“THE EXCESS OF THE PHOTOGRAPHIC ARCHIVE”
In the late 1980s, Bill Gates, cofounder of Microsoft, recognized, as many others did, that the market for images was to become one of the growth sectors of the new digital economy. With the creation of Interactive Home Systems, which was renamed Corbis in 1995, he envisioned a “databank of a million pictures” comprising photographs and photographic reproductions of art, “unique and comprehensive archive of images of all kinds.” Targeting the individual consumer rather than traditional clients in the press, publishing and advertising industry, Corbis, and its major competitor Getty Images, accumulated vast amounts of visual material and copyrights to dispose of endlessly reproducible resources and by this, control the future dissemination of digital images. What made the digital form of photography (and film) so appealing to Corbis, as to anyone involved in the archiving and the management of photography collections and facing space, time and monetary constraints, both in the commercial and institutional context, was the idea of immateriality. The economic potential of the digital image relied on the idea of immateriality and the substitution of the material object. The article traces the creation of a digital image archive fuelled by the fantasy of “collecting everything” and the potential of selling it – and the ultimate failure in doing so. Through the case of Corbis, it will explore the implications of the economical paradigm on the archive and on the accessibility of collective visual memory.
The invention of digital imaging reanimated the fantasy that had infused photography from its early days on: the fantasy of ‘collecting everything’, of accessing “enormous collections of forms”, as Oliver Wendell Holmes had advocated for with regard to stereoscopic views, ‘classified and arranged in vast libraries, as books are now’. To facilitate the development of such collections, Holmes suggested the formation of a “comprehensive system of exchanges, so that there may grow up something like a universal currency of these bank-notes.” Like Holmes, Paul Valéry applauded the economic potential of photography and its siteless and time-less reproducibility when stating in his 1928 essay ‘The Conquest of Ubiquity’: ‘like science, it [the reproduced work of art] becomes an international need and commodity.’” Yet, understood in relation to the development of an increasingly democratic process, Valéry explicitly related the gains of mechanical reproduction to the empowerment of the individual. Liberated from the location of its performance and its temporal continuum, the mechanical reproduction would enable the individual not only to access and participate in the experience of art, but also to choose when and where to do so. Thus, as Valéry envisioned, various types of reproductions would soon spread into the private space, individually programmed, to ‘satisfy our needs in response to a minimal effort[…] at a simple movement of the hand, hardly more than a sign’, and as simple as ‘water, gas, and electricity are coming to our houses from far off.’
It is precisely this juxtaposition between the liberation of experience on the one hand and its privatization on the other hand that is illustrated in the digital painting by Louie Psihoyos, ‘The Information Revolution. 500 monitors’ of 2003 (Figure 1). Situated in the very centre of a panoptic structure, we see the dark silhouette of a person in front of a white, glaring light, comfortably leaning back in a chair. Conveying an atmosphere of leisure, the person is surrounded by a grid of colorful, brightly illuminated screens. While the image suggests a maximized visual experience, the boundaries between entertainment and control, monitors and monitoring, the observer and the observed are blurred. Thus, it may be interpreted as corresponding perfectly to the Zeitgeist increasingly present since the early 1990s: the idea of the ultimate transformation of everyday life facilitated by the progress of communications technology and its electronic apparatuses. Yet, it embodies, intentionally or not, the very critique of such a concept.
But rather than discussing the dichotomy between the enhanced accessibility provided by digital imaging on the one hand and on the other hand the loss of information in the process of digitization, this chapter focuses on the intrinsic relation between the fantasy of ‘collecting everything’ and the idea of earning money from it. It thus aims at drawing attention to the economic potential of both, archiving and reproducing photography. This idea of commodification seems to appear with every major technical development of the photographic method as a medium of storage and dissemination, creating new standards, new markets and new networks. Building on the recent scholarship that has been critically investigating the multiple material manifestations of photography, this chapter furthermore aims at exploring the idea of materiality with regard to the digital form of photography and the way digital materiality shapes the archive and is shaped by the archive. This relation and the implications of the economical paradigm on managing vast photographic collections will be examined through the history of the ‘visual content provider’ Corbis, a company, founded by Bill Gates in 1989 under the name Interactive Home Systems (IHS). What made the digital form of photography so appealing to anyone involved in the archiving and the management of collections and facing space, time and monetary constraints, both in the commercial and institutional context, was the idea of immateriality. The digital image was understood as an image without physical carrier, reduced to a binary code, an image ‘produced without the intermediaries of film, paper or chemicals.’ (Sasson 2004: 186). The economic potential of digital imaging as well as other forms of photography precisely relied on this: the idea of immateriality and enhanced mobility. As Holmes had suggested in 1859 in his plea for the sales potential of stereoscopic views: ‘Form is henceforth divorced from matter’. The anticipated potential of analogue photography lay in substituting an object by an image, and by this in replacing the ‘immobile and expensive’ form of an object depicted in the photograph. The same reasoning was applied to digital imaging. While the digital reproduction of images was
not considered the equivalent of the original photograph, its quality was adequate and could therefore be rendered marketable. And, since the picture market, in contrast to the art market of photographs, does not deal with the photographic object, but with the intangible rights for reproducing, using and publishing the images, the shift from the analogue to the digital was interpreted as a mere continuation of existing practices. Digital imaging promised to be a further step in the continuous search for higher efficiency in the reproduction technique, management and exploitation of images. Thus, it was assumed that the digital reproduction, compared to analogue ‘matter’, was mobile and inexpensive and as such, would inevitably replace of its analogue predecessor. And just as photography had superseded previous reproduction techniques, above all lithography and engraving and their related professions and businesses during the early nineteenth century, a new market with new rules would emerge and eventually replace existing businesses.
