The Big Bang Data Theory

Reflections on the recent Big Bang Data exhibition held at Somerset House, London.

 

They say that you should never meet your heroes, for they will always disappoint you.

Writers more gifted than I have told me to always avoid using a cliché to begin a piece of writing; and indeed to avoid cliches throughout a piece of writing. It is good advice, and advice I try to follow, but in this instance I felt the above words were the most concise to describe the feelings I wish to express. While my encounter with the Big Bang Data exhibition at Somerset House was not a case of meeting my heroes per se, the sense of anticipation I had before going into the exhibition was not met by what I found when I got there.

As I sat in Caffé Nero on the corner of The Strand and Waterloo Bridge shortly before walking round to Somerset House, munching on an all day breakfast panini, I confessed in my notebook that I was “a little nervous that I’m not going to like it – or that some of the works I’ve found really inspirational in books will not live up to my expectations.” I know better than to judge an exhibition by its venue, but I had a sense Somerset House was a strange venue for such an exhibition, which I would normally, rightly or wrongly, associate more with a venue such as the Science Museum, or perhaps The Barbican. Though I confess, this might be coloured by my previous experiences of Somerset House as a venue for exhibitions of fashion and jewellery, and Paloma Faith concerts.

The exhibition, staged in the Embankment Galleries, opened with Timo Arnall’s 2014 piece Internet Machine [1], a three screen work shot inside the Telefónica data centre in Alcalá, Spain, to which Arnall was granted exclusive access. [2] The piece depicts the server rooms, power generators, and cabling of the data centre – just one small part of the physical structure of the Internet – though a series of moving stills. At first, the piece gives the impression of a slow-moving camera panning over the space, but after a few moments it becomes clear that it is not a video, but not exactly a photograph either. In a similar manner to Nancy Davenport’s Weekend Campus (2008), the piece combines elements of stillness and movement: the virtual camera pans up over the stitched photograph; the lights on the server cages appear to blink, but the fans are not spinning. [3]

The images themselves are clinical and likely familiar to anyone who has ever typed the phrase ‘data centre’ into Google Image Search; but their scale, ultra-high definition, and the slow, deliberate motion of the camera conspire to make them quite seductive. According to its wall-mounted interpretation, the piece encourages us to think about the Internet as having a physical presence, to counter metaphors like ‘The Cloud’, which suggests ethereality and weightlessness. This is does, but this statement also makes its seductiveness problematic, because it does little to address the environmental impact of such data centres; and risks contributing to the fallacy that digital storage is an unlimited, clean, resource. Data centres are estimated to consume around three per-cent of the world’s electricity, [4] much of which is supplied by high-pollution coal fire power stations. While some companies – Apple and Facebook, for instance – have been trailing renewable energy at some of their data centres, there is no escaping the fact that these data centres leave a deep carbon footprint. Electricity, of course, is not the only resource demanded by these data centres, and if we start to factor in the raw materials required simply to build their hardware, the picture appears even less clinical. [5]

Moving deeper into the exhibition, the first section, ‘The Weight of the Cloud’, looked at the physical supports required by our data: the media it is stored on and the cables laid under the ocean needed to send it around the world. In fact, it was here that I learned the first submarine transatlantic telephone cable was laid between Scotland and Canada in 1956; more precisely between Oban and Clarenville, Newfoundland. [6] Trivia aside, the physical structure of the Internet is both an interesting and pressing topic, but while this section of the exhibition was informative, it offered too little in the way of critical analysis, feeling more like an introductory summary than an in-depth investigation. It pointed to they way the Internet reshapes cities, as financial companies vie to get their servers closer and closer to the main exchange, but makes no mention of the re-shaping of the landscape required to lay a cable in a straight line between Chicago and New York to shave a few milliseconds off the time it takes data to travel between the two. It gives us a timeline of data storage devices from 1950s punch cards to a USB stick, but says nothing of the raw materials and the labour required to make these storage devices.

This is not to suggest that every exhibition or every work of art that approaches the subject of ‘The Internet’, ‘data’, or, even more generally, ‘computing’ must engage with these issues, but it struck me as odd ­– and a little disappointing – this show did not take the opportunity. In fairness, the show does include a number of works that attempt to illuminate what we might call the ‘dark side’ of ‘Big Data’ and the ever-increasing connectivity the Internet enables, particularly with regard to our personal privacy, [7] but on the whole I felt it to be informative, yet somewhat lacking in critical depth. Though I personally found this disappointing, I can accept that this particular exhibition was perhaps not the venue for an incisive analysis of all the issues connected to ‘data’, and may be better viewed as a conversation-starter.

