 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
 |
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
|
|
 |
 |
|
| Big Data Lisa Jevbratt, Christina McPhee and Andrea Polli Brett Stalbaum, editor Abstract Artists confront the problems of data density and range in the aesthetic of the sublime. Together with an introduction by Brett Stalbaum, these essays by Lisa Jevbratt, Andrea Polli and Christina McPhee were first published in print for YLEM Journal, Volume 24 Number 6, May-June 2004 (McPhee) and Volume 24 Number 8, July-August 2004 (Jevbratt & Polli), at the suggestion of Loren Means. The YLEM Journal is the bimonthly publication of YLEM, a twenty-three-year-old organization dedicated to the nexus of art, science, & technology. For more information on joining YLEM and to view the YLEM Journal online, visit www.ylem.org. Introduction Brett Stalbaum Moore's Law, Gordon Moore's famous prediction that processing speeds double approximately every 18 months, has proven to be so prescient that it long ago rose past the status of provocative futurist claim to the level of pedestrian cultural assumption. But what has not yet become an accepted cultural assumption is that Moore's law is at least matched, and possibly exceeded by the exponential growth of data to be processed. The relationship between humankind's ability to collect data and to process and understand data is co-exponential: both are exploding. Data sets from genomics, astrophysics, geography, geology, particle physics, climatology, meteorology, nanotechnology, materials science and even the search for ET are producing quantities of data that challenge the technical limits of super computers, distributed computing, grid computing, and superscalar simulation techniques. Even given Moore's law, optical networks, and cheap mass storage, the problem of big data is nevertheless looming larger as our ability to collect data actively competes with our ability to process and digest it. Computation has already become a nominal, if not tacit assumption in contemporary art practice due to the ubiquitous implementation of computer and communications technologies in all aspects of our emerging global culture. How does big data impinge on the present generation of representational artists who operate under the assumption of a rich computational environment? And what are the emerging aesthetic and conceptual parameters that impinge on the practice of artists who consciously recognize data and coding as the primary expressions of an art practice wherein the notions of "representation" are not limited to narrowly prescribed assumptions regarding a specifically graphical or interactive interface and networked distribution as the primary cultural operatives between artist and audience? What other questions arise in an environment where we live in a constant streaming wash of data, and what are the issues surrounding how artists might help interpret both cultural and scientific phenomena? Lev Manovich raises a particularly interesting issue in his 2002 essay titled "The Anti-Sublime Ideal in Data Art". In it, Manovich identified an aesthetic approach to big data that seeks to interpret large data sets on much the same terms as designers and scientists seek to analyze data; a pursuit which he describes as the exact opposite goal of romantic art. "If Romantic artists thought of certain phenomena and effects as un-representable, as something which goes beyond the limits of human senses and reason, data visualization artists aim at precisely the opposite: to map such phenomena into a representation whose scale is comparable to the scales of human perception and cognition." He goes on to form a critique of such practice, and raises the question of "How new media can represent the ambiguity, the otherness, the multi-dimensionality of our experience... In short, rather than trying hard to pursue the anti-sublime ideal, data visualization artists should also not forget that art has the unique license to portray human subjectivity – including its fundamental new dimension of being 'immersed in data.'" Look to the writings of three artists whose practice conspicuously intersects with questions relating to the romantic and the sublime. Their writings, each in a different manner, suggest possible paths toward answering the many issues that have been raised by the explosion of, and our immersion in, big data. Interestingly, Andrea Polli's "Atmospherics/Weather Works: Artistic Sonification of Meteorological Data" begins with a quotation from the romantic American poet Walt Whitman's "Proud Music of the Storm". Polli is interested in how sonification of large data sets differs aesthetically from visualization, and in helping a sonic "language or series of languages for communicating this mass of data needs to evolve." Not only does Polli's text clearly describe the types of aesthetic choices that were necessary in the sonification of the President's Day Snowstorm and Hurricane Bob data, but also reveals a successful example of