The idea of immateriality and enhanced mobility was so powerful that it ‘captured the imagination of the cultural heritage community’ and private companies alike (Trant 1995: 262). Numerous institutions and private companies involved in the archiving of collections sensed the opportunity not only to reduce the volume of their collection by preserving the digital surrogates, but also to earn income from their digitized holdings. Accompanied by a widespread debate on the impacts on knowledge production and the dissolution of ‘the archive’ as an authoritative body, digital imaging led to major financial investments and emergence of new business ventures. While most of the players in the traditional picture market, embodied by photography agencies and commercial image banks, were hesitant or simply did not have the financial means and the technological know-how to invest in digital technology, the possibilities of digital imaging were considered so promising, that it attracted companies alien to the picture market. Companies, such as Corbis and its major competitor Getty Images.
When forming the company IHS in 1989, Gates envisioned the creation of a distribution service for digital images, simulating and satisfying the alleged needs of modern society and the modern individual. In his book The Road Ahead published in 1995, Gates unfolds his vision of an information society and his motivation for creating IHS. The chapter ‘Cyber-Home’ presents in particular the construction plan for Gates’ own private home, which was planned to be equipped with “news coverage and entertainment at the touch of a button”. An individual choice of images, recordings, films and TV programs would be displayed on several synchronized wall monitors. Gates specifically mentions photography: as the first private user of a ‘databank of a million pictures’ comprising photographs and photographic reproductions of art, he imagined a “unique and comprehensive archive of images of all kinds.” With the help of wide-ranging data registers these images were intended to be easily retrievable. The establishment of IHS would therefore serve the implementation of the illusion of ubiquity for profit-making purposes. In contrast to most photographic agencies and commercial image banks, that primarily target the press, publishing and advertising industry, the IHS business model was designed for an additional and highly potent group of customers: the individual consumer.
Parallel to the idea of immateriality, the business model of the IHS distribution service was based on two assumptions: First, it relied on the hypothesis that the ‘information high-way’, whose infrastructure in the early 1990s was still rudimentary, would soon allow for the transmission and reception of large amounts of data, including images. This technological development would automatically stimulate the demand and create a desire for digital images and subsequently generate new marketing opportunities. IHS would benefit from this infrastructure by installing, as many feared, “a tollgate on the information super high-way.” And indeed, by heavily investing into the digitization and the technology for storing and disseminating images, IHS aimed at controlling the anticipated stream of digital images as such. Second, Gates believed that ‘just as software had replaced hardware as technology’s most valuable product, so too would content eventually replace instruction sets as a basis of digital value.’ Digital content, be it visual, aural or textual information, would become the capital of information technology. And this capital needed to be amassed.
For Corbis and Getty Images the imperative of the subsequent years was the accumulation of as much visual material as possible to establish a comprehensive collection of images, or as Corbis projected, a ‘digital Alexandria’ or an ‘Encyclopedia Britannica without body text’. The company’s aim was to be ‘the place for pictures on the Internet’, assuming that there was a clearly defined amount of images worth owing. The development of a particular thematic grid, or ‘gridded territory’ (Rose 2000: 564) determined on the images needed to built this place, a place composed of potentially profitable images, and, in view of the prospective clients, a place dominated by American and Western visual regimes and visual histories. To distinguish itself from existing photographic agencies and image banks, the Corbis collection would allow the
client to find everything in one place, thus saving time and accounting efforts. However, as the history of Corbis shows, the accumulation of ‘virtually anything imaginable’ proved to be far more difficult than anticipated and bore unforeseen consequences.
Banking on images
In the early 1990s, Corbis negotiated non-exclusive licensing rights with photographers and a number of museums to market their collections. The latter seemed particularly appealing: museum collections represented a pre-selection of acclaimed art works certified by the authority of the cultural institution and thus promised to be valuable and profitable. In turn, many museums, especially smaller ones were seeking co-operations with private or institutional partners at the time, as they were aware of the need to engage in the digitization of their holdings for promotion and editing purposes. Thus, the museums were to benefit not only financially from the arrangement, since part of the generated profit was to be shared with Corbis, but also through the indirect promotion of their collections. However, many museums balked, or gave their consent for a very limited period of time, as they feared a loss of control over the use of their holdings especially given that copyright legislation had yet to be adjusted to reflect the rapidly changing technology. The key concern of photographers and institutions was that Corbis initially claimed a separate copyright protection, arguing that the digital reproduction of an artwork or photograph could be considered as fundamentally distinct from its original. Through the potential adjustment of color, brightness, and contrast, the digital reproduction could be interpreted as a unique work of art. In addition, the museums feared that Corbis would take possession and commercialize a wide range of art works, which were regarded as common property. The company countered these allegations by stating that the copyright protection applied only to the digital image, the visual surrogates of the primary materials produced by Corbis and argued that it was not preventing anyone from reproducing and subsequently disseminating the same original work of art.
At that time, the marketing opportunities for digital images were virtually non-existent. Both, marketing opportunities and the desire for digital images had to be created. The initial product, comprising largely of digital reproductions of artworks, was first directed towards private individuals rather than professionals. It was only later that the company would target the established customers of the picture market, namely advertisers, editors, and designers.
