The Business of Data Protection

The effort to establish photography, especially in the form of microfilm, as an information technology was grounded in the hope that it could mobilise all sorts of original material that – as was the case with engineering plans – contained layers of information so dense that the content was impossible to translate into codes and thus needed to be preserved as an image. The investment in these utilitarian forms of photography emphasised the status of images not as mere illustrations but as central tools of knowledge production and management. 1Wolfgang Coy, “Die Konstruktion technischer Bilder: Eine Einheit von Bild, Schrift und Zahl”, in Bild – Schrift – Zahl, ed. Sybille Krämer and Horst Bredekamp (Munich: Fink, 2003), 143–53.

But, once again, the retrieval process was the Achilles heel of the micro technology: aperture cards, which combined the image and coded information, were costly to produce and tricky to maintain. The system could be set up for selected content but certainly not for all business documentation.

When microfilm took on a more passive role as a storage medium, the rhetoric adopted by the photography and IT industries pivoted towards data protection rather than data circulation. Photography was increasingly propagated as a means of protecting sensitive data, such as in the form of the COM system (Computer Output Microfilm). Unstable digital data from magnetic tapes was automatically converted into an abstract visual representation and saved on rectangular sheets of photographic film, known as microfiche (fig. 1), that could be stored as external memory.

The concept of data protection was also mirrored in the multiple product and service advertisements that highlighted the value of data in modern information society and the subsequent need to manage and protect these assets. The gathering of ever-increasing amounts of data went hand in hand with data protection. The response to the potential endangerment of original material (through data theft or other threats) was the creation of backup versions and the protection of the original material. As a 1958 advertisement for the private information management company Iron Mountain (fig. 2) claimed, “Records … they log a company’s past … mark its progress … determine its future.” No history, no future.

In the post-war and cold war context, this sentence followed a key rationale that has shaped archival and cultural heritage theories and policies over decades: Without history, without memory, there is no culture, there is no future. 2For a discussion of the inclusiveness and exclusiveness of archives, see Ariella Azoulay’s work, including her “Unlearning …” series for the Still Searching blog. Creating a business history and business archives was deemed necessary for the sustainability and expansion of a company’s operations. They also legitimised and cemented a company’s economic and cultural standing. At Iron Mountain, these records were stored as photographs, on large film rolls tightly packed and stacked several metres high; since the mid-1990s, Iron Mountain, which has turned into a multi-billion-dollar company operating worldwide, has specialised in digital data management. And it is only one of several in the field. 3See, among others, Monika Dommann, Hannes Rickli and Max Stadler, eds., Data Centers: Edges of a Wired Nation (Zurich: Lars Müller, 2020). Data gathering, data capitalism and server parks, after all, have been around for some time.

The digital form of photography, meaning the merging of image data and metadata, marked a paradigm shift with regard to the organisation, searchability and retrieval of images. An image file consists of encoded pixels, descriptive and administrative metadata, mechanically captured data (date and time of production, aperture, exposure time, camera type, etc.) and geodata. The informational status of an image can be further enhanced by adding captions, tags, copyright information, hashtags, likes and comments. All this metadata and attached information makes photographs legible to software. It constitutes the image infrastructure, the underlying condition for the circulation and networking of images. 4Daniel Rubinstein and Katrina Sluis popularised the term networked image in their essay “A Life More Photographic”, Photographies 1, no. 1 (2008): 9–10. They argue that the photographic image becomes a new type of photography in alliance with the Internet, through which it necessarily interacts with other media forms. See also Jonathan Beller, “The Programmable Image of Capital, M-I-C-I’-M’ and the Computer”, Postmodern Culture 26, no. 2 (January 2016), https://muse.jhu.edu/article/635542.

In the 1990s, when companies like Getty Images and Corbis aimed at disrupting the picture industry by investing in digital image distribution, their initial edge was based on their technological know-how. As tech companies, they invested heavily in the development of databases, linked data indexing and classification methods and the establishment of an e-shop system. In 2016 Corbis, initially founded and owned by Bill Gates, was bought by its major competitor Getty Images via the Chinese investor Visual China Group. Through the acquisition, the databank contents and the licensing rights for around 100 million images, including historical collections, news pictures, art reproductions and stock photography, were transferred to Getty Images. What the deal did not include was the takeover of Corbis’s employees, physical structures and analogue archives. it was a real coup in the eyes of Jonathan D. Klein, the president of Getty Images, who announced the merger on Twitter: “Almost 21 years but got it. Lovely to get the milk, the cream, cheese, yoghurt and the meat without buying the cow.” 5For a more detailed account of the Corbis / Getty Images merger and its implications, see Estelle Blaschke, “Bilder als Kapital: Corbis Getty Images und der digitale Bildermarkt”, Fotogeschichte: Beiträge zur Geschichte und Ästhetik der Fotografie 142 (Winter 2016): 49–54.

In order to sell the whole range of dairy goods, you do not need the cow, just the digital files, metadata and licensing rights. What sounds like the wet dream of a venture capitalist mirrors a development that took shape over the last three decades: data and metadata are key in organising the overabundance of images and, in digital environments, they increasingly determine what is searchable, and thus visible, or not. In the context of the picture industry, the searchability determines the value of an image. In a broader sense, photography as a medium plays an essential role here: not only is the camera an image-making device; it is also a highly efficient tool for capturing metadata automatically. This is also reflected in the camera design. While up until the 2010s only selected cameras collected the technical data (known as EXIF data), this is now standard in any digital camera (fig. 3). The same goes for geodata, and in recent years it has also become possible to pre-programme user and copyright information into the camera, so that the copyright is attached to each image in the metadata protocol.

Digital images, thus, encapsulate different levels of information that go way beyond the “surface” of what can be seen.