interaction between and artist and scientist(s) to reinforce and potentially uncover new knowledge through what she claims is a potentially more visceral sonic experience of data. Christina McPhee is also interested in the sonification of environmental data in "Sense of Place and Sonic Topologies: Towards a Telemimetic Sublime in the Data Landscape". Her text is a theoretical riff based in part on her experience to date with her nascent++ Slipstreamkonza project, which processes data representing the carbon fluctuation of tall prairie grasses. McPhee's text is connected to Polli's not only in terms of their shared interest in data sonification and collaboration with science, but they also meet up in something of a rapprochement with the romantic tradition that Manovich discusses. Polli's notion of how data sonification might lend to a "physical and emotional exhilaration [that] enhances the scientist's understanding" is obviously congruent with McPhee's notion that "...one may turn a gaze to what cannot be 'seen'. Here we move into a zone of the sublime. Sublimity refers to that which is below, beyond or immanent relative to an ontological or cognitive threshold. I assume that there is a way of expressing this indeterminate zone, or invisible condition, in both the realms of the physical, cultural landscape and in the interior, 'behind the screen' landscape of the net." It should be noted that this rapprochement with the romantic and the sublime is in no way a conservative one. The sublime, which can also be described as a particularly human cognitive response to decision-making circumstances wherein the amount of data overwhelms one's deductive reasoning capabilities, yet under which humans are more often than not able to think and act to yield successful outcomes, is one of the general capabilities to date that has evaded machine intelligence. It seems that the prodigious deductive abilities of computational systems cannot yet simulate the prodigious inferential capabilities of the human mind. We have not yet entered the period of strong AI predicted in JCR Licklider's 1960 essay "Human-Computer Symbiosis", but rather we continue to exist in the symbiotic phase where "computing machines can do readily, well, and rapidly many things that are difficult or impossible for man, and men can do readily and well, though not very rapidly, many things that are difficult or impossible for computers." Big data, as it turns out, is a challenge even to this successful symbiosis. The work of Polli and Jevbratt, particularly through their engagement with issues relating to the sublime, express congruence with the problem spaces of both data mining, artificial intelligence and many other disciplines faced with big data problems. Jevbratt's "A Prospect of the Sublime in Data Visualizations", builds on McPhee's assumption regarding expression of the "indeterminate zone" of the sublime, and is also an answer to Manovich's use of her work as an example of the anti-sublime ideal. In her essay, she explores the potential for a symbiotic human-machine space to be understood via the sublime in terms of a "methodological distancing" including the concept of "Via Negativa" and a proper appreciation of the opportunistic nature of meaning that would allow us to take into account (romantic) philosopher Emmanuel Kant’s notion regarding the "mobilizing effect the sublime has on our organizing abilities." Jevbratt thinks this would help us avoid "the most common mistake in data visualizations...", that being "not too much information but too little, their 'images' of the data landscape are not high resolution enough for an esthetic decision to be made." With special thanks to Loren Means and the board of YLEM, I hope that the exploration of data and the sublime in the essays to follow will stimulate further discussion of the issues and opportunities presented to artists by the problem of big data. Brett Stalbaum La Jolla, California May 2004  |
|
[1] Manovich, Lev The Anti-Sublime
Ideal in Data Art, (2002) http://www.manovich.net/DOCS/data_art.doc |
| The Prospect of the Sublime in Data Visualizations Lisa Jevbratt www.jevbratt.com Looking out and up  The Polar Sea, Caspar David Friedrich 1823-1824 In Caspar David Friedrich’s (German, 1774-1840) “The Polar Sea” (1823-24) we look out on an endless inhospitable ocean of ice, and a shipwreck - a trace of an attempt to do the impossible, to go “there”, to reach for and understand the unbearable void.  Starry Sky We look up at the starry sky and we sense a fear of not comprehending and being engulfed, a fear of the unknown, and simultaneously we experience a longing for the inaccessible, impenetrable darkness. These are the classical visuals of the sublime. Images of a sense of grandeur we can’t reach, which we can’t penetrate or grasp. It is in the very far distant, it is hidden in layers of mist, or made inaccessible by a climate not suited for us and it instills a sensation of deep fear. Yet we urge for it, we are fascinated and attracted by it. Looking down and in  The Rocky Mountains, Colorado. Picture sixty-three taken with my new camera. 