new video of Cultural Analytics software @HIPerSpace (287 megapixels)



We have created a new video which showcases some of the capabilities of Cultural Analytics software running on HIPerSpace (a supervisualization system with the resolution of 35,840 x 8,000 pixels; 286,720,000 pixels total.)

The software is being developed by Software Studies Initiative and GRAVITY lab (Graphics, Visualization and Virtual Reality Laboratory) at Calit2. The development is supported by Interdisciplinary Collaboratory Grant from UCSD Chancellor office (2008-2010).

At present software allows a user to load up to 4000 images of ANY size on the HIPerSpace wall along with accompanying metadata (which can include manually entered annotations, automatic measurements of visual properties, and other information). The metadata format is a simple tab-delimited text file (so it can be prepared in Excel etc.) Using our interactive software, a user can sort image set and also make X, Y plots of any sets of metadata types. The plots can include the actual images placed over the plotted points; the size of these images can be changed at any time interactively by a user.

Because HIPerSpace is powered by a cluster of PCs, it allows for interactive performance impossible with current desktop computers or the web. Thousands of images can be scaled and graphed instantly. Therefore, HIPerSpace is a perfect platform to develop techniques for software-supported cultural analysis of large media sets.

The N^3 Report

Researchers:
Andrew Wilson | PhD student, Art History, UCSD
Andy Rice | PhD student, Communication, UCSD
Lara Bullock | PhD student, Art History, UCSD
Chris Head | MMFA student, Visual Art, UCSD
Tara Zepel | PhD student, Art History, UCSD

The NBC Nightly News is an American institution that has been broadcast nightly in its current format since August 1, 1970. Arguably, the Nightly News, alongside newspapers, has served as America's primary source of information dissemination over the past 50 years. Over this time, changes in technology, visual culture, and the market dynamics for news shows have led certain aspects of nightly news production to evolve in subtle ways.

The N^3 Report presents a meta-broadcast of the NBC Nightly News (1980-2008) unpacked by cultural analytics. Through techniques examining visual characteristics and technological shifts that both are and are not obvious to the human eye, the N^3 Report performs what could be called "producer measurement systems"* in search of visual patterns, trends, or variations and asks what defines the "look" of the news institution over the past 50 years. How and when did aesthetic changes take place? And what might be the "look" of the future?

* a reverse of audience measure systems such as Nielson Ratings


Introductory segments from each year within the specified range were collected from Vanderbilt's Television News Archive. Each segment, beginning with the program's opening and ending with the first stationary shot of the anchor, was analyzed for intensity, color distribution, graphic content and temporal patterns. We report the following:

1) An increase in intensity and brightness by year.





AVG Intensity Mean Brightness (NBC 1980-2008): This graph presents the average measurement of brightness for all shots in each of the NBC Nightly News segments measured. The data reveals a trend of increasing brightness over time.





AVG Mean/Median Gray Value (NBC 1980-2008); An average composite image, which essentially merges the moving images footage into a single frame, was created for each of the introductory segments from NBC Nightly News episodes ranging from 1980 to 2008. This graph presents the mean and median gray values in these composite images. The data indicates an increase in average brightness (on a 0-255 scale) over time.


2) A graphic inundation throughout the nineties and into the millennium that appears to have subsided in the last two years in favor of a renewed and stable focus on the anchor.




Each image in this visualization is created by adding all frames in a given news video on top of each other. Each image on the top row represents one year of the 1980s, the middle row corresponds to the 1990s, and the bottom the 2000s. Note how in the middle of the 1990s row, the visage of the anchorman is replaced by an image dominated by graphics and motion. The focus on the anchorman returns in 2007 and 2008.



3) A visualization of temporal patterns using montages of segmented frames.



Hi-Res Image (7MB)
Each block within the larger montage is an introductory segment montage sampled at 30 frames per
second. Each row shows a 5 year span with two shows per year.

Row 1+2 = 1980s
Row 3+4 = 1990s
Row 5+6 = 2000s


4) A high speed video montage of NBC Nightly News introductions over the past 30 years.



Our investigation explores potential correlations, which may then be further developed through further research. The goal of this project was to take a mainstream source of information, the NBC Nightly News, and analyze the aesthetic techniques used in tandem with the message and reputation that is attached to this specific news program. It is also useful for viewing cultural shifts in the news over time. We focused on introductions to the Nightly News broadcasts because of the relative level of producer control and the concise presentation of what is to follow. By using visualizations of the visual aspects of introductory segments from 1980-2008, we hope to broaden the understanding of this cultural mainstay in American culture and open new questions that, without such techniques, may be overlooked or not realized, but that could contribute greatly to the growth of visual humanities and cultural studies.




Cultural Analytics


Interface design for Cultural Analytics research environment.

The explosive growth of cultural content on the web including social media and the digitization efforts by museums, libraries, and companies since the 1990s make possible fundamentally new paradigm for the study of both contemporary and historical cultures. We can use computer-based techniques for quantitative analysis and interactive visualization already commonly employed in sciences to begin analyzing patterns in massive cultural data sets. To make an analogy with "visual analytics," "business analytics," and "web analytics," we call this new paradigm cultural analytics.

