

This article provides a visual investigation into scientific communities through the lens of language. Inspired by actor-network theory, the study examines how individuals establish connections through shared vocabularies and, consequently, how communities organize themselves into linguistic groups. Using scientific texts to map the lexical dimension, the premise posits that research communities can be visually represented by their members and the words they employ, favouring the comprehension of social structures. The research draws from a decade-long personal experimentation with language-based visual models, to explore how research communities appear according to their lexicon, in which each individual is intricately intertwined. Employing cutting-edge techniques of text analysis and network visualization, the study analyses, organizes and maps scientific communities, clustering individuals into thematic groups based on their language use. The findings are presented through a series of projects that delve into the analytical power of images and unveil novel visual methods to better understand the spatial dynamics of language and communities.

Mapping Techniques for an Automated Library Classification: The Case Study of Library Loans at Bibliotheca Hertziana

2024-01-01T23:59:01+00:00

This paper introduces an innovative analytical method for visualising research libraries, overcoming the limitations of the assumptions made by their classification systems. The approach combines user loan data with deep mapping techniques to graphically display usage patterns and thematic clusters. Dimensionality reduction is used to visualise the catalogue by book loans, and prompt engineering with large language models is used to describe loan clusters with detailed summaries and titles. This approach was applied to the library collection owned by Bibliotheca Hertziana, a renowned research institute for art history based in Rome. The final output was assessed by a group of experts through interviews supported by an atlas providing statistical information on clusters. This yielded promising results towards a more general framework for visually mapping textual collections and capturing their transformation and usage from an interdisciplinary perspective.

Super-Vision: Tracing EPFL History Through 8,000 Doctoral Theses

2024-01-01T23:59:00+00:00

The fiftieth anniversary of EPFL (École Polytechnique Fédérale de Lausanne) offered the opportunity to retrace its history through the digital archives housed by the institute itself. Part of the exhibition Infinity Room 2, the Super-Vision project investigates the practice of academic advising by visualizing 8,000 doctoral theses in a work at the intersection of art, science, and technology. Inaugurated in September 2019 at EPFL Pavilions, Super-Vision presents a diachronic mapping that uses artificial intelligence to shed light on an institutional dataset that would be unobservable otherwise. To achieve such a goal, 8,000 doctoral theses are analysed with natural language processing and mapped with techniques of dimensionality reduction, combining language and time within in an interactive visualization accessible to the public. The project title has a twofold meaning: on the one hand, it refers to the educational practice that connects doctoral students to supervisors; on the other hand, it employs information design like a macroscope to grasp complex phenomena from a distant standpoint. The result offers EPFL employees and museum visitors an original perspective to look at the institute with different eyes.

Advanced Interface Design for IIIF: A Digital Tool to Explore Image Collections at Different Scales

2023-01-01T23:59:02+00:00

This article introduces a proposal for an experimental interface design that uses the International Image Interoperability Framework (IIIF) to facilitate the exploration of image collections through the navigation of relational models created by the scholarly practice of annotation. Within the project From Data To Wisdom, an innovative digital tool was designed by harnessing IIIF resources and leveraging close and distant reading on three levels of detail: micro, meso, and macro. The proposed tool integrates annotation features that enable scholars to analyze individual images and interpret broader connections and patterns across image sets. This article outlines the experimental interface’s theoretical framework, design principles, and significant advantages, highlighting the potential to support interdisciplinary research and advancements in digital art.

Orchestrating Cultural Heritage: Exploring the Automated Analysis and Organization of Charles S. Peirce’s PAP Manuscript

2023-01-01T23:59:01+00:00

This preliminary study introduces an innovative approach to the analysis and organization of cultural heritage materials, focusing on the archive of Charles S. Peirce. Given the diverse range of artifacts, objects, and documents comprising cultural heritage, it is essential to efficiently organize and provide access to these materials for the wider public. However, Peirce’s manuscripts pose a particular challenge due to their extensive quantity, which makes comprehensive organization through manual classification practically impossible. In response to this challenge, our paper proposes a methodology for the automated analysis and organization of Peirce’s manuscripts. We have specifically tested this approach on the renowned 115-page manuscript known as PAP. This study represents a significant step forward in establishing a research direction for the development of a larger project. By incorporating novel computational methods, this larger project has the potential to greatly enhance the field of cultural heritage organization.

Weather Map: A Diachronic Visual Model for Controversy Mapping

2023-01-01T23:59:00+00:00

The Weather Map is a visual model to investigate public debates onmedia. Relying on the Media Cloud archives, the visual model transforms a simplequery into a sophisticated visualization by employing the visual grammar of synop-tic weather charts. Peaks of pressure and clashes between airmasses are used to de-scribe the conflicts in media through the temporal dimension, diving into thehuman and non-human dynamics that make the controversy alive. The WeatherMap was conceived as a digital tool to help students and scholars analyze publicdebates, according to the controversy mapping field founded by Bruno Latour. Inparticular, the visual model pushes the boundaries of network visualization, explor-ing advanced techniques of graphic design. The outcome is a web-based applicationdeveloped in JavaScript and Python at the disposal of education and research.

A Visual Translation of the Pandemic

2022-01-01T23:59:05+00:00

In 1923, Walter Benjamin published translations of Baudelaire’s poetry with a prefatory essay the idea that translation is therefore not only the practice of addressing foreign readerships, but rather a process of authorship in which the original text is amplified with further significance. The authors use the term translation with a meaning that is not only linguistic but also visual. They analyze the coronavirus pandemic by translating scientific literacy through the techniques of natural language processing and data visualization. The Cartography of COVID-19 results from a visual translation that invites readers to explore the current pandemic from a different point of view that extends their perception.

COVIC: Collecting Visualizations of COVID-19 to Outline a Space of Possibilities

2022-01-01T23:59:04+00:00

We describe the COVID-19 Online Visualization Collection (COVIC), its goals, how it came to be, and why we propose such a collection as a new path for design research. The COVIC database contains a collective visualization response to the COVID-19 pandemic gathered from approximately 3,000 articles, each containing one or more visualizations (about 12,000 in total). We have sought to create a resource for design research—a boundary object—that will be useful to any of the disciplines brought together through their response to the pandemic event.

Extending Museum beyond Physical Space: A Data-Driven Study of Aldo Rossi’s Analogous City as a Mobile Museum Object

2022-01-01T23:59:03+00:00

Aldo Rossi composed the famous collage known as Analogous City for the Venice Biennale in 1976. This text presents a visual study of the collage through both physical and digital means: a mobile app works in conjunction with a reprint of the Analogous City in the format of a city map. Forty years after its creation, the collage’s original elements are finally identified and collected, and the mechanisms of composition are disclosed thanks to Fabio Reinhart’s contribution. The map of the Analogous City is analyzed in both historical and museum viewpoints, focusing on the reflections that emerged when exhibiting in Maastricht, Milan, Lausanne, Bergamo, and Rome. Although the map was designed as an interactive installation for these exhibitions, it has turned out to be also an educational tool useful outside museums. If Aldo Rossi created an artwork to think about the reconstruction of the city, likewise, the map of the Analogous City helps to rethink museums by designing their objects in a way they can leave the exhibition for a second life in the city.

An issue of accessibility

The introduction of the printing press to Europe in the mid-1400s was not a complete cultural revolution as it is often made out to be. Johannes Gutenberg’s books were novelty items that only wealthy collectors could afford. The real revolution took place fifty years later in Venice, where the publisher Aldus Manutius introduced typographic and editorial innovations that led to the creation of the paperback (Marzo Magno 2020). In the late fifteenth century, Manutius printed a small book of about sixty pages entitled De Aetna, which radically changed the way of reading. Readers were previously forced to handle heavy books, but the paperback was mobile, increasing the book’s context and reach. Books were purged from glosses and formatted using the italic type, an innovation introduced by the font designer Griffo which entirely revolutionized typography. The combination of these innovations lowered the price of books, making them accessible to a wider audience (Davies 1999, 42). Manutius’ story teaches us how the revolution does not take place with the printing press but with the later wider access to literary works. Contextualizing this teaching in museums, it can be said that the real revolution is not found in new architectures or digital infrastructure but in the concept of accessibility understood as the circulation of collections, in their multiple possible forms.

Museum accessibility originates from private collections often veiled by an exotic and mysterious aura, called cabinet of curiosities. Those exercises of personal curatorship were initially aimed at providing a sense of amazement and aesthetic satisfaction for the sole collector, but over time their audience shifted toward a general public (Cellauro 2012). Today museums extend the concept of the cabinet of curiosities in the form of institutions, making valuable objects that would be otherwise hidden accessible to the public (Anderson and Malouf 2018). Once exhibited, these objects become a form of cultural commons at the disposal of the members of society, so that knowledge can be assessed by a larger public than a few elected individuals as it used to be in the past.

More recently, the notion of museum has been further developed when Franco Russoli, the director of Pinacoteca di Brera, developed the concept of living museum. Initially introduced by his forerunner Fernanda Wittgens, the living museum is premised on the idea that institutions must address a larger public to create a relationship with the city through education and learning (Russoli 2017). The museum is a living organism not limited to the building itself, but extends beyond it, making the objects of the collection reachable by a wide audience.

Following this concept, André Malraux, as other intellectuals of the same period, created a traveling exhibition composed of copies of the most famous artworks, called the imaginary museum (Malraux [1952] 1997). The goal was to exhibit copies of the most famous artworks in more peripheral cities (Rowley and Völlnagel 2020, 23) by subverting Walter Benjamin’s theory, which claims reproduction deprives artworks of the authenticity of their presence, or aura (Benjamin [1968] 2007). Today, aura has lost its relevance, leaving space to artworks’ circulation through sophisticated techniques of reproduction, which Adam Lowe refers to as digital materiality (Lowe 2020). As implicitly stated in the writings of Antoine Hennion and Bruno Latour (1996) or Salvatore Settis (2015), reproduction has always existed and there is no harm in the technique itself. It is the responsibility of society to charge an object with an aura through celebrations, stories, and spaces.

This article embraces the idea of an age of digital reproductions (Davis 1995), relying on a research project about Aldo Rossi’s Analogous City to amplify the work of museums in the city through digital and physical objects. If social networks (Vrana et al. 2021) and digital collections (Snydman, Sanderson, and Cramer 2015) have already opened up institutions on digital reproduction, the map of the Analogous City is a museum object capable of living after the exhibition, like a virus that finds the way of flourishing through transmission. Museum objects can be reinvented in new forms to circulate just like Aldus Manutius’ paperback or Adam Lowe’s digital materiality: the digital and the material forms can be designed as a sole object for education, preservation, and reproduction.

The Five Dimensions of Analogy

The seventies were a period in which architects appropriated the collage, an expressive technique already explored by artists and painters (Braghieri 2019; 2020). For the 1976 Venice Biennale, Aldo Rossi deliberately employs such a technique to talk about the reconstruction of the city (Lampariello 2017). By doing so, he molded a visual representation of analogy, a concept that he already developed in the second edition of The Architecture of the City (Rossi [1966] 1982). Together with those he considered more like friends than colleagues — Eraldo Consolascio, Bruno Reichlin, and Fabio Reinhart — Rossi composed an imaginary city by expanding upon and in some ways diverging from the work that Vitruvius developed on analogy in antiquity (Ortelli 2015; Cache 2018). The collage was composed in Zurich by photocopying images from the private library of Werner Oechslin, the director of ETH architecture department, when the library was still located in the city center.

The authors were inspired by the painting Capriccio with Palladian buildings (Canaletto 1756) in which Canaletto arranges three Andrea Palladio’s buildings in a fictional Venice, in such a credible way to fool citizens at the time (Algarotti 1759). The composition is created using three forms of analogy, discussed by Peter Eisenman in the editor’s introduction of The Architecture of the City (Rossi [1966] 1982). One analogy is temporal: elements from different times cohabit the same space; although they belong to different periods, the Palladian Basilica and Palazzo Chiericati appear together in that overlapping of buildings that architects call a palimpsest. The second analogy is spatial, whereby Canaletto relocates different monuments to Venice. The basilica of Vicenza, for example, appears on the banks of the Grand Canal with gondolas sailing around. The third analogy is operational, as the architectural tension between planning and construction represents two different stages of the design process (De Michelis 2014). Andrea Palladio’s Rialto Bridge, at the center of the composition, was never selected to replace the old wooden bridge over the Grand Canal, and thus portrays built and never-built buildings.

Peter Eisenman also discusses a fourth analogy of scale that is attributed to Capriccio but is explicitly expressed in the technology used for the collage (Rossi [1966] 1982). There is no doubt that the topographical scale is related to the act of reassembling buildings from distant places, but the photocopier allowed Rossi and his friends to play also with the physical scale. The photocopier, a tool also used for artistic experimentation (Urbons 1991), was for them a means of experimenting with relationships of scale between the elements. The analogy becomes scalar by transforming the Analogous City into the imaginary place where architecture plays out of the rules. In an evident conflict of scale, the Notre-Dame du Haut by Le Corbusier (Boesiger and Girsberger 1967) is placed inside one Como’s courtyard, drawn by Gianfranco Caniggia (1963). The analogy lies in the tension between architecture and urban scale (Campanile 2020), but also between reality and imagination as written by Aldo Rossi himself (Rossi 1976). Just like the playfulness that characterizes Aldo Rossi’s drawings, the Duomo of Milan becomes an interior furnishing (Celant and Huijts 2015), and the colored pencil becomes a postmodern building (Venturi, Izenour, and Scott Brown 1977). In the Analogous City, the authors reuse the analogy as a tool to reinvent the city through a creative process that recalls Gianni Rodari’s arbitrary association ([1974] 2010).

Yet, there is still a fifth dimension that is visible but barely examined, which can be called the analogy of reverse. Although Aldo Rossi already investigated the relationship between inside and outside of the house in Spazio chiuso, interno (Savi 1976), the Analogous City developed it further. The most valuable example is represented by the Laurentian Medicean Library. While Michelangelo brings the exterior facade inside the library’s entrance (Murray 1971), Rossi draws a garden inside the library, displacing an external element to an interior environment. Although this detail might seem small, it tells a lot about the approach to the collage, both theoretical and playful.

Fabio Reinhart’s Interpretation

As in all complex objects, we undertake a path of deconstruction to understand the Analogous City. The occasion presents itself with the exhibition curated by Ton Quik on the initiative of the Bonnefanten Museum in Maastricht, called Aldo Rossi: The Window of the Poet (Celant and Huijts 2015). The map represents the scientific contribution of the École Polytechnique Fédérale de Lausanne (EPFL), coordinated by the Archizoom gallery and the digital humanities laboratory, respectively directed by Cyril Veillon and Frédéric Kaplan. The project was brought to light thanks to various authors, including Fabio Reinhart whose contribution is essential in order to fully understand the Analogous City. Forty years later, two precious texts (Reinhart 2015a; 2015b) complete the original thoughts expressed by Aldo Rossi (1976) and Manfredo Tafuri (1976), previously published on Lotus International during Venice Biennale 1976. The city is an object that needs to be read through buildings, urban plans, drawings, images, and architectural projects. As stated by Cameron McEwan, these elements “were understood as ‘texts’ linked within a discursive chain connected to thought and ideology made ‘readable’ by critique” (McEwan 2020). In the years of the semiotic revolution guided by Umberto Eco, the interpretation of a text had to be extended to a large variety of artifacts, including the collage of the Analogous City.

In his first text, entitled “Captions for the Analogous City,” Reinhart (2015a) invites visitors to enter the city through its three entrances. The main entrance can be reached by following Lake Maggiore’s waters through the walls of Castelgrande in Bellinzona (Reichlin and Reinhart 1969). The metaphysical entrance is indicated with a gesture by David (Tanzio da Varallo 1625) whose outstretched arm, redrawn by Reinhart himself, replaced Goliath’s head. The third entrance, the intimate one, hides within the walls of Bellinzona in a tortuous path originating from the drawings of Scandicci’s town hall (Rossi, Fortis, and Scolari 1970), taken up in turn from Karl Friedrich Schinkel’s work for Charlottenhof Palace (Whyte 2000).

Reinhart also describes the collage of the Analogous City as a mechanical machine, inspired by Jean Tinguely’s mechanisms (Isgro 2020), which pivots on three gears placed at the center of the composition. These gears correspond to the Monument to Moncenisio by Giuseppe Pistocchi (Godoli 1974), the Vitruvius’ ideal city by Giovanni Battista Caporali (1536), and the plan of the San Cataldo Cemetery by Aldo Rossi and Gianni Braghieri (Savi 1976).