2004. Jevbratt 2004. We look down. We consume satellite and aerial photography in all its forms; on the web we can access detailed satellite and aerial photographs looking down on our houses or whatever we want to surveil from above, we are capturing mountains far below with our first digital camera, we have the poster of “lights emitted from the earth” on our walls (maybe pondering what it says to bypassing intelligences – gods, aliens and others), and we rely on satellite imagery to predict weather and track fires.  Peripheral evidence: two dimensional polyacrylamide gelWe look in. The genome is mapped and we are trying to figure out how to look at it. New technologies for looking in towards and inside cells, RNA and nano structures are rapidly developing, and the methods of making peripheral evidence of them and their processes are constantly refined. We look at our networks that produce data about ourselves in sublime quantities. Anti-Sublime The datasets we are looking at now are of no less dimension, vastness and grandeur than the datasets that were the subject of the classical sublime; and the sensations of the sublime harvested by the romantic artist and others is of great interest to us when trying to make sense out of our datasets today. However, a quite logical argument against the possibility of the sublime acting within data visualization can be made. It has been well formulated by Lev Manovich in “The Anti-Sublime Ideal in Data Art”. “If Romantic artists thought of certain phenomena and effects as un-representable, as something which goes beyond the limits of human senses and reason, data visualization artists target the exact opposite: to map such phenomena into a representation whose scale is comparable to the scales of human perception and cognition. For instance, Jevbratt’s 1:1 reduces the cyberspace – usually imagined as vast and maybe even infinite – to a single image that fits within the browser frame.”  The whole Internet with The Web as it existed in 1999 (red) and 2002 (green).
1:1(2), Interface: Migration, Lisa Jevbratt 2002The reasoning is very clear and it troubled me because I instinctively knew that it was wrong – both in making the case that data visualization by definition is anti-sublime and that my project 1:1 would be a good example of this case. How then can data visualizations utilize the (or be) sublime? Why should they aim to? Push, pull While the datasets of today are as substantial as the ones dealt with in the classical romantic sublime, there is a difference in direction and force. In the original sublime the force is attraction. The object of desire is over there, far away and we want to reach it. We want to go there, we are scared and intimidated but our longing and effort is ‘towards’. When our force (engine, energy, luck) fails the ship stops, it does not get closer. The forces of nature push us away - we urge to approach. The classical sublime was the extreme tension of not knowing and wanting to know; we were attracted by the fact that we didn’t know. Now, looking in and down the force is reversed. If the engine in a plane stops, it approaches the ground; the natural force is gravity and we want to stay up and away. We are pulled down and respond by retracting. The forces of nature pull us down, in - we urge to repel. The sublime now is the extreme tension between (hypothetical) familiarity - the earth is our home, the cells and DNA are in our bodies, the networks are our creation - and a methodological distancing. Esthetic decision-making In the article “Systems Esthetics” Jack Burnham wrote about the new complex process or systems oriented society, culture and economics he saw emerging: a new era in which systems analysis would be the most relevant method for making understandings in any discourse. Burnham argues that because we can’t grasp all the details of our highly complex systems (economic, cultural, technical, etc), we cannot make “rational” decisions within them or understand them by analyzing the systems or their parts. The way to make decisions within them and to understand them is by making more intuitive, “esthetic decisions”, a concept he borrows from the economist J. K. Galbraith. This idea has an intriguing parallel in the philosopher Emmanuel Kant’s reasoning about the mobilizing effect the sublime has on our organizing abilities. He claims that in experiencing the sublime, by facing large amounts of information, huge distances and ungraspable quantities, our senses and our organizing abilities are mobilized. Contrary to what might be believed, we feel empowered, able to make decisions, and capable to act. Many strategies for aiding people in the task of turning any large set of data into knowledge assumes that they should be presented less information and fewer options in order to be able to make sense out of the data. However, humans are capable of sorting through enormous amounts of visual information