Between Re-production and Re-presentation: The Implementation of Photographic Art Reproduction in the Documentation of Museum Collections Online

Museums extend their visibility beyond the physical institutions by providing online collections. By doing this, the museums seek to make a whole collection accessible to visitors online and, as a result, to make the cultural heritage accessible to a broader spectrum of society. Although the collections are represented under the inherent conditions of the graphical user interface, online collections are based on earlier forms of representation, and have implemented their principles. To find precursors and influences on current online collections the following research questions are discussed: What influence has photography had on the inventory processes and the visibility of museum collections? What kind of remaking happened with the combination of text and image in classifying systems at the turn of the 20 th century? What effect has this early remake had on current online collections? This article explores one early remaking of a museum collection. The Museum fur Kunst und Gewerbe Hamburg (The Museum of Arts and Crafts Hamburg) started to use photography as an addition to text based classification systems. Contemporaneously the photographic reproductions were used as an illustration for the depicted museum objects in publications. The first employee of the museum, Wilhelm Weimar, primarily made sketches of museum objects and in 1897 started to photograph them. As a result 1700 glass plate negatives in three different sizes were produced, which mainly show museum objects or details of them. In this article the photographic art reproduction on glass plate negatives from Wilhelm Weimar are analyzed. In a second step the history of index cards introduces a classification system that became omnipresent in libraries and all kind of offices since the late 19 th century. In the Museum of Arts and Crafts Hamburg index cards were illustrated with images, and prints mounted on cardboard provided more accessible representations of the depicted objects, which were held in storage.

Museums extend their visibility beyond the physical institutions by providing online collections. By doing this, the museums seek to make a whole collection accessible to visitors online and, as a result, to make the cultural heritage accessible to a broader spectrum of society. Although the collections are represented under the inherent conditions of the graphical user interface, online collections are based on earlier forms of representation, and have implemented their principles. To find precursors and influences on current online collections the following research questions are discussed: What influence has photography had on the inventory processes and the visibility of museum collections? What kind of remaking happened with the combination of text and image in classifying systems at the turn of the 20 th century? What effect has this early remake had on current online collections?
This article explores one early remaking of a museum collection. The Museum für Kunst und Gewerbe Hamburg (The Museum of Arts and Crafts Hamburg) started to use photography as an addition to text based classification systems. Contemporaneously the photographic reproductions were used as an illustration for the depicted museum objects in publications. The first employee of the museum, Wilhelm Weimar, primarily made sketches of museum objects and in 1897 started to photograph them. As a result 1700 glass plate negatives in three different sizes were produced, which mainly show museum objects or details of them. In this article the photographic art reproduction on glass plate negatives from Wilhelm Weimar are analyzed. In a second step the history of index cards introduces a classification system that became omnipresent in libraries and all kind of offices since the late 19 th century. In the Museum of Arts and Crafts Hamburg index cards were illustrated with images, and prints mounted on cardboard provided more accessible representations of the depicted objects, which were held in storage.

Introduction
Art reproduction has a long history, encompassing many techniques. The appearance of photography in the middle of the 19 th century changed art reproduction irrevocably. From the invention of photography in 1839 onwards, works of fine art and other objects were made its subjects. Photography simplified the reproduction process of images since Henry Fox Talbot's invention of the calotype process (Fox Talbot, 1844). While the report of François Arago in front of the French Chamber of Deputies only imagined reproduction using the daguerreotype process, invented by Louis-Jacques-Mandé Daguerre, the calotype process enabled reproduction out of one negative (Arago, 1839: 53;Mirjam Brusius, 2014). This triggered questions over the value of the original compared with photographic reproduction. Shortly after the invention of photography, museums, with their large collections, hired photographers to document their collections or exhibition rooms; photo studios specialized in photographic art reproduction and created distribution networks; and museums changed permissions for making reproductions of their exhibits (Brusius, 2016: 162;Hamber, 1996: 393f;Hauswald, 2016: 99;Bader, 2013: 330-340). These photographs, taken for research use, functioned as a working base for custodians, and were produced on commission for publications. Today, these photographic art reproductions of the 19 th and 20 th centuries, kept in the museum or other archives, have themselves become significant witnesses to a particular period. As such, they have shifted in status from working materials to collection objects, as the interest in exploring photographic art reproductions increases. Thereby, the view held on these photographs also changes. The photograph of an artwork is no longer analysed as a proxy for the depicted object but as an independent object; investigations are now based on the material of photography such as presentation in the image space, lighting or the image material itself. With this twist, many institutions began to review the image material in their archives and look at them as a source for research.