In addition, Reinhart explains how the Analogous City is organized into four semantic areas that correspond to four types of cities: the topographical city by Guillaume-Henri Dufour, which represents an area of Ticino (1865), the urban city mapped by Gianfranco Caniggia, which illustrates Como’s urban structure (1963), the ideal city by Vitruvius, which is described in De Architectura (Caporali 1536), and the mnemonic city by Giovanni Battista Piranesi, which recalls an imaginary Rome (1913).

Fabio Reinhart’s reflections are historically interesting because it is the first time that one of the authors of the Analogous City recorded their reflections regarding the composition. The semantic areas of the collage have richer meaning when combined with the entrances, which delineate the passages between areas. Furthermore, the three gears moving the entire mechanism make the viewer think of a complex machine moving heavily and loudly, almost to the point of listening noise. These two views express the notion that the city is an intricate and complex mechanism, also for those who designed and built it.

Deconstructing the Analogous City

Deconstructing the Analogous City means collecting the original references used by Rossi and his friends to compose the collage in order to explore their visual culture. The work had already been set up by Reinhart, who identified more than thirty references (Figure 1), a majority of which were extracted from the series directed by Pier Luigi Nervi, Storia universale dell’architettura (Universal History of Architecture) and includes volumes such as Architettura del Rinascimento (Architecture of the Renaissance) by Peter Murray (1971) and Architettura contemporanea (Modern Architecture) by Manfredo Tafuri and Francesco Dal Co (1976). Other references included Aldo Rossi’s projects, such as the housing unit in the Gallaratese district of Milan (Rossi, Fortis, and Scolari 1970), part of the apartment complex Monte Amiata by Carlo Aymonino, or the San Rocco housing unit designed in collaboration with Giorgio Grassi (Savi 1976).

Figure 1. Fabio Reinhart compiled the first list of references on which the research was based (private archive). Each single reference is located in the collage on the front page.

It is worth noting that some references stand for the authors’ regions of origin, located between Milan, the Lake Maggiore, and Ticino. Among them, the house by Bruno Reichlin and Fabio Reinhart in Vezio (Fera and Conti 2007), David and Goliath by Tanzio da Varallo in Varallo Sesia (1625), Casa Croci by Antonio Croci in Mendrisio (Fera and Conti 2007), the monument to the partisans by Aldo Rossi in Segrate (Rossi, Fortis, and Scolari 1970), and the floor plan by Max Bosshard, Eraldo Consolascio, and Orlando Pampuri in Brontallo (Rossi, Consolascio, and Bosshard 1979). This demonstrates that their visual choices were deeply rooted in their culture, and therefore strongly related to the territories where they were born and lived.

The collection of the original references was made possible thanks to the support of the EPFL library, which owns an extensive section of architecture, and the collection of E-rara, a Swiss digital platform that houses a large number of old prints. Research on physical and digital archives was a moment of great interest but also of frustration due to the references difficult to find. For example, the geometric solids by Augustin-Charles d’Aviler (D’Aviler 1738) and the Doric column by Andrea Palladio (Palladio 1786) were incredibly challenging to identify because of the large number of versions, often being too similar to be distinguished. The correct version of d’Aviler’s geometric solids was identified by using a magnifying glass to count the shading lines of the cone (Figure 2), while Andrea Palladio’s Doric column was recognized in the version drawn by Ottavio Bertotti Scamozzi through a process of comparison (Figure 3). It is worth noticing that the image was not created by Palladio’s historical collaborator Vincenzo Scamozzi but by his heir Ottavio Bertotti Scamozzi.

Figure 2. Augustin-Charles d'Aviler's geometric solids are a reference to Alberto Savinio, who exhibited at Palazzo Reale in Milan in 1976, the same year of the Analogous City (Savinio 1976).

Figure 3. The Doric column originally drawn by Palladio is reproduced by Ottavio Bertotti Scamozzi, heir of Andrea Palladio's collaborator, Vincenzo Scamozzi (Palladio 1786).

On the other hand, the coffee maker was the latest reference to be found thanks to the personal archive of Beatrice Lampariello; the Spanish magazine Construcción de la ciudad had dedicated to Aldo Rossi two volumes that were difficult to find (Rossi 1975a; 1975b). It is curious to observe how, in the context of a study on Aldo Rossi’s graphic work, the most difficult item to find was the drawing of a coffee maker, an object that Rossi loved to represent most.

The final list includes 45 references in all that have been summarized in two visuals. Figure 4 shows all the references as they appear in the collage in separated squared frames, making it possible to identify there the most extensive elements that characterize the semantic areas: the mnemonic city by Piranesi (first row, last column), the ideal city by Vitruvius (second row, first column), the urban city by Caniggia (second row, fourth column), and the topographical city by Dufour (second row, last column). In Figure 5, all the squared frames are reassembled in a unique image that unequivocally recalls the Analogous City.

Figure 4. This figure collects the 45 references that make up the Analogous City. Their position in the frames corresponds to the position they occupy in the collage, and their image has not been cropped on the contrary of the elements present, almost always partial in the composition.

Figure 5. This image collects most of the references that make up the Analogous City, positioned as in the original. It is surprising to recognize how the simple overlap immediately recalls Aldo Rossi's collage.

Recovering the elements of the Analogous City was an investigation into the visual culture of Aldo Rossi and his friends. The final list of references is a sort of visual bibliography that authors left behind, which gives importance to the visual representation as an alternative to writing. When the Analogous City is dissected and deconstructed, we are met with an utterly intimate representation of the city thanks to the personal experiences, origins, and biases of its authors. Tracing the elements that compose the collage is a historical work that gives a deeper understanding of the intellectual context in which the work was created.

Transformation Into a City Map

Scientific research often takes shape in articles, but recently researchers have been experimenting with new formats of communication (Gorman 2020). The Swiss typographer Adrian Frutiger, for example, printed a map about the evolution of signs and symbols (Frutiger 1997). Frutiger’s experiment is interesting given that maps — objects falling somewhere between books and posters — are characterized by a high portability despite their large printing surface. Although the spread of portable devices has forced these artifacts into disuse, they have been reborn through their satisfying rediscovery, further giving us an innovative perspective.

Reflecting on the possible ways to publish the study about the Analogous City, one idea came to mind: what better way to orient yourself in a city than a city map? The city that was transformed into a picture panel for Venice Biennale undergoes a second transformation into a city map as Fabio Reinhart brilliantly underpins, “The map is an interpretation of the picture panel, a sort of re-invention within the framework of a new social and cultural reality: of identical content, but materially and dimensionally mutant” (Reinhart 2015b).

Taking up Adrian Frutiger’s idea, the Analogous City was reprinted as a map (Rodighiero 2015). On the front side, the Analogous City is printed as a city map divided into sectors. The map’s format is appropriate at first glance as the large surface allows the viewer to grasp more detail compared to any book (Figure 6a). Next to it, references appear in alphabetical order with their spatial coordinates, which allow the viewer to identify images in the map, while the introductory text by Reinhart illustrates the proper interpretation of the collage (2015a). The backside hosts the collage’s elements in their wholeness, together with full bibliographic references (Figure 6b). The original article by Aldo Rossi (1976) is also included on the back side, in addition to the second text authored by Fabio Reinhart (2015b), and two introductory texts by Cyril Veillon and Dario Rodighiero. Developing creative or unconventional formats for this research on the Analogous City paves the way for a new perspective, and new possibilities for user interaction.

Figure 6. The study of the Analogous City takes the form of a map in which a grid divides the collage in sectors to identify its references. The publication (Rodighiero 2015) is enriched by two unpublished texts by Fabio Reinhart (Reinhart 2015a; 2015b), in addition to the original texts by Aldo Rossi (Rossi 1976).

The Exhibitions

The map of the Analogous City must be contextualized within the exhibition Aldo Rossi: The Window of the Poet, organized by the Bonnefanten Museum in Maastricht and hosted later by the GAMeC of Bergamo and the Archizoom in Lausanne. Out of this exhibition, the digital installation has been also presented in Triennale di Milano and MAXXI in Rome.

The map of the Analogous City makes use of augmented reality with the idea of developing a digital humanities project. The application called The Analogous City — freely available for tablets and smartphones on Android and iOS — captures the camera video and recognizes the artwork to return an augmented version of it on the screen. In augmented reality, the elements of the Analogous City are indicated by blue circles floating on the map. By clicking them, the visitor identifies the proper bibliographic reference and the photograph when available. In addition, the uncut images of references are visualized in their wholeness, perfectly aligned in terms of size and rotation with the collage, in order to make the viewer aware about eventual cutout portions and modifications of the original image.

The map of the Analogous City thus becomes part of the exhibition, both physically and digitally. In Maastricht and Bergamo, the map is displayed on a wooden table equipped with two tablet computers. The horizontal surface allows visitors to approach the map from multiple sides, creating a participatory space of interaction in which people can talk to each other (Figure 7). Although the application had been conceived to work with the map, some of the visitors started using the tablet with the original artwork, pointing to the Analogous City hanging on the wall, probably attracted by its aura. The map, which was supposed to act as a link between the artwork and the visitor, disappears like the perfect mediator by bringing the visitor even closer to the Analogous City.

Figure 7. This photo, taken at the Bonnefanten Museum, shows a moment of collective interaction around the map that was placed on a table. In the background, the Analogous City is in its original version.

Once any interactive project comes to exhibition space, it is the unexpected uses that intrigue designers and architects. During the exhibition in Lausanne, where the museum surface at our disposal was smaller compared to that of Maastricht, visitors were invited to interact directly with the original artwork and the map was excluded from the exhibition — the idea was suggested thanks to observing these previous experiences. Needless to say, the experiment was successful as the interaction was clear even without the map working as a mediator between the artwork and the visitors. To reaffirm the centrality of the visitor in the museum, a father encouraged his daughter to use the tablet with a smaller print of the Analogous City, showcased at a more appropriate height for her (Figure 8). What is interesting about this image-recognition algorithm is that it works at different scales, as the image it receives through the camera does not bring any information about scale but simply one flat video recording. Once again, the visitor had changed the rules of the game.

Figure 8. A child interacts with a reproduction of the Analogous City at EPFL Archizoom in Lausanne. When invited to play with the original, visitors always bring unexpected behaviors to the project.

The same no-map configuration was presented at Triennale di Milano in the exhibition Comunità Italia (Ferlenga and Biraghi 2015) and at the MAXXI in Rome during a personal exhibition dedicated to Aldo Rossi (Ferlenga 2021). In Milan, the Analogous City was hanging too high to allow a comfortable interaction because the artwork appeared extremely delicate during its unpacking: the two sleeves that compose the collage were evidently unglued, meaning that the piece was at risk for severe damage. In Rome, the Analogous City was displayed in a vast space; however, the interaction was weakened by the anti-theft cable that secured the tablet. Although this system works properly on a table, it was far from ideal for viewing from afar, especially when the Analogous City was exhibited in its 2-meter original format, obliging the viewers to pull the cable repeatedly.

The presence of design in museum projects brings a certain dose of creativity that shines through when the physical constraints of the exhibition need adaptive capabilities. Adjustments to the digital installation become an opportunity to experience new variations on the theme and stimulate the visitors differently from which changes in their behavior can be observed.

The Map as Educational Tool

Descartes once said that things must always be seen in the past, the present, and their future. If we think about the map of the Analogous City in these terms, the past would be the study, the present is the exhibition, and the future is the education. The idea of extending the museum beyond the physical space becomes more concrete when the publication has been distributed. In that sense, the map is a flexible object that the visitor can buy after the exhibition to relive the experience at home or can be bought or downloaded on the Internet without entering the museum.

Yet the map also proved to be an excellent teaching tool, as demonstrated during a lecture given at the Technical University of Delft in December 2017. The organizers were enthusiastic about how an architectural study was delivered to scholars and architects. Each student and researcher attending the lesson was gifted with a copy of the map. After an introductory frontal lesson that was roughly based on the content of this text, attendees were invited to open their maps in a collective gesture that fostered a creative atmosphere, perfect for a workshop (Figure 9). While examining the collage’s images is a suitable exercise for individuals, the workshop was more appropriate in identifying the position of single elements starting from images shown through the projector. Once an element is projected for the class, the first question for the attendees was to think about the semantic collocation of it, including if a specific reference belongs to the city of memory. Questions like the latter helped narrow down the area of research. The workshop was also the right moment to show details such as the garden within the Laurentian Medicean Library and the David with his outstretched arm redrawn by Reinhart. Such traits are barely noticed, yet tell a lot about the technical and creative process behind the Analogous City. Similar details and questions were able to resurface when the map was explored in an educational environment.

Figure 9. The photo, taken at the Technical University of Delft, shows a moment of collective reading. The map becomes an educational instrument that encourages readers to identify the elements of the Analogous City.

Conclusions

The map of the Analogous City is an interdisciplinary project in which aspects of design and typography have been blended with architecture and museology, according to the approach that characterizes digital humanities. The project results in a hybrid object that adheres to digital materiality, a notion in which the boundary between digital and physical is blurred and often crossed by using reproduction techniques. If the city map is a publication that brings to readers the opportunity to look at the Analogous City in detail, it is also worth noticing how the “original” artwork is itself a copy created from an actual collage kept in the archives of Centre Pompidou.

Walter Benjamin’s aura is nothing but a cultural construction of society, which future generations will see increasingly applied to digital objects, as it is happening today with the recent attention to non-fungible tokens (NFTs). When charged with auras, objects of digital materiality can be seen differently and take on a life of their own, like Adam Lowe’s reproduction of Wedding at Cana, which makes more sense in San Giorgio Maggiore, its birthplace, rather than in front of Leonardo’s Mona Lisa at the Louvre Museum.

Objects of digital materiality have the capacity of moving in space to find their personal aura. As Wedding at Cana leaves Paris to come back to Venice to be relocated in its original context, museum objects — already charged with aura by institutions (Alpers 1991) — can likewise be moved to new environments that may increase or modify their aura. For example, some scholars were fond of hanging the map of the Analogous City like a poster so as to decorate their working environment. During the study at EPFL, Frédéric Kaplan used to have a copy of the map in his office. This copy is still charged with the story of how the original elements were collected and transformed into an exhibition.

Examining these examples clarifies how Franco Russoli’s concept of the living museums can be reversed toward a living museum object, capable of being reproduced, multiplied, and distributed out of the museum building itself without losing its aura. In this hybrid form of digital materiality, the object can leave the museum for its second life, without forgetting the museum as a point of origin where the object is empowered with a charge of aura that can increase or diminish during the displacement, according to each personal object’s trajectory (Rigal and Rodighiero 2015).

Returning to Aldus Manutius’ story, knowledge and circulation spark the actual revolution. Bruno Latour refers to objects, such as maps, using the term immutable mobile (Latour 1986), expressing the idea that the intellectual revolution happens through the circulation of objects capable of comparing different cultures. In this context, the digital age resets time and costs related to the displacement, which can happen instantaneously so that museum objects become a form of cultural commons at the disposal of society.

The role of museums is to facilitate the circulation of knowledge, and digital objects represent perfect mediators to connect knowledge with society. Accessibility, in this sense, is a property that establishes connectivity with the territory, like a network whose ramifications extend the museum to the city itself and beyond. Supporting and developing objects of digital materiality has to be seen at this point as the key to adding a new life and value to museums, by extending the reach, relationships, and interactions that were previously impossible.