One marketing tool, Corbis’s CD-ROM’s, such as A Passion for Art: Renoir, Cezanne, Matisse and Dr. Barnes, Volvanos: Life on the Edge and Leonardo da Vinci, were conceived as virtual multimedia exhibitions. Through the licensing of a selection of images, ‘students, teachers and surfers’ were encouraged to ‘create their own documentaries by doing moulded searches.’ This description suggests that Corbis had only a vague idea of the potential users and uses of digital images in the beginning. However, the sales figures for the CD-ROMs were hardly satisfying and bore no relation to the considerable financial investments in the digitization and the development of a comprehensive databank system. And, as one might expect, the concept of a separate copyright, the ‘divorce from matter’ and the appropriation of images through the digital medium was not viable. The amassing of visual content by means of cooperation with museums and photographers proved to be too complicated, too time-consuming and was soon to be abandoned.
From 1995 on, the year that Corbis adopted its name, the company was forced to reconfigure its business model to counter the rights issue and to gain a leading position in the picture market. What followed was the acquisition of a myriad of photographic collections and the takeover of a wide range of photographic agencies and commercial picture archives. The policy of acquisition and the merger of collections is a common practice in the history of photographic agencies and commercial archives: holding as many pictures as possible is the principle and the catalyst of the picture market. Yet, with the emergence of digital images and the creation of a new industry, the picture market was radically transformed. Owing to their substantial investments, Corbis and Getty Images absorbed a diversity of collections and companies replicating the pre-digital market environment. As a consequence, photographic collections of unparalleled scale emerged.
Corbis’ aim was to cover a wide variety of subject areas, including fine arts; political, social and cultural history; entertainment; science and technology. In 1995, Corbis purchased the Bettmann Archive, which had previously merged with United Press International (UPI), which in itself was an amalgam of various collection such as ACME and Pacific & Atlantic; in 1996 the holdings of the agency Sygma, with a corpus of approximately 40 millions images were added, as well as those of various smaller collections, such as the Ansel Adams estate and the Andy Warhol Foundation. Aside from the accumulation of material stock, such as the Bettmann Archive, UPI and Sygma, thereby alleviating problems of copyright, Corbis established several commission contracts for mining institutional collections, among them parts of the prints and photographs collection of the Library of Congress. Interestingly, both Corbis and Getty Images consciously sought out and purchased historic collections of 19th and early 20th century photography, in an attempt to recycle ‘visual history’ and exploit the vague notion of collective memory. Enriched through time, these photographs were considered highly valuable. They deepened the collection and provided the company with credibility. And with the digitization, these historical photographs gained an added market value – the digital form, and the ability to be used and reused in new and different ways. It is indeed through the digitization that these photographs were integrated into a new economic cycle. Moreover, in order to flesh out the thematic grid,
several photographers hired by Corbis set out to systematically capture potentially profitable objects, themes and sites that were yet missing from the collection. Of particularly promising subjects, various views and constellations were photographed: the Eiffel tower during all seasons; by day or by night; as close-up, panoramic or aerial view; with one person, two or more, crowded; in black and white or color, etc. These contemporary photographs added to the purchased photographs of the Eiffel tower in construction, portraits of its engineer, The Eiffel tower during various decades, drawn or painted. Taken together, Corbis pursued the approach of purchasing and commissioning photographs to be able to respond to any given request, and by this, building customer loyalty. According to Corbis’s corporate information, the company today represents the work of more than 30,000 photographers and has a stock that totals over 100 million images.
Operating at that time with the largest concentration of high-quality scanners, Corbis engaged in digitizing vast numbers of photographs. In 1996 the digitization of the purchased collections operated at full stretch. Soon, about 1,000,000 scans were at Corbis’ disposal; with 40,000 images added each month, the company was scanning around the clock.
As scholars have argued, the digitization and the digital image displayed on the screen reemphasize the illusion of transparency, privileging, once more, the perception of the image content, rather than its context. Digital imaging accentuates an aspect of photography that is indeed an important characteristic of the medium, but one, which may have prevented us from understanding the multiple functions of the medium (Edwards, Hart 2004):  that is its quality as a depictive device and as a means of creating multiple reproductions. The seemingly effortless reproducibility of digital images and the ease with which digital information is copied, altered and combined, has prompted the comparison of digital imaging with the idea of reproductive cloning. A concept that may well be invoked by the opening page of the Corbis website of 2002, ultimately pointing to the company’s understanding of the medium (Figure 2). The microphotography of spermatozoon buzzing in various directions and accompanied by the heading ‘The possibilities are endless’, construes the notion of digital imaging as a quasi-natural resource, endlessly reproducible, endlessly combinable.
The revenge of materiality
However, while the digital technology certainly accelerated the mobility of images and changed the way images were produced, stored, distributed and viewed, digital imaging was far from being ‘inexpensive’ and ‘immaterial’. Corbis, as many others invested in the digitization, the development and the recycling of historical collections, encountered a myriad of problems, mainly originating in the materiality of photography. These problems related not only to the materiality of the analogue holdings that Corbis had purchased, but also to the materiality of the digital images, an aspect that had been neglected for a long time. It is indeed only since recent years that the myth of the immateriality of digital imaging, and digital holdings in general, has started unraveling. Yet, the materiality of digital images is distributed across many more levels and appears more scattered, and is therefore harder to identify and to interpret. The materiality of digital imaging and its undeniable dependency from the analogue holdings unfold when describing the functioning of Corbis and the problems that the company encountered with the creation of their products.