Unfortunately, what I saw as critical shallowness was not the only reason I found I found the exhibition to be somewhat disappointing. The show covered a huge range of works – ranging from hand-drawn data visualisations on postcards, to technically-complex room-sized installations, and, unfortunately, some of these works were badly let down by their presentation; Thomson and Craighead’s Horizon [8] was a case in point. Originally commissioned for Dundee Contemporary Art’s Timecode exhibition in 2009, Horizon brings together live footage from public webcams located in every time zone around the world in what the artists describe as an “idiosyncratic global sundial”. Practically, this is a large, portrait-orientated grid of webcam images, each line annotated with the location of the webcam and its time zone. Reading the piece from left to right, gives the viewer an impression of what has been going on in that place over the course of the day, and, over the course of twenty-four hours, the dark band of images from the part of the world where it is currently night passes from the top to the bottom of the screen like a shadow. In Dundee, the work was presented as a large-scale projection, reaching from the floor to the top of a false wall installed in the gallery space, but here it was presented on a flat-panel display mounted vertically on the wall. Inevitably, this changes how the viewer looks at and interacts with the work, instead of standing back to survey the whole piece, one immediately moves in closer. With only a simple change of scale, the ‘feeling’ of the work changed considerably – it lost some of its presence, it sense of scale, which felt appropriate for a work that in effect spans the entire world. 

The reduced scale of the work was not, however, its only, nor its most pressing, problem. The images were still legible and one could still get the sense of narrative and of a sundial the artists describe. The most pressing problem was that the Mac desktop and dock were visible underneath the work. The works does not need to disguise the fact it is running on a computer, everyone visiting the show can probably tell us without seeing the desktop behind it. While some of the other works in the exhibition were originally intended to be viewed online, Horizon was not, meaning that the desktop and dock were never meant to be seen with the work. The effect of seeing the desktop is to foreground the technology; prompting us to think about what is going on behind the scenes rather than focus attention on the work itself. It feels like a mistake, and, worse still, a mistake that could have been – and indeed should have been – easily corrected.

Any exhibition involving as much technology as Big Bang Data is bound to experience some technical difficulties, but Horizon was not the only work to be let down by its presentation. Rather than expend more words on disappointing presentation, however, I want to touch on presentation more generally, as this was the area of the exhibition that perhaps has the most bearing on my own research.

I already noted the wall-mounted interpretation of Timo Arnall’s Internet Machine, but it was a notable feature of the exhibition that every work was accompanied by a fluorescent yellow interpretation. I use the word ‘interpretation’ quite deliberately for, while labels are an expected part of most exhibitions, it felt as if, without this interpretation many of the works would be completely illegible. ‘Illegible’ here should not necessarily be conflated with ‘uninteresting’, what I mean is that it is impossible to discern, without some kind of description or key, precisely what data is being visualised in, or is acting upon, a work.

Take, for example, Jaime Serra’s 2010 work Vida sexual d’una parella estable (Jaime i Francisca) [9] a portrait-orientated, printed data visualisation, which records the sexual encounters between the artist and his wife over the course of one year. Nights without sex, sex with and without penetration, with and without orgasm, are amongst the possible sexual acts recorded and presented as a series of coloured lines. The work’s visual appearance is entirely divorced from its subject matter, each coloured line is entirely symbolic and could just as easily be substituted for any other colour of line, weight of line, or a different shapes altogether. In other words, the same data could be presented in any number of different forms. It is not a stretch to suggest that, if a viewer were to encounter this work in isolation, without its accompanying title, key, and textual description, there is no possible way they could connect the work to a history of sexual encounters, or lack thereof. The work could just as easily be a visualisation of the different blends of tea the artist drank in the course of a year, or a random collection of coloured lines.

One might argue that clean lines and geometric shapes are, in a sense, the ‘safest’ means to visualise data, assuming the aim is to present data with the minimum of intervention from the artist; to present the data as a kind of ‘transparent interface to reality’. Presenting data as a series of geometric shapes or a progression of different colours reduces a diverse set of data points (whether it be a sex act or a cup of tea) to a single form, what Mitchell Whitelaw describes a s kind of “uniform diversity”, [10] which does not give added weight to any one particular data point. In this approach, the work’s key, or in some cases simply its title, becomes absolutely crucial to our ability to read and understand the work. This might be an incredibly obvious point to make, but it strikes me that, in the case of work like Serra’s, our way into the work, the hook if you will, is realising what the data represents. In other words, when it crosses the line between ‘data’ and ‘information’.