We believe that a systematic use of large-scale computational analysis and interactive visualization of cultural data sets and data streams will become a major trend in cultural criticism and culture industries in the coming decades. What will happen when humanists start using interactive visualizations as a standard tool in their work, the way many scientists do already? If slides made possible art history, and if a movie projector and video recorder enabled film studies, what new cultural disciplines may emerge out of the use of interactive visualization and data analysis of large cultural data sets?

The idea of Cultural Analytics was first presented by Lev Manovich in 2005. Software Studies Initiative founded at Calit2 in 2007 made possible to turn this vision into a research program. By drawing on the cutting-edge cyberinfrastructure and visualization research at Calit2 as well as world reputation of UCSD in digital arts and theory, we are able to develop a unique research agenda which complements other projects in digital humanities and "cyberscholarship":

  • while most projects in digital humanities deal with text, we focus on automatic analysis of visual and media cultures and artifacts: video games, visual art, media design, cinema, animation, AMV, machinema, photography, etc.;
  • we place emphasis on interactive visualization of large cultural data sets (as opposed to only analysis);
  • we are developing techniques for analysis and visualization of born digital content such as video games, web sites and social media.

References:

Software Studies Initiative. Cultural Analytics: vision (last update: 06/2009). [key 26.3 MB] [ppt 12.8 MB]

Software Studies Initiative. Cultural Analytics: case studies (last update: 06/2009). [key 32.4 MB]. [ppt 8.2 MB]


Lev Manovich (with contribution by Noah Wardrip-Fruin). Cultural Analytics: white paper (5/2007; latest update 11/2008): [doc 2.4 MB]


Lev Manovich. "Cultural Analytics: Visualing Cultural Patterns in the Era of “More Media.”(published in DOMUS, spring 2009.) [doc 44KB]

Lev Manovich. "How to Follow Global Digital Cultures, or Cultural Analytics for Beginners." Deep Search, ed. Felix Stalder and Konrad Becker. Transaction Publishers ( English version) and Studienverlag (German version), in press. [ doc 92 KB]

So Yamaoka, Lev Manovich, Jeremy Douglass, Falko Kuester. "Cultural Analytics on Ultra High-Resolution Displays." Paper submitted to ACM Multimedia 2009 conference.

Lev Manovich and Jeremy Douglass. "Visualizing Temporal Patterns in Visual and Interactive Media." Forthcoming in Visualising the 21st Century, ed. Oliver Grau. MIT Press. [pdf 5 MB]

PUBLICATIONS

Books

Lev Manovich. Software Takes Command. Forthcoming from The MIT Press in 2010 in Software Studies Series . Italian translation forthcoming in 2009.

Noah Wardrip-Fruin. Expressive Processing: Digital Fictions, Computer Games, and Software Studies. The MIT Press: forthcoming September 2009.

Articles, book chapters and white papers

Jeremy Douglass. “Computer Visions of Computer Games: analysis and visualization of play recordings.” Workshop on Media Arts, Science, and Technology (MAST) 2009: The Future of Interactive Media. UC Santa Barbara, January 2009. [slides]

Jeremy Douglass. "Playpower: Radically Affordable Computer-Aided Learning with $12 TV-Computers.” co-authored with Derek Lomas and Daniel Rehn. Meaningful Play 2008. Michigan State U., October 2008.

William Huber. "Epic spatialities: the production of space in Final Fantasy games." Book chapter in Wardrip-Fruin and Harrigan (eds.) Third Person: Authoring and Exploring Vast Narratives. MIT Press, 2009


Lev Manovich. "How to Follow Global Digital Cultures, or Cultural Analytics for Beginners," Deep Search, ed. Felix Stalder and Konrad Becker. Transaction Publishers ( English version) and Studienverlag (German version), in press.

Lev Manovich. White paper: Cultural Analytics: Analysis and Visualizations of Large Cultural Data Sets, May 2007. With contributions from Noah Wardrip-Fruin.

Lev Manovich and Jeremy Douglass. "Visualizing Temporal Patterns in Visual and Interactive Media." Forthcoming in Visualising the 21st Century, ed. Oliver Grau. MIT Press.

Lev Manovich, Jeremy Douglass, and William Huber(In preparation) "Cultural Analytics: Methods and Techniques." For submission to Digital Humanities Quarterly.

So Yamaoka, Lev Manovich, Jeremy Douglass, and Falko Kuester. "Cultural Analytics on Ultra High-Resolution Displays." Paper submitted to ACM Multimedia 2009 conference (pending.)