and make sensible and complex decisions in a split second, (the ability of driving a car is one example). Supported by Kant’s idea I propose that under the right circumstances, drawing on sensations of the sublime, people can, when faced with huge quantities of data, be mobilized to make intuitive understandings of the data. Many information visualizations and displays are a result of the mistake of compressing the information too much and decreasing the amount of information through calculations that embody assumptions that are never explained. The most common mistake in data visualizations, artistic or scientific, is not too much information but too little, their “images” of the data landscape are not high resolution enough for an esthetic decision to be made. Meaning is opportunistic Why is low-resolution highly compressed data representation less meaningful? If it is counteractive to a sublime, why is that? How does that sense of awe and 'aha' that the fear and force of the sublime helps us experience transpire? Meaning behaves like a parasite. It is opportunistic, taking “immediate advantage, often unethically, of any circumstance of possible benefit” (the definition of opportunistic at http://dictionary.com). If meaning in fact is opportunistic, and opportunism implies an unethical stance then it could follow that meaning does not thrive in an ethical environment. This reasoning is more interesting if one understands the term ethic as an opposition to faith. Ethic is a stance in which one in any moment is aware of ones goals and choices. One has a plan and a way in which to carry it out. Faith is a stance in which we let go, were we are submerged and surrendered, when we are trustingly accepting a “truth”, an emotion or a calling. (At the conference “Derrida and The Question of Religion” at UCSB in fall of 2003, Derrida mentioned during a discussion between him and presenter the concept of the calling and reflected on how that concept is not that different from how animals follow traces. This constitutes an interesting point for the thoughts in this paper.) Culture then is extremely meaningless because so many choices have been made, and nature is extremely meaningful since no choices have been made. It seems like we strive to cut the extremes, the very meaningful and the extremely meaningless. To make culture more meaningful we create unstable conditions for decision making: i.e. to reduce the number of ready-made choices we create unpredictable and arbitrary events and expressions within it. It is interesting to see that younger people are more prone to produce these. Quite likely a young mind has more difficulty dealing with the burden of meaninglessness, and thus tries to minimize it by generating arbitrary signs (such as the expressions, fashion and sounds of various subcultures). To make nature less meaningful we organize and categorize it and our experiences of it. (Of course nature is only void of choices if one does not believe in a creating god. In fact the very idea of a creationist god could be seen as another attempt to decrease the meaningfulness of nature). However, another, contradictory reaction to nature might be that our ability to perceive meaning is numbed by the loudness of it. Just as our retina gets saturated after looking at one color and creates a ghost image of the opposite color when we look away briefly, nature can (falsely) appear as if completely void of meaning. The result of this reasoning is that as soon as we are trying to make what we experience ethical, i.e. succumb to a plan and direction by making deliberate choices, the experience and its data decreases in meaning. If we semantically categorize and search for meaning it is as if we try to look at the dust on our corneas, we can’t see it unless we stop looking at it. Everything becomes meaningless when we attempt to “capture” the meaning. In the task of visualizing huge datasets this means that we need to avoid making assumptions about the meaning of the data in order to allow meaning to find an opportunity to occur. Perhaps the answer to the question in the beginning of this paragraph is that we need to allow the interplay between the extremes, allowing the meaningfulness and the meaningless to happen by not attempting to reduce either. Identity in the non-intended Some years ago a student of mine made an interesting discovery in a project he made. It was Web software that returned the result of a search for something on a selection of search-engines in the reversed order. I.e. the most relevant, however the search-engines define that, was last on the list and the least relevant of the relevant sites was shown first on the list. The result was striking. The least relevant sites, the ones usually so many clicks away we don’t bother to look at them, varied greatly between the different search engines. The most relevant results, the ones usually displayed on top, were all the same. A similar finding was made some centuries earlier by Giovanni Morelli (1874-1876). He sought