Lev Manovich' Lecture @ Centre Pompidou (Paris)

Conference by Lev Manovich and reading by Olivier Cadiot

Mapping Time: How Big Data and Visualization Makes Visible Evolution of Cultural Artifacts. 
When: Friday, June 15 2012, 8pm 
Where: Centre Pompidou, Petite Salle
web: http://www.ircam.fr/transmission.html?event=1119&L=1

In 2007 we created Software Studies Initiative (www.softwarestudies.com) at University of California, San Diego (UCSD) to develop techniques and software tools that will enable humanists and social scientists work with large visual data sets. We call our approach "cultural analytics." In my talk I will show how we use cultural analytics techniques to study temporal patterns in sets of cultural artifacts. The examples include including one million manga pages, all paintings by Vincent van Gogh, films by Dziga Vertov, 4535 cover of Times magazine (1923-2009), and 20,000 pages of Science (1880-) and Popular Science (1872-) magazines. Use of visualization allows us for the first time to see the "shapes" of cultural time. Each visualization reveals the unexpected and intricate patterns of temporal change in a particular artifact - or our experience of these artifacts. Taken together, they demonstrate how we can visualize different kinds of gradual changes over time at a number of scales, ranging from a few seconds of an animated film to dozens of years of magazine and newspaper publication. 

Lev Manovich (http://manovich.net) is a professor at the Visual Arts Department, University of California - San Diego where he teaches courses in digital humanities, visualization, digital art, and new media theory. 

Reading by Olivier Cadiot 
Using the examples from experiments over the years - Le colonel des zouaves (1997),Retour définitif et durable de l'être aimé (2003), and Un mage en été (2010) - the subject of this lecture is to discuss the work of writing for theater when carried out in connection with IRCAM's technological context as well as the relationship between the elements of style and temporality in the performance. In 1993 Olivier Cadiot made his acquaintance with the theater via Ludovic Lagarde, beginning a long questioning of writing for the theater with the complicity of the actor Laurent Poitrenaux.

veja.vis project @ Digital Humanities 2012

The project veja.vis will be presented @ the Digital Humanities Conference in Sheffield, UK, September 2012
From veja.vis
Description of the project:
The increasing capacity of computational data analysis is driving computer scientists and designers into the development of new features to visualize and understand cultural artifacts in a different manner. Social scientists, digital humanities researchers are investigating how to create what we can call "cultural algorithms" to discover or reveal new trends about a field of investigation that can be related to film studies, literature, communication and so on and so forth. Following the theoretical approach created by Lev Manovich about his studies related to "cultural analytics", this paper will present a one year research on the visualization of the entire collection of covers of the Veja magazine, considered the most important weekly magazine in Brazil. The visualization that we created in our Lab at the Federal University of Juiz de Fora (www.ufjf.br/sws) analyses and demonstrates practical uses for cultural visualization, since we can have critical analytical details about all the covers such as the gender that is more frequent in the covers (masculine), the colors that the magazine uses regularly etc. We can also use image recognition algorithms so we can cross data with wikipedia, for example, and discover who was more frequently featured in the covers: e.g. politicians or media stars.
The project is coordinated by Cicero Silva and Marcio Santos. Images by Marcio Santos.

From veja.vis

From veja.vis

Video with the entire collection of Veja covers.

Manovich' seminar and public lecture at Bruno Latour's Media Lab | SciencesPo

medialab | Sciences Po
June 12

Public lecture:

How to see one million images?

The explosive growth of cultural content on the web including social
media, and the digitization by museums, libraries, and other agencies
opened up fundamentally new possibilities for the studies of both
contemporary and historical cultures. But how we navigate massive
visual collections of user-generated content which may contain
billions of images? What new theoretical concepts do we need to deal
with the new scale of born-digital culture? How do we use data mining
of massive cultural data sets to question everything we know about
culture? In 2007 we have established Software Studies Initiative
(softwarestudies.com) at University of California, San Diego to begin
working on these questions. I will show a number of our projects
highlighting how visualization allows us to see patterns in cultural
data which were not visible before. The examples include analysis of
art, photography, film, animation, motion graphics, video games,
magazines, and other visual media, including 1 million pages of manga
(Japanese comics) pages and 1 million images from deviantArt (largest
social network for non-professional art).


Visualization as the New Language of theory

Drawing on the practical projects done in our lab softwarestudies.com,
I will discuss how computational analysis and visualization of big
cultural data sets leads us to question traditional discrete
categories used for cultural categorization (such as "style" and
"period."). But while computers can keep track of million of points
without the need of such categories, how do we resist our conventional
urge to use language to divide the world into sharp boundaries and
give them names? How can we learn from software to think differently?
Can visualization provide the new language of theory?