The initial objective of reproducing 40,000 photographs per months and ultimately reproducing the entire collection was soon curtailed. Not only were the costs for the digitization, including the handling, the indexation, and post-production soaring, but further complications arouse from bringing together the different collections and their individual classification systems. Some collections were not necessarily conceived for re-use, information was missing or erroneous, files were lost, the copyright could not be traced or existed under different rights regimes. Corbis, like many others, had misjudged the difficulties in migrating the existing, often very heterogeneous metadata into a new visual database, and underestimated the very importance of both, a well functioning database and the development of an effective search engine. By primarily concentrating on the accumulation of image content and by conceiving the photograph and the digital image merely as two-dimensional resources, the company had underestimated the time and labor needed for turning digital data into a valuable product. Yet, the economic value does not only rely on the image content, but is primarily constituted through the information attributed and the services associated to the image. In other words: in the picture market, the value of a photograph or a digital image is composed of the image content and the way it is formed, interpreted, distributed etc., thus, the context it appears in. Both are intertwined and cannot separated. The ‘materialization’ of the digital image product, however, is not some kind of automated process, but needs to be carried out for each individual image and relies on the evaluation and interpretation by individuals. The materialization of the digital image is, therefore, not only the precondition for the commercial exploitation, but also for its very existence.
The heavy investment in the purchase of ‘content’ of the early years was followed by the comprehensive editing of the analogue and the digital material and the development of a ‘context’. This process is carried out to this day. The conversion of the holdings into digital form and their development into digital products consist in multiple steps. First the analogue material is thoroughly examined, with duplicates identified. Information related to the photograph (i.e. caption, photographer, copyright) are verified, researched and eventually corrected in order to render the image exploitable. In a second step, the Corbis editors select photographs for the digital reproduction according to their relevance, their re-use and sales potential, and the physical condition of the analogue negatives and prints. The photographs are digitized as highquality scans of approximately 60MB, serving as the raw, uncropped version. This master file is considered the ‘original’ digital file. The actual visual products are modified and resized
versions of the raw scan. Formatted in different dimensions, they are available as low-resolution image for ‘Web & Mobile’, as small, medium, or large size images. The image resolution, thus, infuses the intended end product and their potential uses. Following the digitization and the formatting, all available text information is incorporated into the electronic file, including the caption, the provenance, the available sizes as well as the copyright. The attribution of categories and the detailed indexing of the image are carried out by the editor and the indexing department. Thus, reflecting the methods of the traditional picture market, the creation of a digital image product is embedded in a detailed production process built upon the idea of economic efficiency and the division of labor. Finally, the digital images products are displayed on the Corbis website, the virtual picture store, searchable and retrievable in different formats and at varying prices.
In the display of the individual images on the Corbis website, the digitized photograph takes up approximately half of the screen surface and appears in combination with all sorts of information listed as ‘Image details’. Massive Crowd on Beach at Coney Island (Figure 3), for example, a black and white image of a crowded beach by the American news photographer Arthur H. Fellig (Weegee) is accompanied by information on the image category (archival), the historical collection the image belongs to (Bettmann), a condensed and more detailed ‘original’ caption and the digital archive number. The location (Brooklyn, New York, USA) and the photographer (Weegee) are hyperlinked, referring to the image classification and the searchable keywords. Special emphasis is put on the indication of the copyright, be it in the form of the digital watermark inserted into the image or information concerning the model and property release. The image, placed on the right-hand side, is framed by a dark grey background, evoking, though inversed, the viewing mode of a light table supported by the luminosity of the computer screen. Besides, the configuration of the screen design, combining both image and text, reminds the design of accession cards used in library or museum collections. The various information accompanying the image and its prominent display points to an essential, yet strangely ignored aspect: one does not search for an image, but for the text associated with the image. Reflecting the archival structure, image query tools have always been, and still are, largely based on text. Hence, the image Massive Crowd on Beach at Coney Island is visualized and materializes through the textual information associated to it. The image echoes the information attributed to the image, deduced from the analogue object, and its interpretation.
As with digital holdings in general, the materiality of digital images is furthermore articulated through the supporting structures required for the archiving, the display and the distribution of the products. As Manoff and others have argued, ‘We access electronic texts and data with machines made of metal, plastic, and polymer. Networks compose of fiber optic cables, wired, switches, routers, and hubs enable us to acquire and make available our electronic collections.’ It is especially the acute energy consumption for the maintenance of the servers, hosting large digital image collections, which has become not only a growing cost factor, but is also increasingly raising ecological concerns.
On a more abstract level, the rhetoric employed by Corbis for advertising their products and services may also be seen as indirectly contributing to their materialization, infusing them with value. Interestingly, this rhetoric oscillated for many years between two concepts: between the appreciation of the photograph as an object with a particular genealogy on the one hand, and the two-dimensional endlessly reproducible and ‘a-historical’ image on the other hand. In its search for a functioning business model and business philosophy, Corbis deliberately positioned itself between philanthropy and business, as evidenced by the visual and verbal rhetoric used by the company.
In The Road Ahead Gates argues against the concerns of museums regarding the uncontrolled proliferation of digital images by claiming that the ‘exposure to the reproductions is likely to increase rather than diminish reverence for real art and encourage more people to get out to the museums and galleries.’ This argument recalls André Malraux’s idea of the ‘Musée Imaginaire’ (Museums without Walls) and the potential of photography as a medium for studying and popularizing art. Yet, as Doug Rowan, the former CEO of Corbis stresses in an interview from 1996: ‘This is not a ‘not-for-profit’ organization’. And Bill Gates is no altruist. Tellingly, the allusion to the concept of the library appears repeatedly. The coined reference to the Library of Alexandria, for instance was eagerly picked up by the media during the mid-1990s, picturing that ‘[Corbis] may swell into the world’s most comprehensive digital reserve of the imagery of mankind.’ For the magazine Wired, ‘Corbis is more than the ultimate digital stock image house. It may be the first online, for-profit library’. In turn, Corbis viewed itself as ‘the prototype of an all-content-on-demand, public access, private library’, further blurring the line between public and private ambition.