The first museum photographer at the Museum für Kunst und Gewerbe Hamburg (The Museum of Arts and Crafts Hamburg) was Wilhelm Weimar. He started to document a mostly three-dimensional collection with an inventorying approach in  1897. While other well-known museum photographers (Roger Fenton at the British Museum and Charles Thurston Thompson at the South Kensington Museum, both in London) started decades earlier, in the Museum of Arts and Crafts Hamburg they chose photography as a medium of documentation only when it became easier to print and reproduce. At the end of the 19 th century the invention of the halftone process enabled the printing of photographs directly by simulating a continuous tone using different sizes and spacing of dots (Meggs, 1998: 141). Print photographs became more affordable, and this crucially was within the museum's budget. Standards on how to photograph two-and three-dimensional objects were established by experimenting with light, perspective, different cameras, and glass plate negatives. These shifts in both printing and technique changed the appearance of the museum's inventory.
Custodians started to work with prints mounted on cardboard that represented museum objects, and photographic prints were added to index cards.
In this article, I aim to show how the implementation of photography changed and enhanced existing systems of classification in the late 19 th century, using the Museum of Arts and Crafts Hamburg as a case study. Photographic art reproductions from this museum (glass plate negatives, mounted prints on cardboard or prints in publications) are the basis for my analysis. The techniques developed there changed the way we look at museum objects; one could say that this increase in images constitutes an important remaking of the presentation of museum collections. Furthermore photographic art reproductions were integrated in the grids of letterpresses and index cards; handwritten information was added beside the cardboard-mounted prints.
In this article I investigate the interaction between text and image around 1900, and the standards of how to photograph arts and crafts. When starting to digitize collections, text-based information was transferred first, with digital images being added gradually. The emerging databases assembled by museums were initially used for internal purposes only and based on older classification systems such as index cards or mounted prints. Today they often form the backup of museums' online collections. The term online collection is often used on museum websites, which Kreiseler: Between Re-production and Re-presentation 5 provide public access to a collection in its entirety instead of presenting a curated selection (Kreiseler et al., 2017). Museum objects, previously in storage and hidden from view, are made publicly available. In the following, when using the term online collection, it refers to museums' online collections.
The thesis of this article is that arrangements and elements in today's online collections are affected by these older classification systems. To test these assertions,

Background
To analyze photographic art reproduction and the network it was part of around 1900 addresses a diversity of research fields, including the history of photography, art history, history of science, media studies and digital humanities. Relevant topics from these fields are introduced below.
Papers and articles written about photography around 1900 mainly discussed photographic techniques -for example daguerreotype, calotype, heliography, or photogravure -alongside emerging chemical processes (Photographische Mitteilungen 1870-1895Photographische Rundschau 1878-1903Schmidt, 1902;Buehler, 1994;Nickel, 1959). Another main topic included different ways of photographing scenarios such as landscapes, portraits, architecture or threedimensional objects like sculptures or crafts (Wölfflin 1896, 1897, 1915, Schmidt 1902, Weimar 1906. Weimar was a part of the photographic community and he published many articles where he shared his knowledge about how to photograph different scenarios and use different plates, filters, or chemicals (1901, 1905, 1906, Kreiseler: Between Re-production and Re-presentation 6 1912, 1917. Furthermore, he wrote a book about daguerreotypes he collected for the museum (1915). This book is considered a standard reference even today (Klemm, 2004: 56).
