The Exotic/The Endotic

Humans are always looking for signals standing out against noise. But as modern society developed techniques to generate progressively more data, this became particularly important. Modern “news” pick out what is important for us to know. Claude Shannon defined information as the amount of unpredictability in a message. Flickr pioneered the use “interestingness” to filter the photos (Flickr, Interestingness Ranking of Media Objects, U.S. Patent #US008732175, Filed February 8, 2006). Google Analytics “monitors your website's traffic to detect significant statistical variations, and then automatically generates alerts, or Intelligence Events, when those variations occur” (, accessed 11/8/ 2014). These are only two examples of the key technology of “data society” - data mining – that use automatic computational techniques “with the intention of uncovering hidden patterns in large data sets” (

The metaphor of mining suggests that you recover what is valuable and discard the rest. But what if we reverse the procedure?

In his 1973 text Species of Spaces and Other Pieces, French writer Georges Perec noted how the news only talk about the exceptional, but never the everyday. He juxtaposes the exceptional and the exotic (the usual subjects of news according to Perec) to the infra-ordinary and endotic:

“What speaks to us, seemingly, is always the big event, the untoward, the extra-ordinary: the front-page splash, the banner headlines. Railway trains only begin to exist when they are derailed, and the more passengers that are killed, the more the trains exist… How should we take account of, question, describe what happens every day and recurs everyday: the banal, the quotidian, the obvious, the common, the ordinary, the infra-ordinary, the background noise, the habitual?” (G. Perec, Species of Spaces and Other Pieces, transl. J. Sturrock. London; New York: Penguin Books, 1997.)

Gerhard Richter. Tourist Office. Oil on Canvas. 150 x 130 cm. 1966.

Gerhard Richter. Toilet Paper. Oil on Canvas. 70 cm x 65 cm. 1965.

During a weekend in October 1974, Perec set out to realize his idea of systematically capturing infra-ordinary. Over three days, he described what he saw from a window of a café in Place Saint-Sulpice: buses, cars, people passing by, ordering coffee, the pigeons, and so on. The results were published as a short book An Attempt to Exhaust a Place in Paris.

Introducing the book, Perec writes:

"Many, if not most, of these things have been described, inventoried, photographed, written about or itemized. My intention in the following pages was to describe what remains; that which we generally don’t notice, which doesn’t call attention to itself, which is of no importance: what happens when nothing happens, what passes when nothing passes, except time, people, cars and clouds." (Species of Spaces and Other Pieces, translated by John Sturrock, Penguin, 1997.)

In our project we analyze and visualize every available public Instagram image shared over a number of days in central part of Kyiv. Similar to Perec, we use a rectangle as a frame, and only take into account what this frame captures. For Perec, it was a window in a café; for us, it’s the rectangle on the map defined by longitude and latitude coordinates passed to Instagram API. This means that as Perec, we don’t see the full trajectories of people before and after they happen to pass through / share photos inside the frame.

Constructing the Everyday

But this is there the similarities end. And this is not only because our frame is intentionally centered on the exceptional as opposed to an ordinary point: Independence Square where police and special government forces confronted tens of thousands of protesters, including people who were already camping at the square for a long time. However, the immediate area of the square is quite small in comparison to the whole area covered by our frame. As we know from the analysis of images and their tags, the larger proportion of shared images is not related to the political events. So in principle, our frame can capture the infra-ordinary along with the exceptional. So does it?

The key difference between Perec’s method and ours is this: “media” people share on social networks is nothing like Perec’s infra-ordinary.

Because hundreds of millions of people share media on hundreds of networks around the world, any generalization about visual social media will only catch part of the pictures. But at least, based on our own research into large samples of Instagram images from different locations (13 global cities in Phototrails, 6 cities in Selfiecity, ongoing work with 8 million images from NYC, and 400,000 images from Kyiv in the present project), we can say that a significant percentage of what people share on Instagram in the areas we looked at is highly curated and designed. People often stage their lives for social media, and they use Instagram’s own and third party image editing tools to refine the aesthetics of their images in ways that in the 20th century were only available to professional photographers.

Another important difference from infra-ordinary is the tiny number of images people intentionally share per day. For example, a part of our dataset for Feb 19-23 has 8,752 images shared by 4,262 Instagram users. How many images an average user shared? One user has 81 images; 79 users have 10 or more; and it quickly goes down after that. 2,669 users (%63) have only one image. The average number of images per user is 2.05 - hardly a comprehensive record of a person’s complete day. (If we were to capture from the complete city area, the number would have been higher).

Thus, even if the Instagram images often show the un-retouched everyday of people’s lives, there are so few of them that even the most mundane image can become exceptional. For instance, while having a coffee every morning is not unique, a single photo of this moment in a person’s Instagram stream becomes a unique event, if it is never repeated again.

Capturing a few days with a stationary video camera fixed at some location, and then cataloging the appearance and trajectories of every object and person would get us closer to the infra-ordinary. Alternatively, if we are to analyze access to video captured by small always on wearable cameras (such as Narrative Clip) from many people, we may get to Perec’s idea of a systematic science of infra-ordinary. Or video from a personal drone following you, or a future full implementation of “Internet of Things” where every object and surface will be surveying us.