Even works like Julie Freeman’s We Need Us (2015), [11] which makes no attempt whatsoever to create ‘information’, seems to rely on some degree of interpretation, or perhaps more accurately framing, to really hook us in. According to its label, We Need Us explores the qualities of data itself without trying to make sense of it, specifically, the qualities of live data related to activity on the Zooniverse website. [12] Set to a soundtrack composed primarily of edited nature recordings, geometric shapes reminiscent of a Malevich painting, drift across the screen; their shape and transparency continually fluctuating. The work is, in itself, visually interesting but there is nothing in the animation to suggest a connection with anything outside of the work; and, for me at any rate, it is this connection that makes the work something ‘more’ than an animation. I do not, however, get this sense from the work alone: I need the label to be able to make this connection.

One of the observations from my own recent work was that, as I rather expected, the labels attached to the work played a huge role in how the viewer came to read the work. One of the most common responses from viewers was that they suspected the work to be responding to some kind of data input, but they were unable to discern precisely what data the work was responding to. This was particularly true of KODK, [13] which responded to changes in Kodak’s share price by changing colour, but no key was provided to relate those colour changes to a fluctuation in stock price, and only the merest hint that it was even being influenced by an external data source. It is true that I never wanted the work to become purely illustrative but, in retrospect and looking to my own experience of these unfamiliar data-based artworks, I wonder if I held too much back.

Though on the whole I came out of Big Bang Data feeling slightly underwhelmed, it did give me some food for thought; though admittedly not about the issues covered in the exhibition itself. One final note: for all that I found disappointing about Big Bang Data, my visit at least ended on a high note. At 17:48:54 on Tuesday 15th March 2016, The BBC News website published the headline, “SFO drops Forex market investigation”, which caused Ellie Harrison’s Vending Machine (2009) [14] to release a packet of prawn-cocktail flavoured Walkers Crisps. I happened to be standing in front of Vending Machine, which was positioned in the exhibition’s shop, at this time and, as such, am now the proud owner of said bag of crisps.

That said, I am a tiny bit disappointed about the flavour.

 

 

 

 

Notes

1. For a short clip from Internet Machine, see Tino Arnall, ‘Internet Machine (Trailer)’ , https://vimeo.com/95044197 (accessed 21 March 2016).

2. The exhibition was originally developed by Centre de Cultura Contemporània de Barcelona (CCCB) with support from Telefónica.

3. Some technical information about the making of Internet Machine can be found on Tino Arnall’s blog. See: Tino Arnall, ‘Internet Machine’ [online] 13 May 2014, http://www.elasticspace.com/2014/05/internet-machine (accessed 21 March 2016).

4. BBC News, ‘Microsoft tests underwater data centre’ [online], 2 February 2106, http://www.bbc.co.uk/news/technology-35472189 (accessed 21 March 2016).

5. Unknown Fields Division’s 2015 work Rare Earthenware is a work closely engaging with the environmental impact and human face of the electronics industry. See: Victoria and Albert Museum, ‘What is Luxury? – Object in Focus: Rare Earthenware by Unknown Fields Division’ [online], http://www.vam.ac.uk/content/exhibitions/what-is-luxury/object-in-focus-rare-earthenware-by-unknown-fields-division/ (accessed 21 March 2016).

6. David Hay, ‘The story of the first transatlantic telephone cable’ [online], BT. com, 26 September 2014 (accessed 21 March 2016).

7. See, for instance, Paulo Cirio and Alessandro Ludovico’s Face to Facebook (2011), Eva and Franco Mattes’ The Others (2011), and Owen Mundy’s I Know Where Your Cat Lives (2014).

8. See: Jon Thomson and Alison Craighead, ‘Horizon’ [online], http://thomson-craighead.net/horizon.html (accessed 21 March 2016).

9. Sexual life of a stable relationship (Jaime and Francisca).

10. Mitchell Whitelaw, ‘Art Against Information: Case Studies in Data Practice,’ [online] The Fibreculture Journal 11, 2008, http://eleven.fibreculturejournal.org/fcj-067-art-against-information-case-studies-in-data-practice/ (accessed 17 December 2014).

11. See: Julie Freeman, ‘We Need Us’ [online], http://www.weneedus.org/ (accessed 21 March 2016).

12. Zooniverse is a website that allows members of the public to assist professional researchers in tasks such a classifying large sets of images. See: https://www.zooniverse.org/.

13. See: Catherine M. Weir, ‘KODK’ [online], http://www.cmweir.com/portfolio/kodk/ (accessed 21 March 2016).

14. See: Ellie Harrison, ‘Vending Machine’ [online], http://v3.ellieharrison.com/index.php?pagecolor=3&pageId=project-vendingmachine (accessed 21 March 2016).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.