References

Algarotti, Francesco. 1759. “Lettera a Prospero Pesci.” September 28, 1759. https://complessopilotta.it/opera/capriccio-con-edifici-palladiani/.
Alpers, Svetlana. 1991. “The Museum as a Way of Seeing.” In Exhibiting Cultures: The Poetics and Politics of Museum Display, edited by Ivan Karp and Steven D. Lavine, 25–32. Washington: Smithsonian Institution Press.
Anderson, Wes, and Juman Malouf. 2018. Spitzmaus Mummy in a Coffin and Other Treasures. Edited by Sabine Haag and Jasper Sharp. Stuttgart: Verlag der Buchhandlung Walther König.
Benjamin, Walter. (1968) 2007. “The Work of Art in the Age of Mechanical Reproduction.” In Illuminations, edited by Hannah Arendt. Reprint. New York: Schocken Books.
Boesiger, Willy, and Hans Girsberger. 1967. Le Corbusier 1910–65. Zurich: Les Editions d’Architecture.
Braghieri, Nicola. 2019. “Alcune indagini e riflessioni intorno a City of Composite Presence.” Piano b. Arti e culture visive 4 (2): 61–87. https://doi.org/10/ghqbdg.
———. 2020. MA2020 — Architettura e Collage. Mantova Architettura 2020.
Cache, Bernard. 2018. “Vers une encyclopédie vitruvienne contemporaine.” Presented at the Leçons inaugurales des Professeurs Emmanuel Rey et Bernard Cache, EPFL, April 9.
Campanile, Nicola. 2020. “The Appropriate Form. The Analogy between the Architectural and Urban Scale.” In From Mega to Nano: The Complexity of a Multiscalar Project, 10–27. Palermo: Palermo University Press.
Canaletto, Giovanni Antonio. 1756. Capriccio with Palladian Buildings. Parma: Parma’s National Gallery.
Caniggia, Gianfranco. 1963. Lettura di una città: Como. Roma: Centro studi di storia urbanistica.
Caporali, Giovanni Battista. 1536. Archittetura. Bigazzini. https://doi.org/10.3931/E-RARA-27122.
Celant, Germano, and Stijn Huijts, eds. 2015. Aldo Rossi: Opera Grafica: Etchings Lithographs Silkscreen Prints. Cinisello Balsamo: Silvana Editoriale.
Cellauro, Louis. 2012. “Venice and the Origins of the Art-Historical Tradition of Display of the Modern Museum.” Konsthistorisk Tidskrift/Journal of Art History 81 (2): 94–122. https://doi.org/10/ghqbdc.
D’Aviler, Augustin-Charles. 1738. Cours d’architecture. Paris: Chez Jean Mariette. https://doi.org/10.3931/E-RARA-5.
Davies, Martin. 1999. Aldus Manutius: Printer and Publisher of Renaissance Venice. 2nd printing. Tempe: Arizona Center for Medieval and Renaissance Studies.
Davis, Douglas. 1995. “The Work of Art in the Age of Digital Reproduction (An Evolving Thesis: 1991–1995).” Leonardo 28 (5): 381. https://doi.org/10/btp8xd.
De Michelis, Giorgio. 2014. “What Design Tells Us about Objects and Things.” Design and Culture 6 (2): 187–202. https://doi.org/10/gc8z22.
Dufour, Guillaume-Henri. 1865. “Topographic Map of Switzerland.” Switzerland: Swiss Confederation. http://map.geo.admin.ch.
Fera, Francesco Saverio, and Luca Conti. 2007. Fabio Reinhart: architettura della coerenza. Bologna: CLUEB.
Ferlenga, Alberto. 2021. “Aldo Rossi: The Architect and the Cities.” MAXXI – Museo Nazionale delle Arti del XXI Secolo. https://www.maxxi.art/en/events/aldo-rossi-larchitetto-e-le-citta/.
Ferlenga, Alberto, and Marco Biraghi. 2015. Comunità Italia: architettura, città, paesaggio 1945–2000. Cinisello Balsamo: Silvana Editoriale.
Frutiger, Adrian. 1997. Symbols and Signs: Explorations. Cham: Syndor Press.
Godoli, Ezio, ed. 1974. Giuseppe Pistocchi (1744–1814) architetto giacobino. Faenza: Rotografica Fiorentina.
Gorman, Michael John. 2020. Idea Colliders: The Future of Science Museums. Cambridge, MA: MIT Press. https://doi.org/10.7551/mitpress/11019.001.0001.
Hennion, Antoine, and Bruno Latour. 1996. “L’art, l’aura et la technique selon Benjamin.” Les Cahiers de médiologie 1 (1): 235–41. https://doi.org/10/gd4vq2.
Isgro, Marina. 2020. “Modernism in Motion: Jean Tinguely’s Meta-Mechanical Reliefs, 1954–59.” Art Journal 79 (2): 6–23. https://doi.org/10.1080/00043249.2020.1765550.
Lampariello, Beatrice. 2017. Aldo Rossi e le forme del razionalismo esaltato. Dai progetti scolastici alla “città analoga”, 1950–1973. Macerata: Quodlibet.
Latour, Bruno. 1986. “Visualisation and Cognition: Thinking with Eyes and Hands.” Edited by Henrica Kuklick. Knowledge and Society Studies in the Sociology of Culture Past and Present 6: 1–40.
Lowe, Adam, ed. 2020. The Aura in the Age of Digital Materiality: Rethinking Preservation in the Shadow of an Uncertain Future. Cinisello Balsamo: Silvana Editoriale.
Malraux, André. (1952) 1997. Le musée imaginaire. Paris: Gallimard.
Marzo Magno, Alessandro. 2020. L’inventore di libri: Aldo Manuzio, Venezia e il suo tempo. Roma: Laterza.
McEwan, Cameron. 2020. “Architecture, Multitude and the Analogical City as a Critical Project.” Architecture and Culture, 1–16. https://doi.org/10/ghqbdd.
Murray, Peter. 1971. Architettura del Rinascimento. Edited by Pier Luigi Nervi. Milano: Electa Editrice.
Ortelli, Luca. 2015. “Analogies.” In Architectura Ludens, edited by Alexandre Blanc. EPFL Magma & Principes / Manslab.
Palladio, Andrea. 1786. Le fabbriche e i disegni di Andrea Palladio. Vicenza: Giovanni Rossi. https://doi.org/10.3931/E-RARA-12411.
Piranesi, Giovanni Battista. 1913. Oeuvres choisies: frontispices, compositions: prisons, trophées, plan et vues de Rome dessinés et gravés de 1746 à 1778. Paris: Vincent.
Reichlin, Bruno, and Fabio Reinhart. 1969. Bellinzona: architettura per la città antica. Casabella.
Reinhart, Fabio. 2015a. “Captions for the ‘Analogous City.’” In The Analogous City, the Map. Lausanne: EPFL Archizoom. https://infoscience.epfl.ch/record/209326.
———. 2015b. “From Picture Panel to City-Map.” In The Analogous City, the Map. Lausanne: EPFL Archizoom. https://infoscience.epfl.ch/record/209326.
Rigal, Alexandre, and Dario Rodighiero. 2015. “Trajectoire d’une représentation cartographique en réseau.” Edited by Armelle Couillet and Jasmine Desclaux-Salachas. Cartes & Géomatique 225: 33–41. https://infoscience.epfl.ch/record/214582/.
Rodari, Gianni. (1974) 2010. Grammatica della fantasia: introduzione all’arte di inventare storie. Torino: Einaudi Ragazzi.
Rodighiero, Dario. 2015. The Analogous City, the Map. Lausanne: EPFL Archizoom. https://infoscience.epfl.ch/record/209326.
Rossi, Aldo. 1975a. Construcción de la ciudad: Aldo Rossi 1 Parte. NovoGraphos.
———. 1975b. Construcción de la ciudad: Aldo Rossi 2 Parte. NovoGraphos.
———. 1976. “La città analoga: tavola.” Lotus International, no. 13 (December): 4–9.
———. (1966) 1982. The Architecture of the City. Edited by Peter Eisenman. American edition. Cambridge, MA: MIT Press.
Rossi, Aldo, Eraldo Consolascio, and Max Bosshard. 1979. La costruzione del territorio nel Cantone Ticino. Lugano: Fondazione Ticino Nostro.
Rossi, Aldo, Massimo Fortis, and Massimo Scolari. 1970. “Architettura di Aldo Rossi 1964–1970.” Controspazio 10.
Rowley, Neville, and Jörg Völlnagel, eds. 2020. Between Cosmos and Pathos: Berlin Works from Aby Warburg’s Mnemosyne Atlas. Berlin: Deutscher Kunstverlag.
Russoli, Franco. 2017. Senza utopia non si fa la realtà: scritti sul museo (1952–1977). Edited by Erica Bernardi. Milano: Skira.
Savi, Vittorio. 1976. L’architettura di Aldo Rossi. Milano: Franco Angeli.
Savinio, Alberto. 1976. Alberto Savinio. Milano: Electa Editrice.
Settis, Salvatore. 2015. “Supremely Original.” In Serial / Portable Classic: The Greek Canon and Its Mutations, edited by Salvatore Settis, Anna Anguissola, and Davide Gasparotto, 51–72. Milano: Fondazione Prada.
Snydman, Stuart, Robert Sanderson, and Tom Cramer. 2015. “The International Image Interoperability Framework (IIIF): A Community & Technology Approach for Web-Based Images.” In Archiving 2015 Final Program and Proceedings, 16–21. Springfield, VA: Society for Imaging Science and Technology.
Tafuri, Manfredo. 1976. “Ceci n’est pas une ville.” Lotus International, no. 13 (December): 10–13.
Tafuri, Manfredo, and Francesco Dal Co. 1976. Architettura contemporanea. Edited by Pier Luigi Nervi. Milano: Electa Editrice.
Tanzio da Varallo. 1625. David and Goliath. Varallo Sesia: Pinacoteca Civica.
Urbons, Klaus. 1991. Copy art: Kunst und Design mit dem Fotokopierer. Köln: DuMont.
Venturi, Robert, Steven Izenour, and Denise Scott Brown. 1977. Learning from Las Vegas: The Forgotten Symbolism of Architectural Form. Cambridge, MA: MIT Press.
Vrana, Vasiliki G., Dimitrios A. Kydros, Evangelos C. Kehris, Anastasios-Ioannis T. Theocharidis, and George I. Kavavasilis. 2021. “Top Museums on Instagram: A Network Analysis.” In Research Anthology on Strategies for Using Social Media as a Service and Tool in Business, edited by Information Resources Management Association, 1029–55. Hershey, PA: IGI Global. https://doi.org/10.4018/978-1-7998-9020-1.ch052.
Whyte, Iain Boyd. 2000. “Charlottenhof: The Prince, the Gardener, the Architect and the Writer.” Architectural History 43: 1–23. https://doi.org/10/dxm6kd.

From Wisdom to Data: Philosophical Atlas on Visual Representations of Knowledge

2022-01-01T23:59:02+00:00

The purpose of this book is to collect in a single volume the multiplicity of approaches, perspectives, and contexts in which the research project From Data to Wisdom has been developed. One of the theoretical assumptions achieved is the idea that contemporary visualizations have abandoned the desire for absolute, synthesizing depiction in favor of a representation that is always imperfect and, so to speak, “in the making.” The very idea that data visualizations are representative, i.e., that they refer to “reality” has been widely discussed and criticized as well. Our intention, therefore, is to present a series of analogies between texts and images. In short, this book will be a mise en abyme of the always incomplete synthesis, but for this very reason richer in meaning, which in our opinion is at the heart, or at least should be, of today’s visual representations of knowledge.

Immersive Architectures for Visual Data Literacy

2022-01-01T23:59:01+00:00

The datafication process transforming society enables us to witness the pandemic from a global perspective. This article provides an example of immersive architecture in which coronavirus-related scientific literature was revealed during Ars Electronica 2021. Like a starry sky, a network visualization representing more than 600,000 articles was showcased in the Deep Space 8K theater, where spectators were accompanied in reading insights. The case study of 3D Cartography of COVID-19 illustrates a novel way to present data in public spaces to foster conversations and reflects on how visual data literacy can be addressed in museums.

Surprise Machines: Revealing Harvard Art Museums’ Image Collection

2022-01-01T23:59:00+00:00

Surprise Machines is a project of experimental museology that sets out to visualize the entire image collection of the Harvard Art Museums, intending to open up unexpected vistas on more than 200,000 objects usually inaccessible to visitors. Part of the exhibition Curatorial A(i)gents organized by metaLAB (at) Harvard, the project explores the limits of artificial intelligence to display a large set of images and create surprise among visitors. To achieve such a feeling of surprise, a choreographic interface was designed to connect the audience’s movement with several unique views of the collection.

Introduction

Although “the humanities so far has focused on literary texts, historical text records, and spatial data,” as stated by Lev Manovich in Cultural Analytics (Manovich 2020, 10), the recent advancements in artificial intelligence are driving more attention to other media. For example, disciplines such as digital humanities now embrace more diverse types of corpora (Champion 2016). Yet this shift of attention is also visible in museums, which recently took a step forward by establishing the field of experimental museology (Kenderdine et al. 2021).

This article illustrates the visualization of an extensive image collection through digital means. Following a growing interest in the digital mapping of images – proved by the various scientific articles published on the subject (Bludau et al. 2021; Crockett 2019; Seguin 2018), Ph.D. theses (Kräutli 2016; Vane 2019), software (American Museum of Natural History [2020] 2022; Diagne et al. 2018; Pietsch [2018] 2022), and presentations (Benedetti 2022; Klinke 2021) – this text describes an interdisciplinary experiment at the intersection of information design, experimental museology, and cultural analytics.

Surprise Machines is a data visualization that maps more than 200,000 digital images of the Harvard Art Museums (HAM) and a digital installation for museum visitors to understand the collection’s vastness. Part of a temporary exhibition organized by metaLAB (at) Harvard and entitled Curatorial A(i)gents, Surprise Machines is enriched by a choreographic interface that allows visitors to interact with the visualization through a camera capturing body gestures. The project is unique for its interdisciplinarity, looking at the prestigious collection of Harvard University through cutting-edge techniques of AI.

The text structure illustrates to the reader the museum and its collection, the curatorship philosophy behind the exhibition, the technical solutions adopted for the visualization, the software for touchless interaction, and how the team coped with technical drawbacks.

Harvard Art Museums

The story of the Harvard Art Museums starts in 1891 with a generous donation by Mrs. Elizabeth Fogg to establish an art museum in memory of her husband, William Hayes. A few years later, the architect Richard Morris Hunt built the Fogg Art Museum at the heart of Harvard University, which witnessed its opening to the public in 1895. On its hundredth anniversary, Director James Cuno describes the rapid growth of art collections by reporting that “more than 450 new art museums were built and more than 100 existing museums added over ten million square feet of new space” (Cuno 1996). Cuno also writes that Harvard University was not an exception, raising new spaces for the Arthur M. Sackler Museum and the Busch-Reisinger Museum to host an extensive collection of 150,000 works. This collection is now part of the Harvard Art Museums, which integrated the three museums into a unique institution in 1983.

Since the publication of James Cuno’s volume in 1996, Harvard Art Museums has exceeded the number of 200,000 works. Although the several extensions of the landmark Fogg building, the collection’s continuous growth led to a complete restoration, completed in 2014 by the architect Renzo Piano. After the demolition of post-1925 expansions, an entirely new section facing Prescott Street integrated the original building of Quincy Street. In addition, a new glass rooftop structure brought natural light into the galleries, the laboratories, and the original courtyard in Travertine marble (Jodidio 2014).

Around the same time, the digital turn entered the world of museums favoring archival work but also a further expansion of archival capacity. Haidy Geismar describes this transformation through the term contact zone, “where old museum collections and new technologies come together, tracing the translation and extension of collections from card catalogues, storerooms and display cases into digital websites, imaging platforms and collection management systems” (Geismar 2018). In this twist, the Harvard Art Museums created an information system that counts almost 240,000 digital objects, composed of high-resolution images and metadata containing titles, descriptions, attributions, dates, classifications, credits, authors, subjects, media, dimensions, and provenances. All these objects are publicly accessible online (Harvard Art Museums 2012) through the International Image Interoperability Framework (IIIF), which provides a web protocol for image and metadata delivery (Snydman et al. 2015).

In the Harvard Art Museums, the contact zone between physical and digital collections materializes in the multimedia space called Lightbox Gallery. Situated on the top floor, the gallery receives its name from the diffused light that comes from the glass rooftop structure. Conceived by metaLAB and HAM to bring visitors digitally in contact with the collection, the gallery features a large wall screen composed of nine interconnected monitors. As stated on the HAM website, “The Lightbox Gallery is a venue for digital experimentation – a space for projects that respond to museums’ collections through new media and emerging technologies.” This definition perfectly fits the domain of experimental museology in which immersive visualization and cultural data are crucial today (Kenderdine et al. 2021).

Can Machines Curate?

The Curatorial A(i)gents exhibition took place at the Harvard Art Museums in spring 2022; the poster in Figure 1 shows the richness of the exhibition’s program. The exhibition presented a series of machine-learning digital installations by metaLAB (at) Harvard, a laboratory described as “an experimental platform that seeks to model new forms of cultural communication, creative practice, and scientific knowledge production” (Birkle and Däwes 2019, 112; metaLAB 2022). The curatorial philosophy of the exhibition appeared in a pamphlet edited by Mike Maizels and designed by Chelsea Qiu, which collects short essays by authors and guests (Maizels and Qiu 2020). In his article entitled iQueries, the metaLAB founder Jeffrey Schnapp discloses the meaning of Curatorial A(i)gents. He alludes to the curator as the figure “who serves as a relay between museum collections and museum programming,” in combination with the neologism a(i)gent, “as the mark of an encroachment on a terrain […] now carried out via the forms of […] ‘artificial intelligence’” (Maizels and Qiu 2020). Then Schnapp gets to the heart of the matter by formulating the question: can machines replace human curators? The provocative approach usually employed to point out digital inequalities (DiMaggio and Hargittai 2001) and technology biases (Crawford and Paglen 2019) is here shifted to investigate innovative curatorial practices through AI techniques.

