Search

In this paper, we devise a machine learning protocol to tackle a complex sociological task: to create a research sample from a few examples of interest, but in the absences of a clear definition of the target subset. As an example, we create a sample of organisational leaders starting from a list of nominees for the Bureau of the Intergovernmental Panel on Climate Change. The difficulty in this task lies in the impossibility to spell out the characteristics that define leadership in a complex and highly distributed organization like the IPCC. To bypass this lack of explicit definition, we use a series of techniques for anomaly detection to identify IPCC contributors with profiles similar to official Bureau nominees. We find that we can construct a precise (albeit implicit) model of IPCC leadership despite its social and political complexity, and that we can usefully use this model to expand our initial sample.

This article examines the organisation of collaborative digital methods and data projects in the context of engaged research-led teaching in the humanities. Drawing on interviews, field notes, projects and practices from across eight research groups associated with the Public Data Lab (publicdatalab. org), it provides considerations for those interested in undertaking such projects, organised around four areas: composing (1) problems and questions;(2) collectives of inquiry;(3) learning devices and infrastructures; and (4) vernacular, boundary and experimental outputs. Informed by constructivist approaches to learning and pragmatist approaches to collective inquiry, these considerations aim to support teaching and learning through digital projects which surface and reflect on the questions, problems, formats, data, methods, materials and means through which they are produced.

“Controversy Mapping shows how we can use social research to bring controversies back to the surface of knowledge and public life, and how it can help to recover the power of controversy to transform what's possible. The book provides everything you need – the ideas, examples, and techniques – to start doing controversy analysis.”Noortje Marres, University of Warwick“Venturini and Munk have produced a significant book that traces the genealogy of controversy mapping back to its origins in actor-network theory to its incarnations in digital methods. Through a lucid and engaging narrative and series of visualizations, they provide a comprehensive ‘field guide’ to the major figures, theories, concepts, and methods that make up the practices of controversy mapping.”Evelyn Ruppert, Goldsmiths, University of London As disputes concerning the environment, the economy, and pandemics occupy public debate, we need to learn to navigate matters of public concern when facts are in doubt and expertise is contested.Controversy Mapping is the first book to introduce readers to the observation and representation of contested issues on digital media. Drawing on actor-network theory and digital methods, Venturini and Munk outline the conceptual underpinnings and the many tools and techniques of controversy mapping. They review its history in science and technology studies, discuss its methodological potential, and unfold its political implications. Through a range of cases and examples, they demonstrate how to chart actors and issues using digital fieldwork and computational techniques. A preface by Richard Rogers and an interview with Bruno Latour are also included. A crucial field guide and hands-on companion for the digital age, Controversy Mapping is an indispensable resource for students and scholars of media and communication, as well as activists, journalists, citizens, and decision makers.

Supplemental material, sj-pdf-1-bds-10.1177_20539517211018488 for What do we see when we look at networks: Visual network analysis, relational ambiguity, and force-directed layouts by Tommaso Venturini, Mathieu Jacomy and Pablo Jensen in Big Data & Society

Supplementary information (PDF 1.5 MB)

Preprint Version; International audience; It is increasingly common in natural and social sciences to rely on network visualizations to explore relational datasets and illustrate findings. Such practices have been around long enough to prove that scholars find it useful to project networks in a two-dimensional space and to use their visual qualities as proxies for their topological features. Yet these practices remain based on intuition, and the foundations and limits of this type of exploration are still implicit. To fill this lack of formalization, this paper offers explicit documentation for the kind of visual network analysis encouraged by force-directed layouts. Using the example of a network of Jazz performers, band and record labels extracted from Wikipedia, the paper provides guidelines on how to make networks readable and how to interpret their visual features. It discusses how the inherent ambiguity of network visualizations can be exploited for exploratory data analysis. Acknowledging that vagueness is a feature of many relational datasets in the humanities and social sciences, the paper contends that visual ambiguity, if properly interpreted, can be an asset for the analysis. Finally, we propose two attempts to distinguish the ambiguity inherited from the represented phenomenon from the distortions coming from fitting a multidimensional object in a two-dimensional space. We discuss why these attempts are only partially successful, and we propose further steps towards a metric of spatialization quality.