The convergent development of art history as a discipline alongside the invention of photography brings a technique into focus which we are used to today: comparative visual analysis. With photography this technique could be performed with the help of lantern slides which became basic equipment for lecture halls in art history seminars and for public presentations. The artworks shown on the lantern slides produced interest rather than the medium of the slide itself or the network they were a part of (Caraffa, 2009;Tietenberg, 1999;Ratzeburg, 2002;Dilly, 2009;Roberts, 1993). Anthony Hamber describes this phenomenon: The photograph is a window on an original and, unlike current art-historical methodology, photographic historians frequently consider the assessment of the characteristics of the window as being as important as the appraisal of the original being viewed through it. (Hamber, 1996: 5) This article focuses on the characteristics of this window, and its influence on contemporary forms of collection presentation.
In art history but also in media studies discussions about a separation between original and reproduction (an art or craft work and its drawn or photographed reproduction) became stronger in the 1920s (Sauerlandt, 1974;Benjamin, 2007).
While sketches or engravings, both as a form of art reproduction and as used for distribution, were handmade and therefore originals in themselves, photographs produced with a mechanical apparatus and chemical processes were seen as a proxy for the original object they showed. The distinction between original and reproduction grew. Discussions arose on the nature of the original and how its perception changes through an increase in reproduced images (Benjamin, 2007).
Another topic is the connection between photography and the approach to creating "objective" images for the emerging and fast changing sciences, whether it be natural sciences or art history. The requirement for objective images grew to prove that it was possible to show visual material that was not influenced by human intervention and to show the "real" nature of the depicted objects. The emerging photographs seemed to be objective because an apparatus had made the image, and this apparatus should simply receive it as it was (Hauswald, 2016;Daston and Galison, 2017;Brusius, 2013).
Today, in online collections (part of discussions in the field of digital humanities) the separated photograph is not the sole focus, but arrangements of image and text; static and dynamic elements; or user behaviour are also considered (Kreiseler et al., 2017). More generally, the interfaces of museum websites are compared or social media activities examined (Lin and Gregor, 2006;Marty, 2007Marty, , 2008Padilla-Meléndez and Águila-Obra, 2013;Pallud and Straub, 2014). But, little has been said specifically about the historical influences of online collections and their presentation through graphical user interfaces (GUIs).
The following analysis draws on each of these topics to argue for the connection between early collection photography and current presentation of online collections.

Wilhelm Weimar and his Photographic Art Reproductions
The photographer Wilhelm Weimar was not a pioneer in art reproduction photography but he started when the process of taking photographs was made simpler and less expensive by the invention of dry plates. Earlier photography used a glass plate negative coated with Wet-collodion. This process "consisted of pouring collodion containing potassium iodide onto a glass plate which was then tilted until the emulsion formed an even coating." As Hamber describes, "once the plate had been coated with the collodion solution it was then immediately sensitised in a bath of nitrate of silver and the camera exposure taken" (1996: 80). This short description shows how much knowledge of chemical processes a photographer had to have and how laborious it was. Furthermore, all darkroom equipment had to be immediately available at the place where the photograph was being taken. The invention of dry plates allowed for easier handling and more time to develop the negatives.
Wilhelm Weimar began taking photographs in 1897, learning autodidactically.
Primarily he was an engraver and one of his key tasks at the museum was to sketch Kreiseler: Between Re-production and Re-presentation 8 objects in its fast-growing collection (see Figures 1 and 2). In 1883, Weimar was the first employee of the museum, which had been founded in 1874. Therefore, he worked closely with the museum's director and founder Justus Brinckmann (Klemm, 2004: 34, 55). Brinckmann later emphasizes the benefit of photographs: What we want, gentlemen, is this: not to make art history but to provide impeccable documents for historical research into the art history that is being  Kreiseler: Between Re-production and Re-presentation 10 filter for high-contrast museum objects and an obscured the window with paper to dim the light when photographing glass objects (Weimar, 1906: 187, 193f). His small studio was situated on the ground floor, with no electric illumination and the window was north-facing (Weimar, 1912: 540). Although many inventions simplified the photographic process, it was still time consuming. Yet, as a result of this, Weimar produced high-quality photographs, which could be used for different purposes.   Kreiseler: Between Re-production and Re-presentation 13 object in a shot. Therefore, he transferred the arrangements he learned from making sketches into photography.
