Visualising Cultural Patterns

Singapore.
The ISEA 2008 conference is pretty much over now - the last event broadly connected with it is a talk by new media theorist Lev Manovich in the beautiful Lasalle arts space. With a title of "Cultural Analytics", I wouldn't be so surprised if this was going to be pretty close to what my colleagues at QUT have in mind when they talk about cultural science...

His aim here is to extrapolate from current to future cultural trends, and he notes that such futurism is traditionally very difficult. Part of his approach, therefore, is to develop new projects with his students which may have the potential to set new trends themselves. Overall, he says, we'll see a very significant new cultural development that builds on data mining and data visualisation technologies.

The background for this development is set by a number of technological trends over the past 30-odd years. There was in the first place a data revolution: an exponential explosion in the amount of data generated. By 2011, the digital universe will be ten times the size of that of 2006; this adds up to a compound annual growth rate of 60%. This requires new tools for data processing, of course (something already realised by Vannevar Bush in his famous 1945 article "As We May Think" - and people like Bush and JC Licklider were scientists at the very forefront of such technology developments).

Computers were created to solve information growth, but ironically, at the same time, they have also contributed to a further increase of information growth. This, then, also gave rise to the development of the first systems for interactive data visualisation. The starting point for this in the modern context was a 1988 National Science Foundation report, Visualisation in Scientific Computing. Much of this required supercomputing resources at first, of course.

From this, Lev suggests, we've moved on more recently (roughly, from 2000 onwards) to the establishment of what he describes as a "data mining society". This was driven in part by financial institutions, anti-terrorism efforts, and more broadly business and government agencies in general. Data mining and analysis is now central to many industries, government activities, and NGO activities. In this, the trend in recent times has been more and more from data mining historical data to the analysis of real-time data (creating a kind of real-time society).

Large-scale informaton still remains relatively ununsual in business and other agencies, but there is a growing use of interactve visual tools such as geo-informaton systems (GIS), information dashboards, perceptual mapping (e.g. the mapping of consumer perceptions of certain brands on scales from conservative to modern, etc.), and so on.

This, then, also translates increasingly to the cultural field. There has been a massive digitisation of existing cultural assets in recent years (including projects such as Artstor, Google Books, or the BBC digital archive), and such material is therefore also available as data to be used in mining and visualisation. How may we take advantage of such data? There has been only very limited progress in the development of the access interfaces for such cultural materials (current user interfaces remain built on basic 19th-century metaphors - the film strip, the light table, etc.).

Further, then, Lev notes the rise of user-generated content (what I would call produsage). Social media sites combine user-generated content with conversations around and through these objects, and build on the widespread availability of the consumer electronics used to capture or create such content. (Lev points now to the massive growth of many produsage sites, and of the amount of content they contain.) We have moved from 'new media' to more media.

(As an aside, Lev asks, what does it still mean to be an artist in this age of such media overload? There are likely to be more images uploaded to Flickr every week than all the art objects in all the art museums of the world...)

User-generated content is one of the fastest growing parts of the expanding information universe. Humans are outperforming computers in generating more data here, and some 70% of the digital universe is today created by individuals, according to a 2008 IDC study. There is also a parallel expansion of the professional cultural universe, though: of agencies (educational institutions, companies, museums), of actors (professional cultural producers, students), of publishing (books, catalogues, Websites, and blogs), and of cultural objects - this growth of the creative and cultural industries is perhaps driven especially also by trends in the developing world.

Mixed with the globalisation of media connection and access, this leads also to an increasing diversification and fragmentation of global culture. It is now no longer possible to speak about centres and provinces - cultural practitioners in newly globalised cultuires are often more ready to embrace the latest ideas than their equivalents in the 'old centres' of world culture. This is visible, Lev says, especially on portals such as Archinect, Coroflot, and Xplsv.tv. What we see is the end of cultural theories based on small datasets ('classical Hollywood', 'Italian renaissance').

How, then, is it possible to develop a theory for specific aspects of this global digital culture with its billions of cultural objects, and hundreds of millions of contributors? Our normal, manual methods of talking about different areas of culture are no longer adequate for this task.

Enter data visualisation. What was once confined to the financial pages of newspapers and scientific applications is now entering the realm of popular culture and everyday digital tools; visualisation is becoming increasingly common across a widening range of environments, including the digital arts. Public visualisations of their history and performance at their headquarters have become as prestigious as their logos for some companies, in fact. Lev points to the exponential growth of a site like Visual Complexity to document these developments, in fact.

Overall, then, there is the potential for the rise of 'culture visualisation' to graph cultural patterns. Lev is in the process of launching CultureVis to help promote this work. As one existing leading-edge example of such culture-visualisation work he notes History Flow by Fernanda B. Viegas and Martin Wattenberg, which maps the multi-authoring history of pages in Wikipedia.

From this, can we develop a new discipline of cultural analytics - providing quantitative measures of cultural innovation? Can we develop real-time interactive maps of global cultural production, consumption, remixing,and collaboration? Can we visualise flows of cultural ideas, images, and trends? Can we visually represent how cultural and lifestyle preferences gradually change over time? Can we jump-start theoretical discussion of all the new cultural areas which have emerged recently (and for which we don't have an analytical language anyway) through the computer-based analysis of large numbers of objects in these areas? Can we track our movement from the 'old world' paradigm through the 'flat world' of 2000 to an 'inverted world' where newly developed countries are more active cultural innovators than the developing world? What is the shape of creativity today - are we inventing new cultural patterns and forms?

It's time to start thinking of culture as data, Lev suggests - providing a 'situational awareness' for 'cultural analysts'. This is also a kind of anti-structuralism: showing the diversity of human cultural expression. He and his colleagues are now building an open (source?) cultural analysis research environment for the data mining of professional and user-generated cultural content. His team's applied for and/or been able to (?) attract funding from the U.S. National Endowment for the Humanities Office of Digital Humanities, in the Humanities High Performance Computing initiative.

Yep, as I thought at the start - this is very compatible with what John Hartley has described as cultural science. It's also very much based on the same logic as our recent work in blog mapping, of course. CultureVis will be interesting to watch as it develops...

And that's the end of ISEA 2008!

Technorati : ISEA 2008, Lev Manovich, cultural analytics, cultural science, data mining, new media, produsage, visualisation
Del.icio.us : ISEA 2008, Lev Manovich, cultural analytics, cultural science, data mining, new media, produsage, visualisation