International audience; API-based research is an approach to computational social sciences and digital sociology based on the extraction of records from the datasets made available by online platforms through their application programming interfaces (or APIs). This type of research has allowed the collection of detailed information on large populations, thereby effectively challenging the distinction between qualitative and quantitative methods. Lately, access to this source of data has been significantly reduced after a series of scandals connected to the use of personal information collected through online platforms. The closure of APIs, however, can have positive effects, if it encourages researchers to reduce their dependence on mainstream platforms and explore new sources and ways to collect records on online interactions that are closer to the digital fieldwork.

Over the past two decades STS scholars have maintained a critical interest in, and displayed a practical engagement with, methods that has allowed us to work with wider publics and/or in broader interdisciplinary contexts. Through practices like participatory modelling (Whatmore et al. 2011), deliberative mapping (Burgess et al. 2007), consensus conferencing (Callon et al. 2009, Horst & Irwin 2010), and various other kinds of collective and user-driven approaches to design and innovation (Woolgar 1990, Hyysaolo 2006, Ehn 2008, Jensen 2012) STS research has fruitfully experimented with, and critically considered the effects of, different modes of intervention. It is a research agenda that coincides with, but has so far not been particularly related to, the rise in digital STS scholarship. This is perhaps surprising considering that broad strands of the interest in digital methods has been driven by a desire to provide navigational aids to actors faced with the challenge of making sense of complicated techno-scientific problems. Natively digital media technologies have thus been re-appropriated by STS researchers specifically for the purpose of mapping controversies in a way that would deploy them and render them more accessible to their concerned parties (Marres & Rogers 2005, Venturini 2010, Yaneva 2012). So far, however, very few concrete experiments have attempted to road test these cartographic ambitions in a way that would put them into direct conversation with our existing experiences with various forms of public engagement and participation.Through a concrete reappropriation of a collaborative format that is indeed native to the digital domain - namely the hackathon - we will show how digital methods can make a difference in participatory STS research. The data sprint, as we have dubbed our reappropriated version of the hackathon, has been tested in a series of four weeklong sessionps where we invited issue experts from the climate change adaptation community to collaboratively map their matters of concern together with a team of STS researchers, digital methods experts, developers and data designers. The sprints took place under the auspices of the E-MAPS project (preliminary results are available here: http://www.emapsproject.com/blog/) and drew on prior experiences from the MACOSPOL project and the winter and summer schools organised by the Digital Methods Initiative in amsterdam. Through a mix of digital methods ranging from web cartography and text mining to scientometrics and social media analysis we took on questions related to climate adaptation funding, vulnerability assessment, project management, and dynamics of the international negotiations. The sprints hardwired the challenges of collaboratively formulating research questions and devising potential avenues of intervention together with concerned group of actors into the practical circumstances of working with (i.e. harvesting, analysing and visualising) digital traces. It thus offered a laboratory in which it became possible to test how the potentials with which digital methods in general, and the cartography of controversies in particular, have been repeatedly invested may or may not be realised.