(What would Perec say if he was alive today to see our data-centered world? Numerous companies and agencies – NSA, telecoms, advertizers and marketers, stores, and so on – are capturing detailed data about behaviors of people: phone numbers dialed; ads clicked on the web; Twitter posts; GPS tracks. But their goal is not the science of un-expectional. On the contrary, the goal is to find the unique: a sudden rize in the volume of people visiting a store web site; a network of people who happen to call each too often; a trend nobody saw before. In short, the goal is “data mining,” defined as the search for unexpected patterns – as opposed to the analysis of infra-ordinary. Even the makers of Narrative Clip, the device for infra-ordinary capture par excellence, don’t see their product in this way. The Sample Photos on company web site look like well composed photos from a hand held camera, as opposed to typical “infra-random” images captured by a device mounted on your moving body., accessed 08/02/2014)


In view of these characteristics of social media, what does our frame captures? Some of it is the exceptional as news define it – but that’s only a one part. The rest is something else, but it is not Perec’s infra-ordinary. People often carefully construct their “ordinary” for Instagram. And they also do share what is exceptional to them: birthday parties, a child’s first day in school, going to a big concert.

So overall, our window captures these categories of images (and of course there are other images, so this list is not meant to be exclusive):

Globally exceptional (e.g, what global news are also likely to report, such as Maidan events).
Locally exceptional (e,g. what local news are likely to cover).
Personally exceptional (e.g., a birthday party).
Personal everyday, staged/edited for Instagram (e.g., many, but not all, selfies).
Personal everyday, un-edited.

Like many sets of categories, these also overlap. A revolution leading to a new government happening in your city can be “globally exceptional,” “locally exceptional”, and personally exceptional. An everyday event captured on video can become viral on YouTube. And where do we put a non-staged image with a Instagram filter – “personal edited” or “personal un-edited”? Rather than discrete either/or boxes, we can think of these categories as the centers of fuzzy clusters – which, with enough work, we can extract using algorithms and visualize.

We can easily introduce new categories and subcategories – but we can’t describe what social media gives us with a single term. Often, it is more than unmediated everyday, but less than highly mediated news. Neither this nor that, but everything side-by-side. And it is this side-by-side which gives social media its uniqueness as genre of representation and communication.

The Tyranny of Geometry

In contrast to ornament and decoration, classical art has always been about the summarization, abstraction, and representation of what is important in human life. Portraits, historical scenes, chronicles of wars. Never infra-ordinary. Finally, in the 19th century literary and visual realism – which, not accidently, emerged at exactly the same time as social statistics – focused on the actual as opposed to idealized life. But it was still processing the raw data into “typical” and “representative” (e.g., “the image of a worker,” as opposed to a single concrete worker.) The most prominent realist super-novel – Balzac’s Divine Comedy – is populated with types such as the noble soldier, the proud workman, the alluring mistress, and others (W. H. Helm, Aspects of Balzac, London: Eveleigh Nash, 1095.) This probably led Friedrich Engels (the collaborator of Karl Marx) to say: "I have learned more [from Balzac] than from all the professional historians, economists and statisticians put together." (Karl Marx and Frederick Engels, Literature and Art: Selections from Their Writings, New York, 1947).

Aleksander Deineka. Left March. 1941. Oil on canvas. Russia.

Andrei Mylnikov. On Peaceful Fields.1950, oil on canvas. Russia.

New medium of photography could in principle capture the “real” real, as opposed to types. But in its first decades first it was too slow and cumbersome for that. Eventually, it was artists rather than photographers or filmmakers who finally focused on making visible the everyday without generalizing, aestheticizing, or editing it: Andy Warhol’s Screen Tests (1964-1966) and Empire (1964), Every Building on the Sunset Strip by Edward Ruscha (1966), Yoko Ono’s Instruction Paintings from late 1950s and 1960s, or John Cage’s 4’33’ (1952).

Situationists created the algorithms for travelling through the everyday city (a derive). In the issue of The Situationist International, Raoul Vaneigem wrote, "All space is occupied by the enemy. We are living under a permanent curfew. Not just the cops—the geometry." (Gray, Christopher, editor, Leaving the 20th Century: the Incomplete Work of the Situationist International, London: Rebel P, 1998. p. 26.)

“I wanted to include everything: sports, politics, even groceries. Everything should be put in a film” - Godard, quoted by Amy Taubin in “2 or 3 Things I Know About Her: The Whole and Its Parts,” The Criterion Collection web site.

Over last ten years, the geometry of the cities did not change much, so it determines our movements as much as it did in the 1960s. But what did change is the presence and role of maps. Digital interactive maps are now built into tens of thousands of apps for mobile devices, and we use them daily. Looking and working with the geometry of the cities and grids of coordinates is now part of our everyday.

If we plot hundreds of thousands of locations of social media contributions (tweets, Instagram photos, etc.) by lots of people over an empty background, the geometry of a city becomes visible. And the more we plot, the more pronounced it becomes. The social – the regular phenomena made up from small behaviors of participants, each seemingly following their own free will – comes into focus. The daily and weekly routines of many individuals become visible as the structure – which situationists’s algorithms were trying to escape.

But as we remove points, this structure starts dissolving. Finally, when we get to a tiny percentage (the number depends on the city size), the locations of points now look random. As we take more points away, the randomness increases. At the end, we are looking at the original empty background. The city has disappeared.