to find a method of determining authorship of paintings and came upon the fact that authorship is more detectable in the parts of a painting done with less intention; the parts which are not significant for the author or the genre in which the painting is made, such as earlobes and fingernails. His method is now called “The Morelli Method”. In art historian Edgar Wind’s words it is interesting that “Personality is found where personal effort is the weakest”. Even more strikingly, what seemed to be true on the Web is also true in biology, according to Albert-Laszlo Barabasi in his booked Linked: The New Science of Networks’. Barabasi is doing research on the network structures and linkage systems of various fields from computer networks to biology. He finds that “For the vast majority of organisms the ten most-connected molecules are he same.” (p. 186) These highly connected molecules, hubs in Barbasi’s terminology, are equivalent to the most relevant pages in a web search or the traditionally most “important” features in a painting. These are the items, nodes, with the most intent. Just as the least relevant web pages are the most dissimilar, and the least important features such as earlobes say more about the painter, the difference between different organisms and the production of their identity lies in the least connected, least used or significant molecules. "[O]nly four percent of the molecules appear in all of them. Though the hubs are identical, when it comes to the less connected molecules, all organisms have their own distinct varieties." (p. 187) Via Negativa These are all evidences that reality does not show itself to us in an expected manner, through intention and expression, but it reveals itself to us indirectly in small fragmentary pieces. The method of searching out those bits and pieces without preconceived notions on what to find has been an important method in various mystic traditions, and the term Via Negativa, possibly coined by Dionysius the Areopagite, a late 5th century mystic, is used to describe it. Via Negativa is a method of distancing, of negation, in which we claim or pretend to not have any preconceived notions of the systems that we are looking at. The method has a lot of similarity with artist methodologies (such as Joseph Beuys) and now also with some contemporary scientific methods. For example, the process of harvesting, sequencing and mapping the human genome has been described as that of a group of people in a dark room fumbling around not knowing what is in the room, how the room looks or what they are looking for. Someone bumps into a thing with four sharp corners and starts to look for other things with four sharp corners. Someone else decides to move along what seem to be walls and feel their texture, yet another sits still and waits for the others in the room to pass by, taking notes on their activities or maybe on their scents. The value in Via Negativa for data visualization is that it creates that opposing force of not falling into, a repelling force counteracting the gravity pulling us down. The Via Negativa enables the sublime to operate. Epilogue 
Juno falling and snacking. Still from Spy Kids 2: The Island of Lost DreamsIf staying up is our (or others, things or beings) effort, then the fall, the ultimate inability to do so, is a trope of interest. There are significant falls ranging from literary, such as the fall of Alice, the girl in Wonderland, to political, such as the fall of the Twin Towers. In "Spy Kids 2: The Island of Lost Dreams", Carmen and Juni, the spy kids, fall into a model of the landscape, through the mouth of a volcano. Their fall lasts for an extended time, so long that they take comfortable positions, eat a snack and discus the possible outcomes of their fall. In the end they might not have been falling for a very long distance. The model that they are falling into has an air vent blowing air sufficiently strong for them to be lifted and they might have spent most of the time in the illusion of falling. In a time period of eight weeks I experienced three events of falling substance that for one reason or another seemed to have significance. On a dreamlike evening just after sunset up on a mountaintop overlooking the beautiful cloud covered southern Californian coastline far from above, a shooting star released itself from its usual celestial path, where we are accustomed to see it disappear far in the distance, and fell towards the earth and us as a real physical object on fire landing not too far away from us. A few weeks later I spent an hour in my closet attempting to clean up after a mold infestation that happened earlier in the year when I, more or less simultaneously, hear my neighbor flush her toilet and felt a liquid substance on my head. Three days into my artist residency in Huddersfield, Yorkshire, England, where I have taken refuge from the daily duties of teaching and meetings, an American bomb fell over Yorkshire; whatever the target was, it missed. Lisa Jevbratt |
|
 |
 |
 |