International audience; Dans ce chapitre, pour des raisons que j'essaierai de clarifier, je parlerai moins des fausses nouvelles que de l'étude des fausses nouvelles. Je ferai de mon mieux pour établir une distinction nette entre les deux, car elle me semble essentielle pour éviter le risque de se perdre dans l'imbroglio de cet objet d'étude bien épineux. Ce risque concerne le débat public dans les démocraties modernes, mais également (et très distinctement) les chercheurs qui étudient les fausses nouvelles. Personnellement, j'ai une relation amour-haine avec les fausses nouvelles. Bien que être qualifie d'expert en la matière me met très mal à l'aise, je n'ai jamais été aussi populaire qu'après avoir commencé à travailler sur ce sujet. Lorsqu'en mars 2017 Liliana Bounegru, Jonathan Gray et moi-même avons lancé le Fake News Field Guide en tant que premier projet du tout nouveau Public Data Lab (je reviendrai plus tard sur cette initiative), nous ne pouvions imaginer que nous serions en mesure d'attirer autant d'attention. En quelques mois, nous avons été contacté par des dizaines de chercheurs et professionnels du monde entier. Plusieurs organisations nous ont invités à utiliser leurs bases de données et leurs technologies et des dizaines de journalistes de différents médias nous ont demandé des entretiens. J'ai souvent entendu des artistes se de se plaindre d'être connus davantage pour leurs oeuvres les plus populaires que pour leurs meilleurs contributions. Je sais maintenant que ce sentiment existe également chez les chercheurs. Cinq raisons pour NE PAS parler des fausses nouvelles Je ne veux pas paraître snob : j'apprécie la réputation d'expert. Je crois simplement que cette réputation n'est pas méritée en ce qui concerne les fausses nouvelles. Je n'essaie pas non plus d'être modeste. Ce n'est pas seulement moi ; personne, à mon avis, ne peut honnêtement prétendre être un « expert des fausses nouvelles », pour la simple raison qu'il n'existe aucun moyen sérieux de délimiter un tel champ d'expertise. Au contraire, les chercheurs qui travaillent sur le sujet s'accordent pour dire qu'on devrait arrêter d'utiliser cette notion et commencer à la dénoncer activement.

This paper discusses the differences and affinities among three types of networks (namely Actor-Networks, Social Networks and Digital Networks) that are playing an increasingly important role in digital STS. In the last few decades, the notion of networks has slowly but steadily struck root across broad strands of STS research. It started with the advent of actor-network theory, which provided a convenient instrument to describe the construction work of socio-technical phenomena. Then came network analysis, and scholars who imported into STS the techniques of investigation and visualization developed in the tradition of social network analysis and scientometrics. Finally, with the increasing ‘computerization’ of STS, scholars turned their attention to digital networks as a way of tracing collective life. Many researchers have more or less explicitly tried to link these three movements in one coherent set of digital methods, betting on the idea that actor-network theory can be operationalized through network analysis thanks to the data provided by digital networks. Yet, to be honest, the affinity between these three objects is sketchy at best. Besides the homonym ‘network’, there is little to is little to show for it. Are we sure that we are talking about the same thing?

International audience; ‘Fake news’ is a key subject of data politics, but also a tricky a one. As this chapter aims to show, various phenomena signified by this misleading label have little in common, except being opposite to the kind of algorithmic intelligence that most other chapters present as the main concern of data politics. This does not mean that ‘fake news’ is not related to computational analytics or political intentions, but it does mean that this relation is not straightforward.To discuss this relation, I will go through a three-stage argument. First, I will criticise the notion of ‘fake news’, dismissing the idea that this type of misinformation can be defined by its relationship to truth. Second, I will propose a different definition of this phenomenon based on its circulation rather than of its contents. Third, I will reintroduce the connection to data politics, by describing the economic, communicational, technological, cultural and political dimensions of junk news.