Surprise Machines owes its name to Alan Turing’s imitation game, described in the pioneering article “Computing Machinery and Intelligence” (Turing 1950). Can machines think? Turing answered using an experiment in which an examiner tries to distinguish between humans and machines while communicating via a typewriter; the machine is understood as engaged in thought when the responses appear indistinguishable. In the article, Turing develops an argument about AI from multiple perspectives, one of which was inspired by the mathematician Ada Lovelace (McCully 2019). When Lovelace argued that machines are incapable of thought due to their inability to take us by surprise, Turing countered by stating that machines are a frequent source of astonishment due to their unpredictable behavior – thereby generating surprises.

Surprise Machines thus aims to surprise visitors by showing them about 213,000 digitized images from the Harvard Art Museums’ collection (a number that is slightly smaller than the entire collection as not all the objects went through digitalization). On the one hand, visitors are amazed by the size of the collection; on the other hand, they are astonished by the small number of objects showcased in the HAM building (a quick estimate from the website proves that less than one percent of the objects have been made visible to visitors over the years). In addition, an experienced eye may be surprised by more specific cues, such as the considerable number of Bauhaus drawings or the unique assortments of powdered pigments.

Figure 1. This image shows the complete program of Curatorial A(i)gents, which alternates between weekly digital installations and online discussions with the authors.

How to Map 200,000 Images

From the very first moment, there was no doubt that Curatorial A(i)gents needed one project to reveal the entirety of the HAM collection. As Norman Foster’s glass-domed Reichstag plays with the analogy of political transparency (Foster 2011), we imagined Renzo Piano’s glass rooftop structure as the way to shed light on the HAM collection by using data visualization.

Data visualization is a computational and design practice aimed at revealing insights by translating tabular data into visual information. In a recent interview, Manuel Lima articulates the translation process through Nathan Shedroff’s diagram (Rodighiero 2021b). Yet this process is more technically and intellectually challenging when the dataset’s richness has to be represented in its integrity, introducing a high level of complexity in the visual outcome. Lima is probably among the first authors to tackle this subject by creating an archive of network visualizations. Ten years after the publication of Visual Complexity (Lima 2011), the landscape of network visualization has changed considerably thanks to scientists such as Albert-László Barabási (Barabási et al. 2020) who has played a central role in retracing the epistemic trajectory of complexity (Weaver 1948). Among new algorithms, the replacement of force-oriented layouts (Bostock et al. 2011; Jacomy et al. 2014) with dimensionality reduction (Maaten and Hinton 2008; McInnes et al. 2018) is key to handling larger relational datasets. The complexity of mapping extensive collections of images stems from the problem of computing and interacting with a considerable number of elements simultaneously.

For Surprise Machines, a local dataset of images was created by downloading the IIIF manifestos and their files. This operation was possible through the Application Programming Interface served by the Harvard Art Museums (Steward [2015] 2021), which provided about 213,000 JSON and JPG files by taking up 50 GB of disk space. Unfortunately, the first attempt to compute lexical metrics from text annotations was proved unsuccessful for two reasons: on the one hand, the high number of elements undermined the computation time of the force-directed graph layout (Bostock et al. 2011); on the other hand, many objects poorly annotated were unconnected to the network visualization. Although this method previously led to promising results (Moon and Rodighiero 2020; Rodighiero 2021a; Rodighiero et al. 2021, 2022), this was not the case. The solution came through a computational study of Aby Warburg’s Atlas Mnemosyne (Impett and Moretti 2017), which led to a collaboration with the Yale University Digital Humanities Laboratory (DHLAB).

The Yale University DHLAB supports scholars interested in digital methods. Among the digital tools provided by the library unit, the PixPlot tool was a significant breakthrough for Surprise Machines. PixPlot is a software composed of two parts, one in Python for processing and one in JavaScript for displaying (Duhaime [2017] 2021). Python code initially processes a set of bitmap images with Inception Convolutional Neural Network (O’Shea and Nash 2015) trained on ImageNet 2012 (Russakovsky et al. 2015). This code transforms JPG files into vectors consisting of numbers. These numbers are successively computed using a technique of dimensionality reduction called UMAP (McInnes et al. 2018) to create a metric of image similarity. This algorithm returns a list of two-dimensional coordinates through a process that may seem opaque to many but that can be explained (Karjus et al. 2022). Finally, JavaScript code creates a web-based, zoomable interface that employs WebGL to speed up the rendering (Danchilla 2012).

The results coming from PixPlot for visualizing the HAM collection were satisfying: the network visualization was balanced and evenly distributed on the screen; more than 200,000 images were spaced out in clusters by visual similarity (see Figure 2). The effect looked like a nebula of points, ready to be explored by zooming through the web interface. Figure 3 enlarges one area of Surprise Machines, revealing a cluster of handmade portraits in which Mahatma Gandhi and Albert Einstein can be recognized just to the top left of the center.

Using a concept developed by Lev Manovich, the fascination of visualizing extensive collections is purely sublime (Manovich 2008), but zooming into details brings information. The attentive use of the interface enables the discovery of specific clusters such as the already-mentioned Bauhaus drawings or powdered pigments. The designers who created the visualization leave room for viewers whose performative insights change from person to person (Drucker 2013).

Figure 2. This network visualization presents Harvard Art Museums’ collection through PixPlot, an open-access software created by the Yale University Digital Humanities Lab. The nebula shows the images in the two-dimensional space by visual similarity.

Figure 3. By using the web-based interactive interface, the visitor may zoom into specific areas of the visualization to reveal image clusters of Harvard Art Museums’ collection. In this area showing hand-made portraits, Mahatma Gandhi and Albert Einstein are visible at the top left of the center.

Designing a Post-Pandemic, Choreographic Interface

Curatorial A(i)gents was slated to open in 2020 but was postponed by two years due to the pandemic. The 11 projects comprising the show were all screen-based, with about half requiring conventional mouse and keyboard interactions like clicking and scrolling. Uncertain about the health protocols in place when the show opened, metaLAB proactively sought a solution for touchless interfacing with the projects. The assembled team saw this as an opportunity to research and develop a “choreographic interface” that would enable museum visitors to use a gestural vocabulary for exploring the projects.

Conceptually, the choreographic interface is a “human-computer interface that increases the kinetic and spatial interactivity between humans and computers through integrating choreographic thinking into the design process” (Derry et al. 2022). Compositional models pertaining to movement, space, and time were used to develop a full-torso gestural vocabulary specific to the projects’ interaction needs. These included tracking, selecting, dragging, zooming, scrolling, right/left advancement, hand-switching, and browser refresh behaviors (see Figure 4). To interpret the choreographed gestures, we used open-source machine vision and machine learning tools such as OpenCV, TeachableMachine, and MediaPipe. Significant iterations were made in response to these technologies’ constraints; these tools are limited to only processing static positions that are extremely distinct from one another. This led us to choreograph torso gestures that are highly geometric and dissimilar, and that sculpt the negative space about the body to make the limbs always visible. Over many iterations, we developed a vocabulary that balances choreographic interest and computational legibility. Much of the prototyping affected Surprise Machines because it included the most challenging pack of interactions. For example, managing the choreography of hand tracking, selecting, dragging, and switching without one method being misinterpreted as another is difficult since all these gestures center on the hand.

Surprise Machines requires simultaneous tracking and identification of the hand position and torso gestures. Our early solutions were computationally heavy and practically non-performing, especially when running the earlier version of Surprise Machines that used PixPlot. We eventually achieved faster performance by using MediaPipe’s Holistic model for pose estimation and a logistic regression model for the classifier; Surprise Machines’ shift to Trails, which is a lighter application, also boosted performance.

Prototyping for Surprise Machines also helped us think through the intuitiveness and relationships of the gestures. For example, we settled on hands-to-shoulders for zooming in; and arms outstretched in a “T” shape for zooming out to echo the sensation of bringing the visualization into the body and away from the body while interacting. Finding it awkward to move one’s dominant hand across the body to the upper corner on the opposite side of the screen, we implemented a simple method for switching dominant hands by looking at the passive hand. Wishing to provide visitors with a performative experience while using the choreographic interface, we added sonification as our last embellishment.

The sonification applies and modulates subtle audio textures to the gestures, providing a dynamic score for the dance and reinforcing to the interface actions. The body’s position and acceleration drive changes to the sound in service of marrying physical and digital movement. While the choreographic interface includes visual feedback on the screen (see Figure 5), aural feedback frees the visitor’s visual attention allowing them to concentrate more fully on the projects.

Figure 4. The choreographic interface invites visitors to interact with the Curatorial A(i)gents projects using a gestural vocabulary. This poster by Pablo Castillo shows the gestures for neutral, zoom in and out, scroll down and up, advance right and left, track, select (also used for dragging), switch hands, and refresh.

Figure 5. Surprise Machines test at the Harvard Art Museum’s Lightbox Gallery showing the choreographic interface in action. The figure also clearly illustrates the Trails’ interface based on a limited set of image previews, distinguishable by the white background.

A Substantial Drawback

Testing was a delicate phase of the project, especially considering that Surprise Machines was developed by a team whose members were based in different parts of the world. Thankfully, some of us were at the Harvard Art Museums when lockdown measures against COVID-19 were less strict, which made testing possible.

When the first test for data visualization took place, the team had mixed feelings. Although the visual was incredibly captivating, the hardware did not support the computation load because of the high number of images shown simultaneously. As a result, another software called Trails – conceived by DHLAB for large screens – was adopted to create a lighter version of Surprise Machines. Even though the back ends of PixPlot and Trails were similar, the front end’s computational load was reduced by replacing the images with distributed samples (see Figures 6 and 7). The new interface was less sophisticated but way faster and more reactive, a relevant feature considering that the Lightbox Gallery’s hardware also managed the choreographic interface’s load.

When part of the team was working on the visualization between the U.S. and Europe, another part was dispersed across U.S. states designing the choreographic interface. While fortunate to run a few remote and in-person tests at HAM, problem-solving for the Lightbox Gallery’s system was challenging as the team was developing Surprise Machines on different hardware setups. The operating system, hardware age, and hardware setup that included the camera, speakers, and nine interconnected monitors were significantly different. Although a choreographic interface for Surprise Machines was fully functional, it was not robust enough to be left installed for the exhibition’s 11-week duration. Since public health protocols did not inhibit the use of shared devices in spring 2022, the final installation featured an air mouse to interact with the projects. Nonetheless, we look forward to including the choreographic interface for future Surprise Machines renditions when the tech driving the system can be managed more closely.

The Lightbox Gallery finally hosts Surprise Machines during the week of May 17–22, 2022. An air mouse was at visitors’ disposal to move the cursor as if it were running on a standard personal computer, featuring dragging and dropping as well as zooming in and out. While most experienced visitors interacted with the digital installation without help, some had difficulty due to limited literacy, as previously discussed in a scientific article by Katy Börner and her colleagues (Börner et al. 2016).

Even with the increase in data literacy, new technologies still struggle to enter museum spaces. The Harvard Art Museums transformed a part of the building Piano envisioned as a meeting room into a laboratory dedicated to experimental museology. However, Curatorial A(i)gents and Living by Protocol, the two back-to-back exhibitions curated by metaLAB, were the penultimate programming in the Lightbox Gallery. Conceived in July 2012 and opened in November 2014, the Lightbox Gallery was converted into a quiet lounge for visitors in summer/fall 2022. The reason for this second conversion was twofold: on the one hand, the Lightbox Gallery was seen as a space with a cycle predestined to end; on the other hand, there was a lack of support within the museum, whose funding for staffing and maintenance ceased in 2018. The lack of funding was the primary reason for the closure of the director’s office in July 2021. Although the detachment of the Lightbox Gallery from the main exhibition stories gave great freedom of planning, the truth is that Renzo Piano’s restoration did not consider the presence of digital devices within the museum at all, and the result is a building extraordinarily crafted but classically designed. It means that the Harvard Art Museums may still live without any digital faculty but also that the digital turn has still to reach its full potential in museums. The experience Lightbox Gallery leaves is an increased understanding of experimental museology for the staff and the authors who exhibited in the gallery and new knowledge for designing future spaces.

Figures 6 and 7. These are two photographs of Surprise Machines, taken by Sarah Newman during the exhibition: the first one shows the nebula of more than 200,000 images from the HAM’s collection, the second one shows a frozen screenshot after zooming in. The latter represents the sampling used to lighten the interface with the remaining images in the background as circles.

Conclusions

Beyond the technical merits, Surprise Machines is a collective project developed in a stimulating research community above all. Resting on a solid curatorship philosophy, the project embraces a collaborative spirit that finds its most prominent expression in the choreographic interface. Although the latter was not part of the exhibition due to technical limitations, it is valuable to look at failures from a future perspective. Failures are indeed necessary sources of knowledge, as was the case with the closure of the Lightbox Gallery after a life cycle of about ten years. When thinking about a project such as Surprise Machines, it is necessary not to isolate the data visualization from the context. Authors, organizations, buildings, and technologies are among the human and non-human actors that provide extraordinary richness and unpredictability to the outcome. In a research environment where works become increasingly sectoral and specific, it is essential to consider multidisciplinarity as one of the most noteworthy qualities of a laboratory (Manzini 2016), and irreductionism as an inexhaustible source of creativity and inspiration (Latour 1988).