Corbis used the idea of the library as an institution dedicated to the public by turning it into a tool of its own self-promotion, and as a catalyst for the digitization. However, the functioning of the company and in particular the organization of its holdings clearly contradicts this idea. First, it benefits stakeholders and not the general public. Second, the Corbis digital archive is not structured according to the metonymic principle of the library, the ‘Nebeneinander’ of elements (juxtaposition), but operates with a hierarchical system, as will be developed later on.
Yet, Corbis does not use this cultural vocabulary for mere marketing purposes. With the references to the Library at Alexandria and the idea of the Museum without Walls, and particularly the liberation and promotion of the work of art through the digital reproduction, the company seeks to legitimize its business model. The digitization as a technological invention, one strongly supported and developed by Corbis as well as Getty Images, is framed within a wider cultural history of technological innovation. These efforts of legitimization emerge at a time, the mid 1990s, when the commercial benefit of digital reproduction, namely its sales potential is still purely hypothetical, and at a time of considerable unease with regard to the new technology. This dichotomy and the appropriation of collective memories becomes evident with the illustration for a Newsweek article, showing a color negative of then Corbis CEO, operating a view camera directed towards us, the spectator (Figure 4). With this representation Corbis explicitly seeks to position its efforts not only into a history of photographic techniques. It also creates a clearly delineated, rather plain, but reassuring visual narrative of human and political achievement, arts and entertainment, deeply entrenched in Western visual history. Corbis mobilizes cultural history as a marketing strategy to legitimize its policy and, by this, alleviating anxieties on the consequences of a new technology. It also mobilizes cultural history to counter the widespread criticism on the appropriation and control exerted by Corbis, and private companies in general, over visual cultural heritage.
However, in addition to the soaring costs for the development of viable products and the archiving, Corbis experienced additional problems with the market for digital images itself. As mentioned earlier, the demand for digital images was developing rather slowly. The transition from a business relying on the service and the direct contact between the client and the agency editor to an electronic commerce based on the individual search and the electronic delivery faced considerable resistance, especially among professionals. This resistance also resulted from the fact that this shift threatened to, and indeed did, significantly reduce the number of picture editors and, thus, replace an established profession.
As the example of Corbis shows, the materiality of digital imaging manifests itself in the visible and invisible metadata attributed to the images, the copyright, as well as the software and the hardware required for their archiving and distribution. The characterization of photography as a ‘multilayered laminated object’, as suggested by Joanna Sasson and others, could therefore also be applied to digital imaging. It is especially the merging of collections and the development of digital imaging products that surfaced the various layers of photographic materiality. And although digital imaging may be perceived or considered as flat or ephemeral, it produces ‘matter’ that is indeed material when considering the substantial investments needed for the development of digital imaging products, the maintenance of the collection as well as the efforts put into the conservation of the digital data. What also becomes evident is that a digital image does not have a value in itself, but needs to be ‘materialized’ to become valuable. Yet, this value, or ‘exchange value’ and the conversion of an image into a ‘visual currency’ depends on the development of a demand and of market structures, but developing these structures proved to be arduous, complicated and time consuming.
The Corbis example also reveals – and this is also valid for others archives and collections, be it in the commercial or the institutional context – that the digitization does not replace the analogue holdings, but forms parallel archives. The digital archive is not a replica but rather a ‘trace’ of the archive. It establishes its own archival paradigm by distinguishing between a digital ‘master’ file and various compressed versions, between the ‘original’ and the ‘copy’. Thus, on the one hand Corbis preserves the analogue holdings (such as the Bettmann Archive, UPI and Sygma) and maintains a digital archive. In the case of Corbis, which transferred its analogue holdings to Iron Mountain, a specialized long-term preservation facility located in a former limestone mine in rural Pennsylvania (Figure 5) in 2001, the digitization has resulted in the geographic divide between the conservation and archiving of photographs, and the circulation and distribution of their digital surrogates. Yet, the artifact does not become obsolete with the digitization. It is quite the contrary. The information related to the photographic object, in particular the copyright indication, and its initial context and use are crucial for their very existence in digital form and a precondition for building a product from digital data.
From a total of 100 million photographs owned or managed by Corbis, ‘only’ four million pictures are displayed on the website today, which drastically reduces the number of visible and circulating images belonging to the Corbis collections.
The question of accessibility is indeed pivotal. On the one hand this question relates, of course, to the control and the ‘authority’ of the archive. On the other hand, the archival systems providing access and allowing the fast retrieval of the holdings are an essential part of the economic potential of archiving. In the commercial context, an image is worthless, if it cannot be found quickly. And an image is non-existent, if it cannot be found at all. Thus, the economic potential of archiving relies not only on the accumulation and long-term storage of holdings, but is also reflected in the various tools and methods for managing and accessing a collection. The history of image query tools is particularly illuminating with regard to photographic agencies and commercial image archives, as their business is largely based on the pertinence and the fast retrieval of their holdings.
In the past, the service of image providers consisted of searching for and filtering a selection of pictures according to the clients’ needs. Essential to this, was the specific knowledge of the archivists and editors that derives from dealing with a collection on an every-day basis. In the case of digital archives, however, it is the client himself, who performs this task with the help of electronic search engines. Consequently, the primary concern and, and the very problem for commercial image suppliers lies in rendering their products as accessible as possible and in navigating the client through the vast quantity, the visual oversupply, the plethora of digital images. The challenge for Corbis with its four millions digital holdings, as with most digital archives, is displaying, on the one hand, the abundance of the collection and the variety of images the company has to offer. On the other hand, the search engines must provide not only a relevant selection, but must find the image.