Resumo Este artigo discute diferenças e afinidades entre três tipos de rede (Ator-Rede, Análise de Redes e Redes Digitais) importantes para o Digital STS.2 Nas últimas décadas, a colonização de pesquisas em STS foi lenta e gradual. Ela se inicia com a Teoria Ator-Rede (TAR), que oferece um conjunto de noções para descrever a construção de fenômenos sociotécnicos. Com o advento da Análise de Redes, estudiosos incorporam técnicas de investigação e visualização desenvolvidas pela Análise de Redes Sociais (ARS) e pela Cientometria aos estudos em STS. Com o crescente uso de recursos computacionais pelos STS, estudiosos atentam para as Redes Digitais como modo de rastrear a vida coletiva. Muitos pesquisadores tentaram relacionar esses três movimentos aos métodos digitais, ao apostarem que a TAR pode ser operacionalizada por meio da Análise de Redes, graças aos dados providos pelas Redes Digitais. No entanto, além da homonímia presente na palavra “rede”, que caracteriza as três abordagens mencionadas, há poucas evidências que comprovam a continuidade entre esses três tipos de rede. Falamos das mesmas redes? Abstract This paper discusses the differences and affinities among three types of networks (Actor-Networks, Social Networks and Digital Networks) that are playing an increasingly important role in digital STS. In the last few decades, the notion of networks has slowly but steadily struck root across broad strands of STS research. It started with the advent of actor-network theory, which provided a convenient instrument to describe the construction work of socio-technical phenomena. Then came network analysis, and scholars who imported into STS the techniques of investigation and visualization developed in the tradition of social network analysis and scientometrics. Finally, with the increasing ‘computerization’ of STS, scholars turned their attention to digital networks as a way of tracing collective life. Many researchers have more or less explicitly tried to link these three movements in one coherent set of digital methods, betting on the idea that actor-network theory can be operationalized through network analysis, thanks to the data provided by digital networks. Yet, to be honest, there is little evidence proving the continuity among these three objects besides the homonymy of the word ‘network’. Are we sure that we are talking about the same thing?

Neste artigo, argumentamos que a nova disponibilidadede conjuntos de dados digitais nos permite revisitar a teoriasocial de Gabriel Tarde (1843-1904), de maneira adispensar inteiramente o uso de noções como "indivíduo"ou "sociedade". Nosso argumento é que, quando era impossível,complicado ou simplesmente lento montar e navegaratravés da massa de informações sobre itens específicos,fazia sentido tratar dados sobre conexões sociaiscom a definição de dois níveis: um para o elemento, outropara os agregados. Porém, uma vez que nós temos a experiênciade seguir os indivíduos através de suas conexões(que é frequentemente o caso com os perfis) poderia sermais gratificante começar a navegar pelos conjuntos dedados sem fazer distinção entre o nível do componenteindividual e o da estrutura agregada. Torna-se possíveldar alguma credibilidade a estranha ideia de "mônadas",de Tarde. Nós afirmamos que esse tipo de prática de navegaçãotornou-se possível somente agora pelas bases dedados disponíveis digitalmente, e que tal prática poderiamodificar a teoria social se pudéssemos visualizar essenovo tipo de exploração de uma forma coerente.Palavras-chave: Teoria Social; Gabriel Tarde; teoria ator--rede; métodos digitais; visualização de dados

Neste artigo, argumentamos que a nova disponibilidade de conjuntos de dados digitais nos permite revisitar a te- oria social de Gabriel Tarde (1843-1904), de maneira a dispensar inteiramente o uso de noções como “indivíduo” ou “sociedade”. Nosso argumento é que, quando era im- possível, complicado ou simplesmente lento montar e na- vegar através da massa de informações sobre itens espe- cíficos, fazia sentido tratar dados sobre conexões sociais com a definição de dois níveis: um para o elemento, outro para os agregados. Porém, uma vez que nós temos a expe- riência de seguir os indivíduos através de suas conexões (que é frequentemente o caso com os perfis) poderia ser mais gratificante começar a navegar pelos conjuntos de dados sem fazer distinção entre o nível do componente individual e o da estrutura agregada. Torna-se possível dar alguma credibilidade a estranha ideia de “mônadas”, de Tarde. Nós afirmamos que esse tipo de prática de na- vegação tornou-se possível somente agora pelas bases de dados disponíveis digitalmente, e que tal prática poderia modificar a teoria social se pudéssemos visualizar esse novo tipo de exploração de uma forma coerente. Palavras-chave: Teoria Social; Gabriel Tarde; teoria ator- -rede; métodos digitais; visualização de dados