Bibliography

American Museum of Natural History. (2020) 2022. Collectionscope. github.com/amnh-sciviz/collectionscope.
Barabási, Albert-László, Mónica Bello, Julia Kluge-Fabényi, et al. 2020. Hidden Patterns: Visualizing Networks at BarabásiLab. Edited by Alanna Stang and Peter Weibel. Berlin: Hatje Cantz Verlag.
Benedetti, A. 2022. “Browsing and Visualizing Collections of Images.” YouTube, January 5. youtube.com/watch?v=bprSgiqZj00.
Birkle, Carmen, and Birgit Däwes. 2019. “‘Old Media Don’t Go Away, They Mutate’: An Interview with Jeffrey Schnapp.” Amerikastudien/American Studies 64 (1): 111–25. doi:10.33675/AMST/2019/1/10.
Bludau, Marian-Jakob, Marian Dörk, and Frank Heidmann. 2021. “Relational Perspectives as Situated Visualizations of Art Collections.” Digital Scholarship in the Humanities 36 (Supplement 2): ii17–ii29. doi:10.1093/llc/fqab003.
Börner, Katy, Adam Maltese, Russell Nelson Balliet, and Joe Heimlich. 2016. “Investigating Aspects of Data Visualization Literacy Using 20 Information Visualizations and 273 Science Museum Visitors.” Information Visualization 15 (3): 198–213. doi:10.1177/1473871615594652.
Bostock, Michael, Vadim Ogievetsky, and Jeffrey Heer. 2011. “D3: Data-Driven Documents.” IEEE Transactions on Visualization and Computer Graphics 17 (12): 2301–9. doi:10.1109/TVCG.2011.185.
Champion, Erik Malcolm. 2016. “Digital Humanities Is Text Heavy, Visualization Light, and Simulation Poor.” Digital Scholarship in the Humanities: i25–i32. doi:10.1093/llc/fqw053.
Crawford, Kate, and Trevor Paglen. 2019. Excavating AI: The Politics of Images in Machine Learning Training Sets. AI Now Institute. excavating.ai.
Crockett, Dennis. 2019. “IVPY: Iconographic Visualization Inside Computational Notebooks.” International Journal for Digital Art History 4: 3.60–3.79. doi:10.11588/dah.2019.4.66401.
Cuno, James B., ed. 1996. Harvard’s Art Museums: 100 Years of Collecting. Cambridge, MA: Harvard University Art Museums.
Danchilla, Brian. 2012. Beginning WebGL for HTML5. Berkeley, CA: Apress. doi:10.1007/978-1-4302-3997-0.
Derry, Lins, Jordan Kruguer, Maximilian Mueller, and Jeffrey Schnapp. 2022. “Designing a Choreographic Interface during Covid-19.” Movement and Computing Conference. doi:10.1145/3537972.3538020.
Diagne, Cyril, Nicolas Barradeau, and Simon Doury. 2018. T-SNE Map. Experiments with Google. experiments.withgoogle.com/t-sne-map.
DiMaggio, Paul, and Eszter Hargittai. 2001. From the “Digital Divide” to “Digital Inequality”: Studying Internet Use as Penetration Increases. Princeton, NJ: Center for Arts and Cultural Policy Studies, Princeton University.
Drucker, Johanna. 2013. “Performative Materiality and Theoretical Approaches to Interface.” Digital Humanities Quarterly 7 (1). digitalhumanities.org/dhq/vol/7/1/000143/000143.html.
Duhaime, Douglas. (2017) 2021. PixPlot. Yale Digital Humanities Lab. github.com/YaleDHLab/pix-plot.
Foster, Hal. 2011. The Art-Architecture Complex. London: Verso.
Geismar, Haidy. 2018. Museum Object Lessons for the Digital Age. London: UCL Press. doi:10.2307/j.ctv1xz0wz.
Harvard Art Museums. 2012. International Image Interoperability Framework at Harvard University. iiif.harvard.edu.
Impett, Leonardo, and Franco Moretti. 2017. “Totentanz: Operationalizing Aby Warburg’s Pathosformeln.” New Left Review 107: 68–97.
Jacomy, Mathieu, Tommaso Venturini, Sebastien Heymann, and Mathieu Bastian. 2014. “ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software.” PLoS ONE 9 (6). doi:10.1371/journal.pone.0098679.
Jodidio, Philip. 2014. Piano: Renzo Piano Building Workshop, Complete Works, 1966–Today. Cologne: Taschen.
Karjus, Andres, Mar Canet Solà, Tillmann Ohm, Sebastian E. Ahnert, and Maximilian Schich. 2022. “Compression Ensembles Quantify Aesthetic Complexity and the Evolution of Visual Art.” arXiv. doi:10.48550/arXiv.2205.10271.
Kenderdine, Sarah, Ingrid Mason, and Lily Hibberd. 2021. “Computational Archives for Experimental Museology,” 3–18. doi:10.1007/978-3-030-83647-4_1.
Klinke, Harald. 2021. “V-Lab Workshop: Visual Analysis in Cultural Data – Image Plots and t-SNE Maps Made Easy.” History of Art Department, October 28.
Kräutli, Florian. 2016. Visualising Cultural Data: Exploring Digital Collections through Timeline Visualisations. PhD thesis, Royal College of Art.
Latour, Bruno. 1988. The Pasteurization of France. Translated by Alan Sheridan and John Law. Cambridge, MA: Harvard University Press.
Lima, Manuel. 2011. Visual Complexity: Mapping Patterns of Information. New York: Princeton Architectural Press.
Maaten, Laurens van der, and Geoffrey Hinton. 2008. “Visualizing Data Using t-SNE.” Journal of Machine Learning Research 9 (86): 2579–605. jmlr.org/papers/v9/vandermaaten08a.html.
Maizels, Mike, and Chelsea Qiu, eds. 2020. Curatorial A(i)gents. Cambridge, MA: metaLAB (at) Harvard. printedmatter.org/catalog/57243.
Manovich, Lev. 2008. “Data Visualization as New Abstraction and Anti-Sublime.” In Small Tech: The Culture of Digital Tools, edited by Byron Hawk, David M. Rieder, and Ollie O. Oviedo. Minneapolis: University of Minnesota Press.
———. 2020. Cultural Analytics. Cambridge, MA: MIT Press. doi:10.7551/mitpress/11214.001.0001.
Manzini, Ezio. 2016. “Design Culture and Dialogic Design.” Design Issues 32 (1): 52–59. doi:10.1162/DESI_a_00364.
McCully, Emily Arnold. 2019. Dreaming in Code: Ada Byron Lovelace, Computer Pioneer. Somerville, MA: Candlewick Press.
McInnes, Leland, John Healy, and James Melville. 2018. “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction.” arXiv. doi:10.48550/arXiv.1802.03426.
metaLAB. 2022. metaLAB (at) Harvard & FU Berlin. mlml.io.
Moon, Christy Y. E., and Dario Rodighiero. 2020. “Mapping as a Contemporary Instrument for Orientation in Conferences.” Atti del IX Convegno Annuale AIUCD. La svolta inevitabile: sfide e prospettive per l’informatica umanistica. doi:10.5281/zenodo.3611340.
O’Shea, Keiron, and Ryan Nash. 2015. “An Introduction to Convolutional Neural Networks.” arXiv. arxiv.org/abs/1511.08458.
Pietsch, Christopher. (2018) 2022. VIKUS Viewer. github.com/cpietsch/vikus-viewer.
Rodighiero, Dario. 2021a. Mapping Affinities: Democratizing Data Visualization. Open-access English edition. Geneva: Métis Presses. doi:10.37866/0563-99-9.
———. 2021b. “Ars Memorativa as the Genesis of Information Design: A Conversation with Manuel Lima.” Nightingale, August 18. nightingaledvs.com/ars-memorativa. doi:10.31235/osf.io/k5unq.
Rodighiero, Dario, Eveline Wandl-Vogt, and Elian Carsenat. 2021. “Making Visible the Invisible Work of Scientists during the Covid-19 Pandemic.” Visual Culture Studies 2. doi:10.31235/osf.io/m4uht.
———. 2022. “A Visual Translation of the Pandemic.” Leonardo 55 (3). doi:10.1162/leon_a_02203.
Russakovsky, Olga, Jia Deng, Hao Su, et al. 2015. “ImageNet Large Scale Visual Recognition Challenge.” arXiv. doi:10.1007/s11263-015-0816-y.
Seguin, Benoit. 2018. “The Replica Project: Building a Visual Search Engine for Art Historians.” XRDS: Crossroads, The ACM Magazine for Students 24 (3): 24–29. doi:10.1145/3186653.
Snydman, Stuart, Robert Sanderson, and Tom Cramer. 2015. “The International Image Interoperability Framework (IIIF): A Community & Technology Approach for Web-Based Images.” Archiving Conference 2015: 16–21. doi:10.2352/issn.2168-3204.2015.12.1.art00005.
Steward, Jeff. (2015) 2021. API Documentation. Harvard Art Museums. github.com/harvardartmuseums/api-docs.
Turing, Alan M. 1950. “Computing Machinery and Intelligence.” Mind LIX (236): 433–60. doi:10.1093/mind/LIX.236.433.
Vane, Olivia. 2019. Timeline Design for Visualising Cultural Heritage Data. PhD thesis, Royal College of Art.
Weaver, Warren. 1948. “Science and Complexity.” American Scientist 36 (4): 536–44. jstor.org/stable/27826254.

Ars memorativa as the genesis of information design: a conversation with Manuel Lima

2021-01-01T23:59:02+00:00

Manuel Lima is one of the most prominent figures of data visualization since the publication of Visual Complexity (Lima 2011). In this conversation, Manuel Lima traces back the origin of data visualization to Ars Memorativa, an ancient mnemonic technique to organize information and facilitate its recall. Going back to the origins is an obsession that brought him to collect and arrange into books images of information design from both physical and digital archives. By doing this, Manuel Lima tackled issues related to the digital objects and their creation, use, and preservation, with a point of view capable of combining the passion for visualizing information and the profession of UX design. This conversation, which took place between Lisbon and Milan on Wednesday 28 July, 2021, comes from a blurb that Manuel Lima wrote for Mapping Affinities (Rodighiero 2021). The discussion is part of the project From Data to Wisdom, and is supported by Fundação para a Ciência e a Tecnologia through the grant POCI-01-0145-FEDER-029717, and the Swiss National Science Foundation through the grant 194442. This text, originally created for the forthcoming book From Data to Wisdom (Higuera Rubio et al. 2022), is published as a preview for Nightingale, the journal of the Data Visualization Society.

Making Visible the Invisible Work of Scientists during the COVID-19 Pandemic

2021-01-01T23:59:01+00:00

Despite the perceptibility of the effects they impart on their hosts, the most incredible capacity of viruses is in their invisibility. Invisibility is the most frightening side of the current pandemic, and invisible is also the work of the scientists striving to find a solution. This proposal presents a data visualization that aims to give visibility to those scientists working on COVID-19. Their scientific publications have been computationally analyzed and transformed into a relational structure based on lexical similarity. The result is a network of scientists whose proximity is given by their closeness in writing. An innovative visual method that hybridizes network visualizations and word clouds shows the scientists in a deep space, explorable through keywords. In such a space, individuals are situated according to their lexical similarity, and keywords are used to clarify their proximity. By zooming, the visualization reveals more information about scientists and their clusters. While a lot of visualizations during the pandemic focused on showing the spread of infection, causing anxiety among the readers, this visualization reveals the efforts of science in eradicating the virus. Making visible the enormous number of scientists working on COVID-19 research will contribute to coping more positively with the pandemic.

Mapping Affinities: Democratizing Data Visualization

2021-01-01T23:59:00+00:00

Organizations are complex systems whose members leave behind a wealth of digital traces, yet the metrics commonly used to read them—the h-index, the impact factor, and similar indicators—tend to rank individuals rather than reveal the collaborative dynamics that hold a collective together. This book proposes affinity as an alternative lens. Affinities take many forms, from shared interests and committee memberships to co-teaching and co-authoring, and they can be both actual, when a collaboration has already taken place, and potential, when a tie is plausible but not yet realized. Translating affinities into a visual representation produces a space where these two dimensions coexist, offering managers and scholars alike a tool to explore the present and plan future synergies, both top-down and bottom-up. Drawing on five years of design research at the École Polytechnique Fédérale de Lausanne (EPFL), the book develops a data visualization method that reconciles the humanities with technology in the spirit of a new European Bauhaus.

The important process of flattening and decentralizing organizations starts first by understanding their structure. Mapping Affinities by Dario Rodighiero introduces an important network visualization model to map large organizations in order to foster collaboration and autonomy. This is a much-needed effort.

— Manuel Lima

The diagrammatic mapping produced by Dario Rodighiero’s designs combines the complexity of esoteric insight with the clarity of rational logic. This well-formulated approach promises to produce ways of ‘reading’ massive quantities of current data so that its statistical richness becomes graphically legible. In the process, the transformation of mere information into schematic visualizations exposes latent, actual, and potential affinities through a practice suggestive of digital alchemy. This is fascinating work, full of possibilities.

— Johanna Drucker

Introduction

Today, organizations are more than ever complex systems. They are so large, ramified, and intertwined that their organic structure seems like a tangle of activities. Day by day organization members contribute to keeping these structures alive with their actions, behaviors, and thoughts. Organizations rely on these daily practices.

Sociology aims to untangle the network of daily practices through the analysis of the digital traces that members leave on the cloud by using desktop computers, smartphones, Wi-Fi networks, identity cards, and online services. The challenge is to recompose structures and behaviors using the data that its members left behind, in various forms and places.

Understanding from daily activities how an organization fluctuates deeply interests the management. The way in which employees work is fundamental to making decisions and planning the future. In particular, managers are interested in having a global perspective to optimize the performance of their employees as much as possible.

The concept of performance deals with the challenge of obtaining the very best from the organization, and management often uses indicators to measure their employees’ performance. Today, however, two interesting things happen: one is that indicators are moving from tabular to graphical form, the other being that the same indicators are at the disposal of everybody as a form of transparency and self-examination.

Nowadays, performance not only interests corporations but also universities. In academia, scholars are often assessed through their publications using the h-index or the impact factor. Directors use such metrics to recruit scholars and, in turn, the same scholars try to improve these metrics to be positively evaluated. This bidirectional use clearly shows that academia adheres to performance-based logic.

Current academic policies do not usually take into account a dimension that plays a critical role in scholarly dynamics, the affinity between scholars.

This book focuses on the concept of affinity and the ways to visually represent it. Affinities are diversified and take many forms: from common interests to committee memberships, from teaching activities to publication co-authoring. Affinities are also multiple as scholars share different kinds of them at the same time, reinforcing their overall ties.

Affinities can be classified as actual and potential. A certain number of potential affinities indicates a predictable tie between scholars; these affinities might be representative for common interests, interdisciplinarity, intellectual culture, professional career, or scientific journals, conferences, and committees. Potential affinities become actual ones when a collaboration takes place; it may be the case of co-authoring a paper or supervising the same doctoral candidate. As a consequence, affinities offer two different dimensions: one is solid and composed of ongoing collaborations, the other is projected towards the possible opportunities to explore the academic environment.

The metric of affinities is crucial for academic organizations. Translating affinities into a visual representation draws a space where actual and potential dimensions can be combined. Contrary to the other metrics that reinforce the ranking between individuals, the logic of affinities is a tool to explore the present for future planning. The attempt is to represent the academic dynamics to foster new synergies. With respect to the logic of governance, planning these synergies is in the interest of both managers and employees to enable top-down as well as bottom-up initiatives.

The metric of affinities has to be, therefore, at the disposal of the whole collective to visually lead individuals in personal choices.

This book is the result of five years working at the École Polytechnique Fédérale de Lausanne (EPFL), during which the problem of mapping affinities was addressed through a design approach. This problem was tackled by visual means, which represent the only solution to manage the enormous mass of data that humans are increasingly producing. The innovation of this work does not stay in the problem itself but rather in the reconciliation of humanities and technology through a new European Bauhaus.

Big Data and the Little Big Bang: An Epistemological (r)Evolution

2020-01-01T23:59:05+00:00

Starting from an analysis of frequently employed definitions of big data, it will be argued that, to overcome the intrinsic weaknesses of big data, it is more appropriate to define the object in relational terms. The excessive emphasis on volume and technological aspects of big data, derived from their current definitions, combined with neglected epistemological issues gave birth to an objectivistic rhetoric surrounding big data as implicitly neutral, omni-comprehensive, and theory-free. This rhetoric contradicts the empirical reality that embraces big data: (1) data collection is not neutral nor objective; (2) exhaustivity is a mathematical limit; and (3) interpretation and knowledge production remain both theoretically informed and subjective. Addressing these issues, big data will be interpreted as a methodological revolution carried over by evolutionary processes in technology and epistemology. By distinguishing between forms of nominal and actual access, we claim that big data promoted a new digital divide changing stakeholders, gatekeepers, and the basic rules of knowledge discovery by radically shaping the power dynamics involved in the processes of production and analysis of data.

Digital Habitus or Personalization without Personality

2020-01-01T23:59:04+00:00

This article aims to offer an original framework to understand the ontological structure of digital media and technologies, along with their effects of subjectivation. In the first section, we confront Bourdieu’s and Latour’s social theories. Indeed, Latour and Bourdieu offered two almost opposite social theories, and both of them can be used to understand digital media and technologies. Our hypothesis is that the digital of today is less Latourian than Bourdieusian. In the second section, we introduce the concept of digital habitus. In particular, we contend that digital machines such as algorithms of machine learning are habitus machines. Although their results present a greater granularity with respect to the standard techniques of the past, these algorithms still reduce individuals to categories, general trends, classes, and behaviors. Such a reduction has flattening effects on the individuals’ self-understanding, especially in terms of identity and interaction with the social world. This is the phenomenon described as the “personalization without personality.” In the third section, we look for proof of our previous insights through a qualitative and comparative analysis between three kinds of data and information visualization. More specifically, we show that contemporary techniques for data visualization with machine learning algorithms are closer to Bourdieu’s use of correspondence analysis (CA) and multiple correspondence analysis (MCA) than to Latour-inspired network visualizations.

Drawing Network Visualizations on a Continuous, Spherical Surface

2020-01-01T23:59:03+00:00

Despite the great literature regarding network visualizations, their graphic representation is hardly an object of investigation. Sometimes it deserves more attention, especially when individuals are represented. Visually translating communities in networks, for example, implies that some individuals are always situated at the borders of the representation. This assumption is clearly unfair, especially if each individual in the community is connected with everybody else. To address this lack of design justice, the community is represented on a spherical network where the surface is continuous. In that space, individuals can be situated in a sparse area, but never on the edges. The spherical network is then projected onto a flat surface to improve readability by making use of cartographic projections.

Introduction

Data visualization was born during the 20th century. Although its history dates back to a more ancient time when diagrams, drawings, and maps were handmade, practitioners do not appear to appreciate this legacy. Despite the separation from historical roots, there is nevertheless no doubt that looking back at its history would be a benefit. Cartography, for example, is one of the disciplines that contributed to shaping the visual language of data. This article investigates the heritage of cartography to improve network visualizations.