The development of editing and managing tools for these massive picture collections, namely the development of electronic databases and efficient, user-friendly search engines is not a straight story forward success, but one of continuous experimentation and slow progress. The changing interfaces and search mechanism of the Corbis website bears witness to this process. The search field for entering keywords for instance, was integrated rather late, in 1999, two ears after the launch of the first Corbis website, which mainly functioned as a billboard for the company’s URL address. (Figure 6) The advanced search options appear more prominently on the portal in the design of 2002 and developed into a multi-optional search field, as shows the current website, combining the search by keywords with search options related to the location, date, photographer, collection, availability, copyright and formats, among others (Figure 7). To respond to different search scenarios, the client is given several options to access and search the collection. With its menu unfolding and the various fields and boxes, the present interface indeed reminds of a form, or control board, if not a mixing console. Corbis’ quest for an evermore efficient search engine points not only to the difficulties in finding images, especially with regard to the radical increase in digital image production; it also demonstrates that image research tools are becoming more important than ever in the accessing and use of digital image archives.
Yet, regardless of the on-going improvement on the search options and the revision of metadata, the Corbis database and the search engine are far from being flawless. The search results often generate a too broad selection lacking pertinence and include repetition. While this is due to the sheer quantity of images corresponding to the image query, it also results from the fact that each image is conceived and treated individually within the digital archive, not considering if it formerly belonged to a series of photographs of one object, topic or event. Moreover, the photographic image has, paradoxically, proven to be quite resistant to its indexing and retrieval, although it has widely shaped archival practices as an efficient medium of information storage. The polysemic nature of photographs, and visual representations in general, often hinders unambiguous classification, especially with regard to large collections.
In recent years, Corbis and other companies involved in digital information and image management have therefore developed additional tools and methods for structuring the data overload. While the continuous efforts in developing a mechanism capable of recognizing the image content by means of color and form are still in progress, Corbis is structuring its vast visual corpus through the rating of images. The rating is hidden to the person accessing the digital archive. The images are assessed with regard to their sales potential and accumulated revenue, to their present-day relevance and artistic quality. Besides the labeling with a basic category, among others fine arts, archival or entertainment, the editor ranks an image according to one of the five levels: the highest rating carries the abbreviation SS for ‘Super Showcase’. The rating of the images has become a crucial tool for structuring the masses of images, as it determines the order in which the search results are displayed on the website.
With the creation of a rating system and other marketing tools to highlight certain collections, the digital archive functions according to a hierarchical system. This system replaces the metonymic system, the principle of the library, which characterizes the analogue photographic archive. However, in an attempt to counterbalance the effects of this hierarchization, that is the reduction of the ‘visible’ digital holdings, and in order to underline the depth and variety of the Corbis collections, the search engine mixes different picture categories, be it documentary, archival, current events, entertainment, etc., and includes lower-rated images along with ‘picture gems’.
Reflecting the industry’s current development, Corbis also works on the widespread ‘profiling’ of its customers and the analysis of search behaviors. With the access to the Corbis website and the image inquiry, and in particular through the registration, the researcher or client leaves numerous traces behind, also referred to as the ‘digital footprint’. These traces are analyzed in order to keep track of the client, but more importantly to automatically anticipate his potential request, his ‘taste’ and consumer attitude. While one may argue that the profiling enhances the navigation, it is also seriously challenging the paradigm of the archive. Because, instead of providing more or less unbiased and universally valid search results, the results are shaped according to the individual client. The results, the visualization of the images, are the sum of one’s previous inquiries and search behaviors. This also means that, to some extent, the client is mirroring himself in each image or piece of information requested. As part of this individualization of the results of an archival request, the displayed selection is furthermore molded according to the client’s specific location, i.e. accessing the Corbis website from the United Kingdom compared to France, for instance, influences the search and its results. Consequently, electronic databases and search engines are in the process of developing from a simple text based documentation to a multilayered mechanism of visible and invisible information and metadata, and of predetermined choices and decisions.
In consequence, Corbis, as any commercial archive in the digital age, exerts control over its collection in two ways. First, it controls the access to its analogue collection by deciding on what is digitized and what remains only in an analogue form. In this respect, the digitization may be regarded ‘as an insidiously repressing technology, enabling institutional control over what is made accessible’ With this in mind, it was especially the transfer of the analogue holdings to the remote preservation facility Iron Mountain that was widely condemned in the media and in the writings and works of artists. Given the geographical location, this transfer theoretically limited access to the analogue holdings and reduced the ‘visible’ pictures to those selected for digitization. An idiosyncratic search, vital to any scientific examination of the collection is thus considerably limited. Moreover, the disproportion between the digitized and non-digitized holdings is rather unlikely to improve, since Corbis has shifted, like many others, from the systematic digitization to more careful and selected projects, such as
the digitization of card catalogues and reverse sides of photographs as well as the digitization on demand. Second, the control is wielded through the search engine. The current development shows that the control over a collection is exercised not only by controlling the digital technology, but also the technology to render a collection accessible through the programming of a search engine – ‘a power that has equal potential to be democratizing and passive, or repressive and active’. Thus, the question about the ‘ownership of the printing press’ that determined ‘the politics of the use and the access to the images’ has shifted towards the technology used for finding and searching for images.