In the last few years, a spirit has been haunting our academic and popular culture — the spirit of networks. Throughout social as well as natural sciences, more and more phenomena have come to be conceived as networks. Telecommunication networks, neural networks, social networks, epigenetic networks, ecological and economic networks , the very fabric of our existence seems to be made of lines and dots. More recently, the interest for graphs overflowed from science to popular culture and images of networks started to appear everywhere. They decorate buildings and objects; they are printed on t-shirts and furniture; they colonize the desktop of our computers and the walls of our airports. Networks have become the emblem of modernity, a way to show and tell our world’s complexity. Our growing fascination for networks is not unjustified. Networks are powerful conceptual tools, encapsulating in a single object multiple affordances for computation (networks as graphs), visualization (networks as maps) and manipulation of data (networks as interfaces).

In this paper we argue that the new availability of digital data sets allows one to revisit Gabriel Tarde’s (1843-1904) social theory that entirely dispensed with using notions such as individual or society. Our argument is that when it was impossible, cumbersome or simply slow to assemble and to navigate through the masses of information on particular items, it made sense to treat data about social connections by defining two levels: one for the element, the other for the aggregates. But once we have the experience of following individuals through their connections (which is often the case with profiles) it might be more rewarding to begin navigating datasets without making the distinction between the level of individual component and that of aggregated structure. It becomes possible to give some credibility to Tarde’s strange notion of ‘monads’. We claim that it is just this sort of navigational practice that is now made possible by digitally available databases and that such a practice could modify social theory if we could visualize this new type of exploration in a coherent way. Cet article vise à démonter que la disponibilité récente de données numériques permet de revisiter la théorie sociale de Gabriel Tarde (1843-1904) qui se passe totalement de notions telles que l’individu ou la société. Notre analyse repose sur l’idée que, tant qu’il était impossible, difficile ou simplement fastidieux d’amasser et de consulter des quantités d’informations liées à des sujets précis, il était logique de traiter les données relatives aux liens sociaux en définissant deux niveaux : un pour l’élément individuel, l’autre pour l’agrégat collectif. Mais dès que l’on suit les individus par le biais de leurs relations (ce que l’on fait généralement dans le cas des profils) il serait plus avantageux de naviguer à travers les fichiers de données sans distinguer le niveau de l’élément individuel de celui de la structure collective. On peut alors accorder une certaine crédibilité à l’étrange notion de « monades » de Tarde. Nous affirmons que c’est justement ce mode de navigation, rendu possible par l’accès aux bases de données numériques, qui permet de modifier la théorie sociologique. Au sens strict du terme, nous ne devrions plus parler de phénomènes collectifs par opposition à des phénomènes individuels, mais seulement d’autant de façons différentes de collecter des phénomènes. British Journal of Sociology Prize 2014.

Le médialab de Sciences Po est un laboratoire de moyens numériques créé, en mai 2009, pour mettre la révolution numérique au cøe ur d'une des plus des plus anciennes et prestigieuses communautés des sciences sociales françaises. Un goulot d'étranglement, pourtant, empêche encore l'équipe du médialab de profiter pleinement de la source de données la plus important de l'âge numérique : le World Wide Web. À ce jour, aucun instrument n'existe pour permettre aux chercheurs en sciences sociales de sélectionner, extraire et archiver un corpus d'informations en ligne. L'objectif du projet HCI (Hypertext Corpus Initiative) au sein du médialab est de développer cet instrument pour le mettre au service non seulement de l'équipe du médialab et des chercheurs de Sciences Po, mais aussi de tout chercheur intéressé par l'étude du Web. En leur offrant des outils pour cartographier le Web, HCI ambitionne de renouveler l'usage des données numériques dans les sciences sociales.