Network visualizations are usually drawn on a flat surface (Abbott 2006), which makes them printable and portable like maps. Despite the advantages of flatness, a major drawback occurs when communities are represented as a network: a part of them is predestined to be at the edges. This physiological characteristic of networks is commonly related to the metric of centrality.

It is noticeable how centrality is one of the major metrics that affect interpretation of network visualizations. Viewers are pleased when they appear at the center, but discontent when their name is placed on the borders. Beyond any metric, it is unfair to predestinate a few individuals on the borders. A network in which all the nodes are connected to each other, for example, should not have borders.

Weighing the idea to draw a network without centrality, it was evident that the spherical surface offered what was needed. Indeed, a sphere like the Earth has a surface to draw a network without edges.

This article proves how a continuous surface can be a more democratic space for visualizing communities. Then, to increase readability, spherical networks are unfolded using techniques of geographical projection such as the Mercator or the quincuncial ones. The variety of such techniques are then applied to a dataset representing the digital humanities community.¹ The lexical distance between authors is so translated into pictures in which functionality fades into a sense of aesthetic and sublime (Manovich 2008).

Being a Flat Networker

Last year, Netflix released a documentary about the Flat Earth Society, titled Behind the Curve (Clark 2018). This society argues that the Earth’s sphericity is not perceivable by humans, whose point of view cannot realize the planet’s sphericity. The government, for them, brewed a plot to keep citizens docile and ignorant (Richardson 2019).

Although science provides enough information to contradict them (Eratosthenes 2010), the Flat Earthers can be considered thought-provoking at least. The fascination for them does not originate from eccentricity, but rather from the audacity in questioning an assumption taken for granted.

At school, children learn many truths such as human evolution or the Big Bang theory. But only over the years, they understand these theories in depth, developing a critical method to evaluate truthfulness. Critical thinking is a valuable means to question every kind of knowledge and form an opinion. Sometimes it can be tortuous and tricky, especially in the scientific community (Kuhn 1962). It was the case for Galileo Galilei who, in front of the Inquisition, was forced to recite the abjuration about something that was empirically proven (Fahie 2005).

Opening the black box like the Flat Earthers can show a problem from a different angle. Reframing this approach to data visualization, we can imagine a Flat Network Society whose only aim is keeping networks flat because of a scientific conspiracy. The article’s goal is demystifying such flatness widely employed by scientists (Barabási and Pósfai 2016; Latour et al. 2012), arguing that thinking in three dimensions sometimes can lead to more appropriate network visualizations.

Portraying Winston Churchill

Painters, photographers, and illustrators know that portraying someone is not an easy task. It requires a mix of abilities such as taste, delicacy, and technical ability. Even when these qualities excel, it might happen that the subject does not agree with the representation as it was the case for Winston Churchill.

Last century Graham Sutherland portrayed Churchill at the age of eighty. The artwork to celebrate Churchill’s 80th anniversary was commissioned by the House of Commons and House of Lords, which named the most famous English painter of that time. However, considering Churchill’s temper, it was not a surprise that something went wrong. His disappointment was so bitter that the portrait disappeared in his country house (Clubbe 2005) and, years later, it seems that the painting was destroyed by his wife Lady Spencer-Churchill.

This story shows the complexity of representing personal identity, which is also true in contemporary media. With the recent, massive quantification operated by personal devices (Rodighiero and Rigal 2019), large datasets bring into question the way in which individuals are represented, as well as the perception of their own self (Rodighiero and Cellard 2019). Severe ethical issues relate to the problem of self-representation, as it intimately concerns all human beings. Nevertheless, although the scientific debate on personal data is going on (Floridi 2019), the discussion about the ethics of visualizing individual information is still in an embryonic state.

The Issue of Network Centrality

During the Digital Humanities Conference in 2014, when speakers were invited to recognize themselves in a network visualization representing them, reactions were different. Someone was offended for not being part of the visualization and someone was pleased to be there, sharing portraits on social media (Rodighiero 2018a). The ENAC Research Day caused the same reactions in a similar setting: when faculty members were invited to recognize themselves in a data visualization, responses ranged from engagement to indifference.

Although most of the complaints were related to technical problems easy to solve, something else was not completely focused. After a period of reflection, it was evident that the reactions that originated from self-recognition were influenced by network centrality. Different interviews clearly indicated that the satisfaction was proportional to the closeness to the center: the more a person was placed in the middle of the network, the more the self-perception was flattering.

The origin of centrality lies in the concept of being the “star” of a social group, which is being popular and worthy of attention. But in network analysis, centrality also determines the node’s relevancy (Scott 2000). Such a measure can be not only mathematically calculated but also diagrammatically interpreted, as in networks it is rather easy to see whether a node is central or not.

Looking for a Continuous Space

At the time of these events, placing individuals at the edges of networks appeared a reasonable choice. But that cannot be really defined as a choice since it relies on a physiological characteristic of network visualizations. Putting some nodes at the margins is a convention that the majority of software and libraries follows.

When network entities are non-humans, there is no significant impact. But when entities are individuals, being on the edge can generate a state of discontent. As in real society individuals might feel discouraged when marginalized, something similar happens in visual representations. Placing someone at the edge of a network corresponds to a form of exclusion, and the cause of this marginalization is identified as centrality, since visual discrimination is observed at its presence. Unfortunately, there is no way to get rid of centrality until networks are drawn on a flat surface. Thus, the question to ponder is, “would it be possible to design space without centrality?” In our own humble opinion, the answer is yes.

Discharging the space from the centrality inherited from planar media, such as painting or photography, implies getting rid of frames that impose a drawing limit. A space without frames is potentially a continuous plain with no interruptions.

The simplest manifestation of the contiguous space belongs to geometry and cartography — it is the sphere. Like the planet Earth, the sphere’s surface represents a continuous, non-infinite space where a line can be drawn endlessly. The hypothesis of this article is that mapping a network on the sphere makes the human representation more democratic.

The Inclusivity of Language

The problem of centrality is stronger in the presence of two conditions: (1) nodes correspond to individuals and (2) a high degree of connectivity exists between these individuals. These conditions often take place in relatively small communities like the scientific communities.

Scientific communities are built on social ties whose multidimensional complexity is proportional to the diversity of practices (Rodighiero, Kaplan, and Beaude 2018; Rodighiero 2018b). Among all the potential metrics that can be employed to represent scientific communities, language represents one of the most interesting ones.

Language is egalitarian, transversal, and inclusive of all communities. It differs from the citation system because language is shared beyond any role or seniority and is not subject to any copyright. A network visualization based on such a metric is then an index of democracy, equality, and impartiality.

In particular, a scientific conference is a specific setting that fulfills the two conditions: (1) a conference is a group composed of individuals and (2) their writing gives life to a dense structure of lexical connections. Conference papers can be analyzed in terms of frequency to create a metric. Text analysis can help to compute a lexical similarity, which can be successively translated into a network where individuals are connected by their words.

The Case Study of DH2019

Attendees usually orientate themselves in conferences using digital and printed programs. Although these programs represent the standard, information is fragmented in pages. More recent approaches aim to describe the conference as a whole, as in the case of lexical cartography created by Moon and Rodighiero (2020).

The lexical cartography is based on vocabulary similarity, inferred from papers. This metric is based on the statistical method called TF-IDF, which computes the term frequency of each speaker according to the entire corpus. The result is a list of the most relevant words for each speaker. Words are then used to create the network links by computing the common dictionary that every pair of speakers share.

The case study of the Digital Humanities Conference (DH2019) confirms the issue of centrality raised in these pages. Despite the high connectivity of lexical ties, a set of authors is always confined towards the network’s borders (see Figure 1). The rich connectivity of DH2019 makes the conference a perfect example to demonstrate the efficacy of the spherical layout.

Figure 1. This network visualization represents the Digital Humanities Conference that took place in 2019 in Utrecht, Netherlands. The visualization is composed of authors connected by their common terminology. Despite the high connectivity, some scholars are nonetheless placed on the edges, transforming the network into a sort of ranking.

Drawing on the Spherical Surface

Network arrangement is based on mathematical methods that simulate the molecular dynamics. Verlet integration is a method that Isaac Newton used to integrate the equations of motion, but today is employed in the force-directed graph layout.² Using a relational dataset where weighted edges connect nodes, the network can be represented on a two-dimensional surface, making its structure readable. D3.js is one of the free libraries available to draw networks using JavaScript (Bostock, Ogievetsky, and Heer 2011).

Network visualizations are the result of a process of stabilization in which edges use their force in a process of spatial negotiation. Unlike the force layout applied to obtain a flat network, designing a three-dimensional network requires a modified system of forces³ able to plot the network in a three-dimensional space. Furthermore, drawing on a spherical surface requires an additional force to the system, acting similarly to gravity. Such a force makes sure that nodes, which are bouncing in the space during the arrangement (see Figure 2), are attracted by an imaginary spherical surface around a hypothetical center. The sum of these forces magically stabilizes the nodes on the spherical surface, but this is not the sole condition.

Parameters play an important role in spherical networks. The optimal coverage of the surface derives from a delicate balance of two parameters that operate on the system: the link strength⁴ and the many-body forces.⁵ When the forces are not correctly balanced, the risk is that of obtaining a roundish, slightly curved flat network like Antarctica. If this were to happen, the link strength would be loosened to slide the network gently over the sphere until it is entirely covered. Likewise, the many-body force has to be increased to untwist any eventual cluster of nodes, which might make the visualization unreadable.

Figure 2. This sequence shows three moments of the network arrangement. In particular, it is noticeable how the size of the nodes changes according to their position: larger nodes are closer to the viewer. During the arrangement, these nodes adopt the same size, which means that they are organizing on the spherical surface at the same distance from the sphere’s center.

The result of the agency is a node arrangement that looks like a sphere (see Figure 3). The likeness to a globe is further reinforced by curved lines that stand for edges. The number of authors (around 1,000) and the number of links (around 140,000) shape a dense network that leads to results favorable to a spherical arrangement.

It is noticeable how the network cannot be read in terms of centrality. Such a measure is absent in favor of another one, density. The network density is defined as the number of nodes in a given area that describes a local neighborhood. When a lot of nodes are condensed, it is an index of a highly shared vocabulary, as in the case of a co-authored conference paper. When single authors are situated in rarefied areas, they are still in a contextual neighborhood despite the distance from the others. In any case, every author always belongs to a 360-degree contextual area.

Figure 3. This snapshot shows a representation of the Digital Humanities Conference 2019. The flat network presented in Figure 1 is differently arranged on a spherical surface. No individuals are placed on the borders, each node is over 360 degrees. In such a condition, there is no way to create a ranking based on centrality, it is rather a space that invites interpretation through the density of different areas.

The Origins of Cartographic Projection

To contradict the flat-Earth arguments, it can be said that our planet’s sphericity was already known before the Common Era. The first text of history on this subject comes from Strabo, a philosopher of Hellenistic Greece who lived between 64–63 BCE and 24 CE. Strabo gives us some interesting information about the Earth’s sphericity and the origin of cartography, collected by Duane W. Roller (Strabo 2014).

Strabo’s pupils were well aware of the planet’s sphericity; they had to know notions like equator and the tropics to attend his class (Strabo 2014). Pupils were also used to studying large wooden globes of statuary dimensions. The need for moving these globes to share knowledge represents the origin of cartographic projection (Latour 1986), which Strabo dates back to the 2nd century BCE (Strabo 2014).

Maps, over time, were printed and drawn on papyrus, textiles, vellum, paper, plastic, etc. Today, however, they are mainly displayed on smartphones or personal computers in what was defined as a cartographic turn (Lévy 2016). This digitalization modified the perception of territory, especially through zooming, a gesture that allows the reader to increase and decrease the point of view rapidly. But zooming is not a mere function; it connects worlds in two and three dimensions, the map and the globe. Maps’ bi-dimensionality is thus connected to the globe’s three-dimensionality like in Google Earth, which merges different levels of detail through interaction (Latour 2014).

The hypothesis of this article is that this connectivity between dimensions is still missing for network visualizations, and one potential gateway is represented by the cartographic projection (Dilke 1987). The cartographic projection, indeed, has the ability to flatten the globe without losing the space’s continuity. If you consider the standard European world map, the distance between the USA and Japan is apparently misleading; but once the viewer understands that the left side continues to the right one, everything is clearer. As a result, the geographic projection unrolls the sphere’s surface, making it readable on a flat surface and preserving the spatial continuity. Furthermore, this spatial continuity allows for remodeling the map, according to the focus that might be on Japan or the USA; the rotation of the sphere, associated with the cartographic projection, lets us think of space without centrality in which every state can be placed at the center of the map.

Figure 4. Ptolemy collected a lot of information about projection techniques, improving them. This figure shows how a portion of the sphere is flattened on a two-dimensional surface (Dilke 1987). This method can be applied to the Earth as to any kind of globe, such as the spherical network.

Cartographic Projection of Spherical Networks

The cartographic projection is one of the most ambitious human gestures. It embodies the gigantic effort to grasp and represent the volume of the Earth. This article presents how it is possible to apply this projection to network visualization and what advantages that operation entails. That operation was technically achievable through the d3-geo component⁶ that extends the d3 library with different projection techniques.

The advantages that characterize the cartographic projections are balanced by inconveniences, as it happens in many processes of transformation. The major trouble is the deformation of the distances between the nodes, which is clearly visible in Figure 5. The Mercator projection (Monmonier 2004) clearly shows a deformation that is proportional to the distance from the equator. The Mercator is one of the major projections as it maintains the correct proportion in numerous populated areas. However, the distortion of Norway is clearly noticeable compared to the grid as well as the false magnitude of Antarctica.

Figure 5. The Mercator projection (Monmonier 2004) shows a deformation that is proportional to the distance from the equator. The distortion of Norway is clearly noticeable compared to the grid, as well as the false magnitude of Antarctica.

Network projections thus have to be interpreted carefully as, like a map, the distances can be represented deceitfully. In general—this is valid for all data visualizations—it is necessary to have a general awareness of the creation process to understand every outcome. Such knowledge will allow the reader to read the media with the necessary precautions.

The results of applying the Mercator projection to DH2019 are shown in Figure 6. The comparison between Figures 5 and 6 gives us an idea of the network deformation, in particular at the top and the bottom of the image where connections are rather sparse. It also has to be considered that these margins are defined by manual intervention as the Mercator projection goes to infinity towards north and south. Furthermore, there is another difference from a standard geographical projection: territory doesn’t have any regular geometrical shape. The lines that are straight on the spherical surface give a sense of the deformation of the plane. Figure 5 shows how lines, which are carefully distributed on the spherical surface, tend to be more curved by moving away from the equator.

Figure 6. The Mercator projection applied to a network shows clearly how the deformation of the space is reflected by the length and the curvature of the edges. Understanding space means having the design process clear above all else.

The darkened zone of the Mercator projection (see Figure 6) shows how the network mainly occupies the area corresponding to the Equator, which enables a horizontal continuity. However, other projections have different characteristics, as is the case with Charles Peirce’s quincuncial projection (Krämer and Ljungberg 2016). In the 18th century, it was pretty common for students and scholars to practice with cartography; Charles Peirce, who is internationally known for his studies in pragmatism, also contributed to cartography with his own projection. The quincuncial projection is particularly relevant in this context as it can be tiled in every direction (see Figure 7). Its continuity extends in both horizontal and vertical directions, creating a space that is continuous to the north, east, south, and west. Contrary to the Mercator projection, the quincuncial one keeps a great precision at the poles making it a perfect tool for Arctic expeditions cruising in the Northern Hemisphere. The North Pole is, indeed, exactly at the map’s center while the South Pole is distributed to the four corners. Looking at Figure 7, it is easy to notice how nodes are massed on the diagonals of the image. More difficult to read, in the same figure, is the deformation of the space which is indicated by the curving lines that draw a double-s shape on the diagonals. As noted earlier, these lines are straight on the spherical surface; keeping this in mind will help to perceive the spatial deformation of the globe.

Figure 7. Charles Peirce’s quincuncial projection can be tiled in all directions. This characteristic makes the network arrangement particularly interesting as the space continuity is guaranteed both vertically and horizontally.

Many are the cartographic projections that have been conceived over time. Each of them allows the viewer to see different aspects of the Earth, which also means exploring different forms in this context. The orthographic (Figure 3), the Mercator (Figure 6), and the quincuncial (Figure 7) projections are the most relevant ones in terms of spatial transformation and continuity. However, the d3 library allowed us to explore more unusual projections, which are hereinafter presented in a showcase that is intentionally more aesthetic than rational. These projections are called Waterman’s Butterfly (Figure 8), Conic Equal-Area (Figure 9), Dodecahedral (Figure 10), Icosahedral (Figure 11), and Lee’s Tetrahedral (Figure 12). By a fortunate coincidence, this selection made on personal taste represents a series of geometrical volumes. That means that, potentially, these network visualizations can be printed and mounted with glue.