Today, the references to a ‘digital Alexandria’ have been dropped from the Corbis business rhetoric. One reason for this may be that the visual content providers face tough competition from a new generation of digital image archives, such as Flickr, Google Images, Facebook and Youtube. In under five years, the picture sharing website Flickr, for instance, has accumulated more than four billion digital images provided by and exchanged among their users. Many professional photographers and public institutions use Flickr for promoting their collections, and since 2008 Flickr has been in cooperation with Getty Images in the area of image licensing. But as the example of Flickr shows: lacking the archival authority (in the form of consistent classification and indexing), the problem of finding a specific image has become even more complex.
The in/discipline of the archive
In conclusion, one may claim that the efforts and difficulties encountered by Corbis in establishing a commercial archive and turning it into a profitable business exemplifies what scholarship has characterized as the ‘in/discipline of the archive’ (Rose 2000: 567). This in/discipline reveals itself in the materiality of both, the analogue and the digital form and the idea that a photograph cannot be conceived without taking into account its performance within the archive as well as the fact that an archival system is predicated on the manual efforts of the archivists and editors, and by extension, their know-how and rigor. This rather fragile dimension of the archive tends to be covered by and contradicts the concept of an archive as a robust, authoritative body.This fragility may also point to the discrepancy between the notion of a quasi-automated, ideal (and idealized) archive and to the actual practice of the archive, especially in a commercial environment. Being an eclectic compilation of collections and lacking the institutional authority, the Corbis digital archive has no epistemic value per se. Through the abundance of styles, themes and categories (Figure 8) and the individualization of search results it becomes a serendipitous juxtaposition of images, an archive of ‘everything’ that is potentially marketable and that has been sold in the past, reflecting and nourishing the notion of taste in consumer culture. The sheer impossibility in setting boundaries and creating a narrative impairs on the essential function and purpose of an archive, that of ‘making sense’. Yet, the Corbis endeavor informs about how the reproducibility of photography repeatedly activates the fantasy of the archive; about the ways by which photographs and digital images are transformed into commodities; about the economic potential of reproducing and archiving, and the ambiguity of photography as an object and as a medium of the archive.
 Bill Gates, Nathan Myhrvold, and Peter Rinearson, The road ahead (New York: Viking, 1995).
 Oliver Wendell Holmes, “The Stereoscope and the Stereograph,” The Atlantic Monthly 3 (June 1859): 738–749.
 Ibid., 740.
 “La Conquête De L’ubiquité,” in Oeuvres, 2nd ed. (Gallimard, 1960), 227. (Original quotation: Telle que la science, elle devient besoin et denrée internationaux.)
 See: Nicholas Negroponte, Being Digital, 1st ed. (Vintage, 1996); Carl Shapiro and Hal R. Varian, Information Rules: A Strategic Guide to the Network Economy, 1st ed. (Harvard Business Press, 1998); Lev Manovich, The Language of New Media (Boston: The MIT Press, 2001).
 Geoffrey Batchen, Burning with desire : the conception of photography (Cambridge, Mass.: The MIT Press, 1997); Elizabeth Edwards and Janice Hart, eds., Photographic Materiality in the Age of Digital Reproduction (London: Routledge, 2004).
 Joanna Sasson, “Photographic Materiality in the Age of Digital Reproduction,” in Photographs objects histories : on the materiality of images, ed. Elizabeth Edwards and Janice Hart (London: Routledge, 2004), 186. See also Joanna Sasson, “Photographic Materiality in the Age of Digital Reproduction,” in Elizabeth Edwards and Janice Hart, Photographs, Objects, Histories. On the Materiality of Images, London: Routledge, 2005, pp. 186-202.
 Oliver Wendell Holmes, The Stereoscope and the Stereograph, The Atlantic Monthly, N°3, June 1859, 738-749.
 Jacques Derrida, Of grammatology (Baltimore: Johns Hopkins University Press, 1998).
 Gates, Myhrvold, and Rinearson, The road ahead.
 Ibid., 257.
 Ibid.; See also: Anne Friedberg, The virtual window : from Alberti to Microsoft (Cambridge, Mass.: MIT Press, 2006).
 Jane Lusaka, Susannah Cassedy O’Donnell, and John Strand, “Whose 800-lb. Gorilla Is It?,” Museum news. (1996): 34.
 Richard Rapaport, “In His Image,” Wired, November 1966. Original quotation: But the philosophical underpinning of Interactive Home Systems and its later incarnations – first Continuum, then Corbis – was based on a grander notion: Gates’s belief that just as software had replaced hardware as technology’s most valuable product, so too would content eventually replace instruction sets as the basis of digital value.
 David Bearman, “Archival Strategies,” The American Archivist 58, no. 4 (1995): 261.
 “Initially, Corbis wanted only “selected images from selected classis”, Swan says. The more they saw, however, the more they wanted. “Then they wanted all the images from the selected classic.” See foot note 21, p. 76.
 Lusaka, Cassedy O’Donnell, and Strand, “Whose 800-lb. Gorilla Is It?”. Original quotation: “The reason Corbis is working with museums is because we believe they are a critical element to have in the archive, an asset we will use to make money through re-licensing commercially and making products – such as CD-ROMs – for consumers.”
 Patricia Failing, “Brave New World Or Just More Profitable?,” ARTnews. 95, no. 9 (1996): 114.
 “Whose 800-lb. Gorilla Is It?”; See further reading see also: Barbara T Hoffman, Exploiting images and image collections in the new media : gold mine or legal minefield? (London, U.K.; Cambridge, Mass.; London, U.K.: Kluwer Law International ; International Bar Association, 1999).
 Rapaport, “In His Image.” Original quotation: “Students, teachers and surfers would be able to create their own documentaries by doing „moulded searches“.
 Consequently, Doug Rowan, former chef executive of Corbis stated that „CD-ROMs technology is seen as an expedient – training for the ultimate transition to online.“ See Rapaport 1966.