Conference Presentations

Lev Manovich
  • April 2008 | London Schol of Economics, | London
  • April 2008 | Royal College of Art (RCA) | London | Lecture
  • April 2008 | Goldsmiths College | London
  • April 2008 | London Schol of Economics, Social Study of ICT Workshop (SSIT8) | London | lecture
  • May 2008 | Department of Design / Media Arts, UCLA | Los Angeles | lecture
  • May 2008 | “Software Studies” international workshop | University of California – San Diego, | San Diego, California | presentation
  • May 2008 | “Software Studies” panel, HASTAC II conference, University of California – Irvine | Irvine, California | lecture with Jeremy Douglass using HIperWall
  • May 2008 | “Software Studies” panel, HASTAC II conference, University of California – Irvine | Irvine, California | Panel organizer, moderator, and speaker
  • June 2008 | Software Cultures” lecture seriies, University of California – Irvine, June 4 | Irvine, California | lecture
  • August 2008 | ISEA 2008 (the International Symposium on Electronic Arts | Singapore | Lecture (satellite event at LASALLE College of the Arts)
  • August 2008 | FILE (Electronic Language International Festival) 2008 | www.file.org.br | Sao Paolo, Brazil via Skype from San Diego | lecture (video)
  • September 2008 | Ciantec (Annual conference on art and technology, Mackenzie University) | Sao Paolo, Brazil via Skype from San Diego | lecture (video)
Jeremy Douglass
  • May 2008 | Jeremy Douglass, “Programming Literary Flow.” Panel on Mapping Process in New Media Landscapes. ELO 2008: Visionary Landscapes. | Vancouver Washington | presentation
  • May 2008 | “Implied Code as Mental Geography.” Panel on ReVisioning Electronic Literature - Origins and Influences. ELO 2008: Visionary Landscapes. | Vancouver, Washington | panelist
  • May 2008 | Jeremy Douglass, “The LA Flood Project” presented with lead Mark Marino, collaborators Dena, Gutierrez, Hight, and Tao. HASTAC II: Techno-Travels. UC Irvine | poster
  • May 2008 | Jeremy Douglass, with Lev Manovich. “HIPerWall Demo: Cultural Analytics” HASTAC II. UC Irvine | presentation
  • May 2008 | Jeremy Douglass, “Visual Rhetoric for Large Displays.” Transcriptions Research Slam. UCSB | presentation
  • May 2008 | HASTAC II: Techno-Travels | “What is Software Studies?” Panel at HASTAC II: Techno-Travels, UCLA | Los Angeles | panelist
  • May 2008 | SoftWhere 2008| Software Studies Initiative at Calit2, UC San Diego | San Diego | co-chaired with Lev Manovich and Noah Wardrip-Fruin
  • May 2008 | “#Include Genre.” | SoftWhere: Software Studies 2008 | Calit2, UC San Diego | presentation
  • September 2008 | “Topics in Software Studies.” | CIANTEC 2008. Mackenzie University, Brazil | presentation with Cícero Silva (video)
  • March 2009 | Jeremy Douglass, with Derek Lomas and Daniel Rehn. “Playpower: Designing 8-bit Learning Games for Radically Affordable Computers.” O’Reilly ETech Emerging Technology Conference 2009: Living, Reinvented | San Jose, California | presentation
  • December 2009 | Digital Arts and Culture 2009 | "Software/Platform Studies" | Irvine, California | track chair
William Huber
  • May 2008 | Softwhere: Software Studies 2008 | "Soft authorship" | Calit2, UC San Diego | presentation
  • April 2009 | Thinking after Dark: Horror Videogames | "Catch and release: the ludological dynamics of horror videogames" | University of Montreal | presentation
  • September 2009 | DIGRA 2009 | Panel on game criticism | London | panelist
  • April 2010 | SCMS Conference 2010 | "Fatal Frames" | Tokyo/Los Angeles | panel chair

Visual Analysis Toolkit

[Tools and source code can be downloaded from the Software Studies Initiative Google Code site. We will document these tools more exhaustively in the near future.]

CAmacros: ImageJ macros for the analysis of images and video

CAjava: Java programs for image and video feature recognition

CAscript: Python script for a single project

CAbatch: workflow management tools for very large and mixed data sets

HIPerSpace visualizer: software for interactive exploration and analysis of collections of images and videos on HIPerSpace

ELogger: custom keylogger application for recording high-speed keyboard events running in emulation.

FSorter: sorting of very large image data using image statistics and metadata (UNIX script / OS X Finder Plugin).

Vilinx: automated image processing for video game play (UNIX script)

Toolkit para Análise Visual

[Ferramentas e código fonte podem ser baixados do Software Studies Initiative Google Code Site. Iniciaremos uma documentação mais exaustiva dessas ferramentas em breve.]


CAmacros: ImageJ macros para análise de imagens e vídeo

CAjava: programas em Java para reconhecimento de imagem e vídeo

CAscript: Script Python para desenvolvimento de projeto

CAbatch: ferramenta de administração do fluxo de trabalho para grandes conjuntos e dados mixados.

HIPerSpace visualizer: software para exploração interativa e análise de conjuntos de dados e vídeos em HIPerSpaces

ELogger: aplicação simples de keylogger para gravação de eventos de alta velocidade do teclado que rodem emulados.

FSorter: classificação de grandes quantidades de dados de imagens utilizando estatística de imagen e metadados (UNIX script / OS X Finder Plugin).

Vilinx: processamento de imagem automatizado para jogos de videogame (UNIX script).

Recently...