Figure 8. Butterfly of Waterman.

Figure 9. Conic equal-area.

Figure 10. Dodecahedral.

Figure 11. Icosahedral.

Figure 12. Lee’s tetrahedral.

Conclusions

Our fictional critique of the Flat-Network Conspiracy is a way to point out how network visualizations can further progress in terms of graphic design since their first appearance in Jacob L. Moreno’s sociograms (Moreno 1934). Although many attempts have been made in the three-dimensional space (Fu et al. 2007; Kwon et al. 2015; Sangole and Knopf 2003; Shelley and Gunes 2012), nobody had ever tried to shape the drawing space in terms of continuity.

Even though today the scientific discussion is mainly aimed at data in terms of ethics and privacy, the group of scholars concerned by the way in which data are presented is still too small. Mark Monmonier (1991) and Alberto Cairo (2012), for example, published two works about the way in which graphics can convey false meanings. In a historical moment in which the personal quantification is growing exponentially, it is necessary to take care not only of the data treatment but also of the way in which information is presented.

The case study of the Digital Humanities Conference is an example of how data visualizations created from personal data can create biases and lead to discrimination. Today more than before, designers have the responsibility to present personal data in the most appropriate way. The aim of this article is finally to foster the discussion among designers and scholars, making them aware of the responsibility to develop guidelines to display individuals and communities fairly, working on simple experiences and observations like this one.

References

Abbott, E. A. 2006. Flatland: A Romance of Many Dimensions. Oxford: Oxford University Press.
Barabási, A.-L., and M. Pósfai. 2016. Network Science. Cambridge: Cambridge University Press.
Bostock, M., V. Ogievetsky, and J. Heer. 2011. “D3: Data-Driven Documents.” IEEE Transactions on Visualization and Computer Graphics 17 (12): 2301–2309. https://doi.org/10.1109/TVCG.2011.185
Cairo, A. 2012. The Functional Art: An Introduction to Information Graphics and Visualization. Berkeley: New Riders.
Clark, D. J. 2018. Behind the Curve. Documentary film. Delta-v Productions.
Clubbe, J. 2005. Byron, Sully, and the Power of Portraiture. Aldershot: Ashgate.
Dilke, O. A. W. 1987. “The Culmination of Greek Cartography in Ptolemy.” In The History of Cartography, Volume 1, edited by J. B. Harley and D. Woodward, 177–200. Chicago: University of Chicago Press.
Eratosthenes. 2010. Eratosthenes’ Geography. Princeton: Princeton University Press.
Fahie, J. J. 2005. Galileo: His Life and Works. Replica Edition. Adamant Media: Elibron Classics.
Floridi, L. 2019. The Logic of Information: A Theory of Philosophy as Conceptual Design. Oxford: Oxford University Press.
Fu, X., S.-H. Hong, N. Nikolov, X. Shen, Y. Wu, and K. Xuk. 2007. “Visualization and Analysis of Email Networks.” In 2007 6th International Asia-Pacific Symposium on Visualization, 1–8. https://doi.org/10.1109/APVIS.2007.329302
Krämer, S., and C. Ljungberg, eds. 2016. Thinking with Diagrams. Berlin: De Gruyter.
Kuhn, T. S. 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago Press.
Kwon, O.-H., C. Muelder, K. Lee, and K.-L. Ma. 2015. “Spherical Layout and Rendering Methods for Immersive Graph Visualization.” In 2015 IEEE Pacific Visualization Symposium, 63–67. https://doi.org/10.1109/PACIFICVIS.2015.7156357
Latour, B. 1986. “Visualisation and Cognition: Thinking with Eyes and Hands.” Knowledge and Society: Studies in the Sociology of Culture Past and Present 6: 1–40.
———. 2014. “Anti-Zoom.” In Eliasson: Contact. Paris: Flammarion and Fondation Louis Vuitton.
Latour, B., P. Jensen, T. Venturini, S. Grauwin, and D. Boullier. 2012. “‘The Whole Is Always Smaller than Its Parts’—A Digital Test of Gabriel Tardes’ Monads.” The British Journal of Sociology 63 (4): 590–615. https://doi.org/10.1111/j.1468-4446.2012.01428.x
Lévy, J., ed. 2016. A Cartographic Turn. Lausanne: EPFL Press.
Manovich, L. 2008. “Data Visualization as New Abstraction and Anti-Sublime.” In Small Tech: The Culture of Digital Tools. Minneapolis: University of Minnesota Press.
Monmonier, M. 1991. How to Lie with Maps. Chicago: University of Chicago Press.
———. 2004. Rhumb Lines and Map Wars: A Social History of the Mercator Projection. Chicago: University of Chicago Press.
Moon, C. Y.-E., and D. Rodighiero. 2020. “Mapping as a Contemporary Instrument for Orientation in Conferences.” In Atti del IX Convegno Annuale AIUCD. La svolta inevitabile: sfide e prospettive per l’Informatica Umanistica. Milan: Università Cattolica del Sacro Cuore.
Moreno, J. L. 1934. Who Shall Survive? Washington, DC: Nervous and Mental Disease Publishing Co.
Richardson, G. 2019. “It’s a Flat Earth.” The New Yorker, April 1.
Rodighiero, D. 2018a. “Printing Walkable Visualizations.” In Proceedings of the 5th Biennial Research Transdisciplinary Imaging Conference, 58–73. https://doi.org/10.6084/m9.figshare.6104693
———. 2018b. Mapping Affinities: Visualizing Academic Practice Through Collaboration. PhD diss., École polytechnique fédérale de Lausanne.
Rodighiero, D., and L. Cellard. 2019. “Self-Recognition in Data Visualization.” EspacesTemps.net, August. https://doi.org/10.26151/espacestemps.net-wztp-cc46
Rodighiero, D., F. Kaplan, and B. Beaude. 2018. “Mapping Affinities in Academic Organizations.” Frontiers in Research Metrics and Analytics 3 (4). https://doi.org/10.3389/frma.2018.00004
Rodighiero, D., and A. Rigal. 2019. “The Daily Design of the Quantified Self.” Swiss Informatics Digital Magazine, February 12.
Sangole, A., and G. K. Knopf. 2003. “Visualization of Randomly Ordered Numeric Data Sets Using Spherical Self-Organizing Feature Maps.” Computers & Graphics 27 (6): 963–976. https://doi.org/10.1016/j.cag.2003.08.012
Scott, J. 2000. Social Network Analysis. London: SAGE Publications.
Shelley, D. S., and M. H. Gunes. 2012. “GerbilSphere: Inner Sphere Network Visualization.” Computer Networks 56 (3): 1016–1028. https://doi.org/10.1016/j.comnet.2011.10.023
Strabo. 2014. The Geography of Strabo. Edited and translated by D. W. Roller. Cambridge: Cambridge University Press.

The accompanying software lays out a network on a sphere with d3-force-3d and unfolds it onto the plane through a dozen geographic projections — Mercator, Orthographic, Equal Earth, and others — rendered live with PIXI.js. Source code at https://github.com/rodighiero/spherical-projection; live demo at https://rodighiero.github.io/spherical-projection/. ↩
The D3 force layout is available at https://github.com/d3/d3-force. ↩
Asturiano’s three-dimensional model is downloadable at https://github.com/vasturiano/d3-force-3d. ↩
Link strength is defined at https://github.com/d3/d3-force#link_strength. ↩
Many-body force is illustrated at https://github.com/d3/d3-force#many-body. ↩
More information about the D3 geographic projections at https://github.com/d3/d3-geo-projection. ↩

Immaginare Gesti-Barriera Contro Il Ritorno Alla Produzione Pre-Crisi

2020-01-01T23:59:02+00:00

Potrebbe esserci qualcosa di inappropriato nel guardare all’era post-crisi quando gli operatori sanitari sono ancora “in prima linea”, milioni di persone stanno perdendo il lavoro, e molte famiglie in lutto non possono nemmeno seppellire i loro morti. Eppure, è proprio questo il momento di lottare affinché la ripresa economica, una volta terminata la crisi, non ci riporti allo stesso vecchio regime climatico contro il quale finora abbiamo tentato, senza successo, di combattere. In effetti, la crisi sanitaria è inserita in quella che non è una crisi – sempre passeggera per definizione – ma piuttosto una mutazione ecologica duratura e irreversibile. Se abbiamo delle buone possibilità di “uscire” dalla prima, ne abbiamo ben poche di “uscire” dalla seconda. Le due situazioni non sono alla stessa scala di grandezza, ma resta illuminante articolarle l’una con l’altra. In ogni caso, sarebbe un peccato non riflettere sulla crisi sanitaria per scoprire altri modi di entrare nella mutazione ecologica, piuttosto che farlo alla cieca.

La prima lezione del coronavirus è anche la più sorprendente: è stato infatti provato che, in poche settimane, è possibile sospendere, in qualsiasi parte del mondo e allo stesso tempo, un sistema economico a detta di tutti impossibile da rallentare o redirezionare. A tutti gli argomenti degli ambientalisti sul cambiamento dei nostri stili di vita, si rispondeva sempre con l’argomento della forza irreversibile del “treno del progresso” che niente poteva far deragliare “a causa”, si diceva, “della globalizzazione”. Tuttavia, è proprio la sua natura globale che rende così fragile questo sviluppo, capace invece di frenare e poi fermarsi improvvisamente.

In effetti, non sono solo le multinazionali o gli accordi commerciali o Internet o gli operatori turistici a globalizzare il pianeta: ogni entità su questo stesso pianeta ha il suo proprio modo di agganciarsi agli altri elementi che costituiscono, a un certo punto, l’insieme. Questo è vero per la CO2 che riscalda l’atmosfera globale attraverso la sua diffusione nell’aria; per gli uccelli migratori che trasportano nuove forme di influenza; ma è anche vero, lo impariamo ancora e dolorosamente, per il coronavirus, la cui capacità di connettere “tutti gli esseri umani” passa attraverso l’intermediario apparentemente innocuo del nostro espettorato. I virus sono dei super-globalizzatori: quando si tratta di risocializzare miliardi di esseri umani, i virus lo fanno in fretta.

Ecco allora l’incredibile scoperta: c’era davvero nel sistema economico mondiale, nascosto a tutti, un segnale di allarme rosso vivo con una grossa maniglia d’acciaio temprato che i capi di Stato, ciascuno a sua volta, potevano tirare subito per fermare “il treno del progresso” con un forte stridio di freni. Se la richiesta di virare di 90 gradi per atterrare sulla Terra sembrava a gennaio ancora una dolce illusione, diventa improvvisamente molto più realistica: qualsiasi automobilista sa che per avere la possibilità di dare un’ultima sterzata senza andare fuori strada, è meglio aver rallentato prima…

Sfortunatamente, questo improvviso arresto nel sistema di produzione globalizzata, non sono solo gli ambientalisti a vederlo come una grande opportunità per far avanzare il loro programma di sbarco. I globalizzatori, quelli che dalla metà del XX secolo hanno inventato l’idea di fuggire dai vincoli planetari, anche loro vedono in esso una formidabile possibilità di rompere ancora più radicalmente con ciò che li ostacola nella loro fuga dal mondo. Per loro, è troppo bella l’occasione di sbarazzarsi del resto dello stato sociale, della rete di sicurezza dei più poveri, di quello che rimane delle normative antinquinamento e, più cinicamente, di sbarazzarsi di tutte queste persone in soprannumero che ingombrano il pianeta¹.

Non dimentichiamo, infatti, che dobbiamo supporre che questi globalizzatori siano consapevoli dei cambiamenti ecologici e che tutti i loro sforzi, da cinquant’anni a questa parte, siano diretti non soltanto a negare l’importanza dei cambiamenti climatici, ma anche a sfuggire alle sue conseguenze costituendo bastioni fortificati di privilegi che devono rimanere inaccessibili a tutti coloro che è bene lasciare in disparte. Nel grande sogno modernista della condivisione universale dei “frutti del progresso”, non sono abbastanza ingenui da crederci, ma la novità è che sono abbastanza sinceri da non darne nemmeno l’illusione². Sono coloro che si esprimono ogni giorno su Fox News e che governano tutti gli Stati del pianeta scettici al cambiamento climatico, da Mosca a Brasilia e da Nuova Delhi a Washington passando per Londra.

Ciò che rende la situazione attuale così pericolosa non sono solo le morti che si accumulano ogni giorno di più, ma è la sospensione generale di un sistema economico che offre, a coloro che vogliono andare molto più lontano nella fuga fuori dal mondo planetario, una meravigliosa opportunità per “rimettere tutto in discussione”. Non dobbiamo dimenticare che ciò che rende i globalizzatori così pericolosi è che sanno evidentemente di aver perso, che la negazione del cambiamento climatico non può durare all’infinito, che non esiste più possibilità di conciliare il loro “sviluppo” con le varie sfere del pianeta in cui sarà necessario finire per inserire l’economia. Questo è ciò che li rende pronti a tentare qualsiasi cosa per ottenere, un’ultima volta, le condizioni che permetteranno loro di durare un po’ più a lungo, di proteggere se stessi e i loro bambini. Il “blocco del mondo”, questa frenata, questa pausa inaspettata, offre loro l’opportunità di fuggire più velocemente e più lontano di quanto abbiano mai immaginato³. I rivoluzionari, al momento, sono loro.

È qui che dobbiamo agire. Se l’opportunità si apre a loro, si apre anche a noi. Se tutto viene fermato, tutto può essere rimesso in discussione, flesso, selezionato, ordinato, interrotto per davvero o, al contrario, accelerato. È ora il momento di fare l’inventario annuale. Alla richiesta data dal buon senso: “Riavviamo la produzione il più rapidamente possibile”, dobbiamo rispondere con un grido: “Assolutamente no!”. L’ultima cosa da fare sarebbe rifare esattamente ciò che abbiamo fatto prima.

Ad esempio, l’altro giorno, un fiorista olandese è stato presentato in televisione, con le lacrime agli occhi, costretto a lanciare tonnellate di tulipani pronti per la spedizione che non poteva più spedire via aerea in tutto il mondo per mancanza di clienti. Non possiamo che compatirlo, ovviamente; è giusto che debba essere compensato. Ma poi la telecamera arretrava, mostrando che i suoi tulipani li faceva crescere fuori terra sotto la luce artificiale prima di consegnarli agli aerei cargo di Schiphol in una pioggia di cherosene; da lì sorge un dubbio: “Ma è davvero utile prolungare questo modo di produrre e vendere questo tipo di fiori?”.

A poco a poco, se iniziamo, ognuno per conto proprio, a porre tali domande su tutti gli aspetti del nostro sistema di produzione, diventiamo degli efficaci interruttori della globalizzazione – efficaci, milioni come siamo, come il famoso coronavirus nel suo modo di globalizzare il pianeta. Ciò che il virus ottiene dall’umile espettorato che passa di bocca in bocca – la sospensione dell’economia mondiale – iniziamo a immaginarlo attraverso i nostri piccoli gesti insignificanti fatti, anche loro, uno per uno: vale a dire la sospensione del sistema di produzione. Ponendoci questo tipo di domande, ognuno di noi inizia a immaginare gesti-barriera non solo contro il virus, ma contro ogni elemento di un modo di produzione che non vogliamo riprendere.

Questo perché non si tratta più di riprendere o influenzare un sistema di produzione, ma di abbandonare la produzione come unico principio di relazione al mondo⁴. Non è una questione di rivoluzione, ma di dissoluzione, pixel dopo pixel. Come dimostrato da Pierre Charbonnier⁵, dopo cento anni di socialismo limitato alla semplice ridistribuzione dei benefici dell’economia, potrebbe essere il momento di inventare un socialismo che sfida la produzione stessa. L’ingiustizia non si limita solo alla ridistribuzione dei frutti del progresso, ma al modo stesso di far fruttare il pianeta. Ciò non significa decrescere o vivere a pane e acqua, ma imparare a selezionare ogni segmento di questo famoso sistema falsamente irreversibile, a mettere in discussione ciascuna delle connessioni date per essenziali, e a sperimentare passo dopo passo ciò che è desiderabile e ciò che ha smesso di esserlo.