 The company name Corbis refers to the Latin word ‘corbis’, meaning basket. The choice of this name can be interpreted as a reference to the ‘shopping cart’ that will later become the icon of electronic commerce.
 In the early 1990s the picture market is composed of a variety of smaller agencies, specialized on specific themes, photojournalistic news agencies as well as stock photography agencies, such as Comstock and The Image Bank, a subcontractor of Eastman Kodak. Stock photography as a genre appeared in the early 1970s. Drawing its name from the concept of producing and storing great quantities of images, stock photography aims at providing inexpensive, often stereotypical pictures, mainly for the advertising industry. Anticipating the radical transformation in the field of communications, Corbis und Getty Images quickly become major competitors on the picture market, as most traditional agencies had underestimated the impact of digitization.
 See Corbis corporate fact sheet, updated November 2009 [www.corbis.com]. One should bear in mind that these figures remain vague, as this information is essentially addressed to the press.
 Scanning technology used : Heidelberg Topaz, Creo Scitex Ever Smart.
 Marlene Manoff, “The Materiality of Digital Collections: Theoretical and Historical Perspectives,” portal: Libraries and the Academy 6, no. 3 (2006): 311–325.
 Edwards and Hart, Photographic Materiality in the Age of Digital Reproduction.
 Geoffrey Batchen, “Photogenics/Fotogenik,” Camera Austria, no. 62–63 (1998): 5–16; Batchen, Burning with desire.
 Manoff, “The Materiality of Digital Collections,” 312.
 Gates, Myhrvold, and Rinearson, The road ahead, 259.
 André Malraux, Le musée imaginaire (Paris: Gallimard, 1965), 123. English translation: The reproduction does not compete with the masterpiece: it evokes or suggests it] „La reproduction ne rivalise pas avec le chef-d’oeuvre présent: elle l’évoque ou le suggère.”
 Rapaport, “In His Image.”
 Sasson, “Photographic Materiality in the Age of Digital Reproduction,” 186.
 Edwards and Hart, Photographic Materiality in the Age of Digital Reproduction, 5.
 The burden of representation : essays on photographies and histories (Amherst: University of Massachusetts Press, 1988).
 Corbis distinguishes between two major categories, that is “creative” and “editorial”. There exist two subcategories for “creative” pictures, namely Rights Managed and Royalty Free and five subcategories for “editorial”: documentary, fine arts, archival, current events and entertainment. See also Frosh, Paul. The Image Factory: Consumer Culture, Photography and the Visual Content Industry. Berg Publishers, 2003 and Bruhn, Mattias. Bildwirtschaft: Verwaltung Und Verwertung Der Sichtbarkeit. Weimar: VDG, Verlag, 2003.
 The different levels are marked SS (Super-Showcase), S (Showcase), B, C, D; most pictures belong to first three categories.
 Gary Hayes, “Under Iron Mountain. Corbis Stores ‘Very Important Photographs’ at Zero Degrees Fahrenheit,” National Press Photographer’s Association, n.d., http://www.nppa.org/news_news_and_events/2005/01/corbis_cave.html. Original quotation: Several freezers currently protect 28,000 “Very Important Photographs” at zero degrees Fahrenheit, including some famous gems like “Albert Einstein Sticking out his tongue” and “Marilyn Monroe Having Skirt Trouble on a Subway Grate.”
 Sasson, “Photographic Materiality in the Age of Digital Reproduction,” 187.
 See the most prominent works addressing the issue: Hal Foster, “The Archive without Museums,” October 77 (1996): 97–119; Batchen, “Photogenics/Fotogenik”; Allan Sekula, “Between the Net and the Deep Blue Sea (Rethinking the Traffic in Photographs),” October 1, no. 102 (2002): 3–34. Further note artworks by Alfredo Jaar: Lament of the Images (2002), installation work displayed at Documenta XI, Kassel, 2002 and Ines Schaber: Culture is our business (2004), photographs and dia-projection presented in the framework of the group exhibition “No Matter How Bright the Light, the Crossing Occurs at Night” at KW Institute for Contemporary Art, Berlin, 2006.
 Corbis solicited a group of photography conservators to develop a preservation plan to stop the deterioration process of large numbers of negatives. In keeping with this plan, in 2002 Corbis transferred the Bettmann Archive and UPI files to Iron Mountain, an underground storage facility located in a former limestone mine north of Pittsburgh. The information protection and storage company, also called Iron Mountain, holds the documents and data of approximately 2,300 clients, including government departments, private companies as well as libraries, museums and media corporations. Temperature and humidity controls ensure the long-term preservation of the analogue materials. For a selection of approximately 28,000 ‘icons of photography’ – a set of vintage negatives of bestselling and best known pictures – the preservation plan foresees storage at -4 degrees Fahrenheit, thereby freezing the negatives. The preservation plan for the Bettmann Archive and UPI was also the model for the long-term preservation project, Sygma Initiative. The analogue archive of the former picture news agency is preserved in Garnay, in Paris since April 2009.
 Sasson, “Photographic Materiality in the Age of Digital Reproduction,” 187.
One would certainly need to question if the concept of the archive applies with regard to the participatory websites mentioned in this paragraph. See André Gunthert, “L’Image Partagée,” Études Photographiques, no. 24 (November 2009), http://etudesphotographiques.revues.org/index2832.html.
 See also Allan Sekula, “The body and the archive,” in The Contest of meaning : critical histories of photography, ed. Richard Bolton (Cambridge, Mass.: MIT Press, 1989); Foster, “The Archive without Museums.”