Da qui l’importanza capitale di usare questo tempo di confinamento imposto per descrivere, prima da soli, poi in gruppo, quello a cui siamo legati; quello da cui siamo pronti a liberarci; i canali che siamo pronti a ricostituire e quelli che, con il nostro comportamento, siamo determinati a interrompere. I globalizzatori, loro, sembrano avere un’idea molto chiara di ciò che vogliono veder rinascere dopo la ripresa: la stessa cosa in peggio, industrie petrolifere e navi da crociera giganti in primis. Sta a noi opporci a loro con un contro-inventario. Se tra un mese o due, miliardi di umani saranno in grado, al volo, di imparare la nuova “distanza sociale”, di allontanarsi per essere più solidali, di stare a casa per non ingombrare gli ospedali, possiamo immaginare bene il potenziale di trasformazione di questi nuovi gesti-barriera eretti contro il ritorno all’identico, o peggio, contro un nuovo attacco violento di coloro che vogliono fuggire per sempre dall’attrazione terrestre.

Uno strumento per aiutarci a scegliere

Dato che è sempre bene accompagnare un’argomentazione con degli esercizi pratici, proponiamo ai lettori di provare a rispondere a questa piccola lista di domande. Questa sarà tanto più utile quanto più si appoggerà su un’esperienza personale direttamente vissuta. Non si tratta di esprimere le prime opinioni che vi vengono in mente, ma piuttosto di descrivere una situazione e se possibile di trasformarla in una breve inchiesta. Solo in seguito, se vi date il tempo di combinare le risposte in un paesaggio caratterizzato dalla sovrapposizione delle descrizioni, riuscirete a giungere ad un’espressione politica incarnata e concreta – non prima.

Attenzione: non si tratta di un questionario e non è nemmeno un sondaggio. È un supporto all’autodescrizione⁶.

Si tratta di fare la lista delle attività di cui vi siete sentiti privati dalla crisi attuale e che vi danno la sensazione di una violazione delle vostre condizioni essenziali di sussistenza. Per ogni attività, indicate se preferireste che riprendessero come prima, meglio di prima, o che non riprendano proprio. Rispondete a queste domande:

Quali sono le attività attualmente sospese che vorreste non ricominciassero più?
Descrivete a) perché queste attività vi sembrano nocive/superflue/pericolose/incoerenti; b) in che misura la loro scomparsa/messa in attesa/sostituzione renderebbe più facile/più coerente svolgere altre attività che preferireste? (Scrivete un paragrafo distinto per ciascuna delle risposte alla domanda 1).
Quali misure prevedereste per facilitare la transizione ad altre attività per tutti gli operai/impiegati/agenti/imprenditori che non potrebbero più continuare le attività soppresse?
Quali sono le attività attualmente sospese che vorreste vedere svilupparsi/ricominciare, o ancora quali attività sostitutive dovrebbero essere inventate?
Descrivete a) perché queste attività vi sembrano positive; b) come rendono più semplici/armoniche/coerenti altre attività che vi piacerebbero e c) che permettono di contrastare quelle che giudicate sfavorevoli? (Scrivete un paragrafo distinto per ciascuna delle risposte alla domanda 4).
Quali misure adottereste per aiutare operai/impiegati/agenti/imprenditori ad acquisire le capacità/i metodi/i guadagni/gli strumenti per permettere la ripresa/lo sviluppo/la nascita di queste attività?

Trovate infine un modo per comparare la vostra descrizione con quella di altri partecipanti. La stesura e la sovrapposizione delle risposte dovrebbero disegnare via via un paesaggio composto da linee conflittuali, alleanze, controversie e contrasti.

Stoller, Mark. 2020. “The Coronavirus Relief Bill Could Turn into a Corporate Coup If We Aren’t Careful.” The Guardian, 24 marzo 2020. ↩
Latour, Bruno. 2019. “Nous ne vivons pas sur la même planète.” AOC, 18 dicembre 2019. ↩
Danowski, Deborah, e Eduardo Viveiros de Castro. 2014. “L’arrêt de monde.” In De l’univers clos au monde infini, a cura di Emilie Hache, 221–339. Parigi: Éditions Dehors. ↩
Kazic, Dusan. 2019. “Plantes animées: De la production aux relations avec les plantes.” Tesi di dottorato, AgroParisTech. ↩
Charbonnier, Pierre. 2020. Abondance et liberté: Une histoire environnementale des idées politiques. Parigi: La Découverte. ↩
Latour, Bruno. 2018. Tracciare la rotta: Come orientarsi in politica. Milano: Raffaello Cortina. L’autodescrizione riprende la procedura dei nuovi cahiers de doléance, qui introdotta e approfondita da un gruppo di artisti e ricercatori. ↩

Mapping as a Contemporary Instrument for Orientation in Conferences

2020-01-01T23:59:01+00:00

This article presents a case study analyzing submissions from the Digital Humanities 2019 conference by visualizing a network of authors situated according to their shared lexicon. This new form of summarizing a conference is an effective way to grasp the whole conference at once. The hope is that this method of visualization will not be employed merely as a retroactive way to reflect on past events, but rather as an instrument to prepare the visit and orientate the attendees during the conference.

The Hermeneutic Circle of Data Visualization: The Case Study of the Affinity Map

2020-01-01T23:59:00+00:00

In this article, we show how postphenomenology can be used to analyze the Affinity Map: a data visualization that reveals the hidden dynamics that exist between individuals within large organizations. We make use of the Affinity Map to expand the classic postphenomenology that privileges a ‘linear’ understanding of technological mediations and introduce the notions of ‘iterativity’ and ‘collectivity.’ In the first section of the paper, we discuss both classic and more recent descriptions of humantechnology-world relations in order to transcendentally approach the discipline of data visualization. In the second section, we use the Affinity Map case study to consider three elements: 1) the collection of data and the design process; 2) the visual grammar of the data visualization, and 3) the process of self-recognition for the map ‘reader.’ In the third section, we introduce the hermeneutic circle of data visualization. Finally, we suggest that the Affinity Map, because of its ethical and political multistability, might be seen as a material encounter between postphenomenology, actor-network theory (ANT), and hermeneutics.

Self-Recognition in Data Visualization: How Individuals See Themselves in Visual Representations

2019-01-01T23:59:00+00:00

This article explores how readers recognize their personal identities represented through data visualizations. The recognition is investigated starting from three definitions captured by the philosopher Paul Ricœur : the identification with the visualization, the recognition of someone in the visualization, and the mutual recognition that happens between readers. Whereas these notions were initially applied to study the role of the book reader, two further concepts complete the shift to data visualization : the digital identity stays for the present-day passport of human actions and the promise is the intimate reflection that projects readers towards their own future. This article reflects on the delicate meaning of digital identity and the way of representing it according to this structure : From Personal Identity to Media is a historical introduction to self-recognition, Data Visualization for Representing Identities moves the focus to visual representation, and The Course of Recognition breaks the self-recognition in through the five concepts above, just before the conclusion.

Mapping Affinities in Academic Organizations

2018-01-01T23:59:01+00:00

Scholarly affinities are one of the most fundamental hidden dynamics that drive scientific development. Some affinities are actual, and consequently can be measured through classical academic metrics such as co-authoring. Other affinities are potential, and therefore do not leave visible traces in information systems; for instance, some peers may share interests without actually knowing it. This article illustrates the development of a map of affinities for academic collectives, designed to be relevant to three audiences: the management, the scholars themselves, and the external public. Our case study involves the School of Architecture, Civil and Environmental Engineering of EPFL, hereinafter ENAC. The school consists of around 1,000 scholars, 70 laboratories, and 3 institutes. The actual affinities are modeled using the data available from the information systems reporting publications, teaching, and advising scholars, whereas the potential affinities are addressed through text mining of the publications. The major challenge for designing such a map is to represent the multi-dimensionality and multi-scale nature of the information. The affinities are not limited to the computation of heterogeneous sources of information; they also apply at different scales. The map, thus, shows local affinities inside a given laboratory, as well as global affinities among laboratories. This article presents a graphical grammar to represent affinities. Its effectiveness is illustrated by two actualizations of the design proposal: an interactive online system in which the map can be parameterized, and a large-scale carpet of 250 square meters. In both cases, we discuss how the materiality influences the representation of data, in particular the way key questions could be appropriately addressed considering the three target audiences: the insights gained by the management and their consequences in terms of governance, the understanding of the scholars’ own positioning in the academic group in order to foster opportunities for new collaborations and, eventually, the interpretation of the structure from a general public to evaluate the relevance of the tool for external communication.

Printing Walkable Visualizations

2018-01-01T23:59:00+00:00

This article concerns a specific actor in the actualization process, the media. The conventional media for visualizations is the computer screen, a visual device that supports the practices of design and reading. However, visualizations also appear in other ways, for example as posters, articles, books, or projections. This article focuses, in particular, on a pretty unusual medium called floor or walkable visualization.

The Analogous City, the Map

2015-01-01T23:59:00+00:00

This new publication of The Analogous City, an artwork produced by Aldo Rossi, Eraldo Consolascio, Bruno Reichlin and Fabio Reinhart for the Venice Biennale of Architecture in 1976, is part of a museographic installation for the exhibition Aldo Rossi – The Window of the Poet (Celant and Huijts 2015) at the Bonnefanten Museum in Maastricht. To gauge and explore this seminal work, Archizoom relied on Dario Rodighiero, candidate in the Doctoral Programme for Architecture and Sciences of the Cities, and designer at the Digital Humanities Lab (DHLAB) at EPFL. Conceived as a genuine urban project, The Analogous City displays an aggregation of architectures drawn from collective and personal memories. What happens if we isolate the forms that Aldo Rossi and his friends so consciously placed in relation to each other? Rodighiero simply decomposed it into the original references and then returned the pieces to the artwork, thus allowing us to simultaneously see the work and its visual vocabulary. An application based on augmented reality has been created to work in tandem with this publication by displaying the complete references belonging to the collage on different layers suspended over the artwork. By downloading the free application and installing it on your tablet or mobile phone, you can recreate the museum experience whenever and wherever you are.

A Subtraction of Weight

The Analogous City is not simply a work that projects over a flat surface a continuous past; rather, it exists as a dialogue with the present and a comparison with one’s ideals and memory. I like to think of The Analogous City as a real alternative. That is how I became an inhabitant of this city and started to patiently reconstruct its history, from its neighborhoods to its streets, from its monuments to its houses. I have talked to the people who lived in this city and listened to the stories of authors, scholars of architecture and simple flâneurs.

I was inspired by Bernardo Secchi (2000), who identifies city planning as a combination of tracks: real practices that model the city, and shape it over time, at their overlapping. Secchi also states that there is another layer that intertwines with the practices and integrates the fragments left by them: the layer of discussions. After months of researching this work, if asked to reflect about it, I would say that The Analogous City is an urban fabric formed by tracks and conversations: tracks of all real and unreal objects that have become a sign on the map; discussions of all those authors, critics, historians, architects, who are equally inhabitants of this city.

This idea already existed, it takes shape in a building designed by Aldo Rossi himself, the Bonnefanten Museum. What better way to show the richness of this urban fabric to visitors – the new inhabitants of this city – than through a city map? An object so simple and recognizable that it does not require any instructions.

Although technologies have transformed the medium over the last few decades, a map preserves its role as an orientation tool that helps to identify an area’s landmarks and to become familiar with them. With a map in hand, one can set out and venture the streets, visit the neighborhoods, marvel at the unexpected sights that only a city can offer, and rediscover the pleasures of wandering, only to find, all of a sudden, one has arrived at destination.

This map has been developed as a tool to know The Analogous City. It is part of a museum installation within the exhibition Aldo Rossi – The Window of the Poet, Prints 1973–1997 (Celant and Huijts 2015) that will be opened in Maastricht next June.¹ This work originates from a cooperation between the Bonnefanten Museum and the École polytechnique fédérale de Lausanne (EPFL), represented namely by the Archizoom Gallery and the Laboratory of Digital Humanities.

The map consists of a simple sheet that is printed on both sides. One part shows a large format print of the work, with its references listed in chronological order; the other, the texts of Fabio Reinhart and Aldo Rossi, together with the list of references and related images, are shown in their entirety, just as they are printed in books.

Each process of establishing a historical identity requires a certain time and so far the fragments of The Analogous City had never been completely revealed. During this practice of reconstruction and deconstruction of the city, some important and hidden stories have surfaced again, those of its inhabitants: the people who have shaped the city and discussed it by exposing their ideas and stories. All of their stories are so interesting that it is a pity that they cannot all figure there, and that there has not been time to listen to all of them.

The references of The Analogous City have been extracted from existing books; a significant part of the archaeological work has been done in both traditional and digital libraries and archives. By identifying myself with the architects of the city, I have tried to recover all publications that the authors selected, photocopied, cut out and put carefully together in creating this collective collage, now kept in the basements of the Centre Pompidou in Paris (Figure 1).

Figure 1. The references identified during the research, returned to their exact positions within the original collage.

For this reason, all bibliographic references already existed as publications in the Spring of 1976, when The Analogous City was composed. Any references from a subsequent date are ascribable to the private archives of the authors.

I can truly find myself in the words of Aldo Rossi, who said that this latest project is particularly dear to me; it is an affection project.

I better understood the esteem among architects (not all) for Rossi’s thinking by talking to those who contributed to the work, as well as the effort of the authors who spent whole nights building the city, because as Fabio Reinhart recalls, during the day they had to work to make a living.

Aldo Rossi, Eraldo Consolascio, Bruno Reichlin and Fabio Reinhart worked for months at composing this work and, similarly, it took me some time to decipher it, find the images, digitize book pages, and sometimes my curiosity led me to read the texts in them, which gave the urban fabric color and depth.

In this way, this installation reveals itself as a modern tool, available to all of us, for pursuing our personal growth, and the time spent using this tool generates our affection towards it.

The work has required time and passion. Initially, references arrived quickly, but little by little as the puzzle was being completed, to recover a certain reference became increasingly difficult and took more and more time. For instance, the last reference was found thanks to the help of Beatrice Lampariello, and required months of research, in the spare moments that I had during my doctorate study at EPFL. Reconstructing the tracks required such a long time, in scattered moments, that even now I could not tell exactly how many tracks make up The Analogous City. I count 42.

The map is part of a digital installation within the exhibition. The installation consists of a table, on top of which The Analogous City has been reproduced and from which, with the use of augmented reality technology, it has been possible to extrude all references that compose it, so as to make them interactive: by framing the map with a tablet, the camera framing is displayed on the screen, enriched with virtual references. These elements are the signs that form The Analogous City. Superimposed upon the references of the plan, they fluctuate and invite the viewer to discover and explore the collage of The Analogous City, by means of a deconstruction (Figure 2).

Figure 2. The entry point of the museum's augmented-reality application, where the references appear as floating elements above the artwork.

As a publication, the map offers a similar experience as the installation exhibited at the museum: by downloading the application, it is possible to interact with the map, which replicates at home what visitors have experienced at the museum.

As Italo Calvino (1988) would say, my work has involved the subtraction of weight: the references of The Analogous City were freed, disconnected from the work and made to fluctuate. This operation has revealed new relations between references, thereby creating new meanings: new conversations that have their foundations in the tracks of the city, which reminds us that The Analogous City is not a static work, but rather a lively, dynamic city.

Finally the work loses the typical vertical orientation of a painting, to become a city map and meeting place for the museum installation. The intention has been to generate a new space that is also a meeting point, like a square, and then to enrich that place with personal discussions, stories, and people, because we should not forget that, as Rossi (1981) reminded in his A Scientific Autobiography, “the project was only a pretext for a general involvement”.

References

Calvino, Italo. 1988. Six Memos for the Next Millennium. Cambridge, MA: Harvard University Press.
Celant, Germano, and Stijn Huijts, eds. 2015. Aldo Rossi: Opera Grafica: Etchings Lithographs Silkscreen Prints. Cinisello Balsamo: Silvana Editoriale.
Rossi, Aldo. 1981. A Scientific Autobiography. Translated by Lawrence Venuti. Cambridge, MA: MIT Press.
Secchi, Bernardo. 2000. Prima lezione di urbanistica. Roma-Bari: Laterza.

The exhibition opened on 25 June 2015 at the Bonnefantenmuseum in Maastricht and ran through 15 November 2015, before traveling to the École polytechnique fédérale de Lausanne (Archizoom) and the Galleria d’Arte Moderna e Contemporanea di Bergamo (GAMeC). It brought together one hundred prints from the museum’s own collection alongside forty drawings and paintings from private collections, complemented by printing plates and proofs. ↩