As a comparison to the described photograph made by Weimar, a contemporary one of the art nouveau cup can be found in the online collection of the Museum of Arts and Crafts Hamburg (see Figure 8). An obvious difference between the old and the contemporary photograph lies in the advanced camera and negative material.
The contemporary image is a color photograph, whereas Weimar's is black and white photography. A smaller difference can be seen in the chosen perspective on the cup.
The contemporary photograph highlights the silver ornament in the middle, and the symmetry of the silver ornament is accentuated. Likewise, the background has changed. No horizontal line separates the foreground from the background. The illumination is soft and the shadow on the right is light. The color photograph was made professionally, but who made it is not recorded in the online collection of the Museum of Arts and Crafts to avoid image copyrights on the art reproduction photograph. at Konrad Gessner's Bibliotheca Universalis (Ibid.,(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24). Although in this Bibliotheca Universalis unmounted notes were organized and fixed in books, it was innovative in that every new thought was written in a new line (Ibid., 13). Not only thoughts but also related books and notes were listed. One invention that might have been influenced and fostered by the new system of index cards was letterpress printing, which involved the separation of each letter and their array in a type case. Krajewski draws a parallel to cabinets used by scholars. Here they saved their unmounted notes in note boxes or even in specially built cabinets, with single ideas separated but held together at the same time (Ibid., 19). The advantage of these arrays was the mobility they brought to the materials they contained. The existing order could be reorganized and new notes could also be added more easily than in books.
In libraries, one innovation was the implementation of catalogs as a search tool.
Even if these catalogs were themselves books, they enabled librarians to classify and rearrange books without changing the order of the library itself. The catalog made the library accessible because a whole book collection became searchable in one single book. Before this system emerged, books often had one specific place on a shelf, with labels on the shelves indicating the books they contained. Therefore, librarians had to know where a particular book had its place. One anecdote features Gotthold Ephraim Lessing finding treasures in numerous strolls like a 'human search engine' (Ibid., 32). He found these treasures by walking through the space, a privilege only a select group of people had.
In the 18 th century, two developments influenced the emergence of index cards as we know them. First of all, the size of each card was standardized in these note boxes for single cataloging projects (Ibid., 42). The precursors of these standardized cards were simple playing cards, which had a standard format and a white back.
Before librarians discovered them as a useful material, they were altered for both death announcements and business cards (Ibid., 33). An advantage of playing cards was their material. The cardboard was more stable than paper and therefore lasted longer. The second standardization concerned the information held on each note.
'The goal of a unified, adequate catalog can be reached only if one can ward off the risk that the data will be arbitrarily diversified whenever employees act randomly' Brinckmann transferred methods from the natural sciences to his increasing collection of arts and crafts in the museum. Equally, the method of using images as an addition to the metadata of object inventories can be attributed to his early experiences in natural science.
Among other things, he had created a visual repertory of images of insects organized by type and place of publication. From this early capability and experience came the illustrated catalogs of his museum, which can today be understood as containing exemplary illustrative images whose scientific accuracy was given the highest importance by Brinckmann, and which provided a model for related publications throughout the world. (Ibid.,16) Brinckmann knew that it is important not only to collect but to inventory, and this ideally included making sketches of the collection's items. The sketches would simplify the work of the director, and other employees of the museum, when they worked with the collection. Furthermore, loan requests coming from other museums or for exhibitions could be answered by sending images. A third role of the object images would be as sets of prototype models that craftsmen and artists could use for their work. In an early printed copy of the museum's regulations most of the paragraphs determine what can be reproduced or taken out of the cupboards.
Importantly, the regulations state that this collection of prototype models could be used without a written request (MKGH-Archiv DirBr 24, 1894-1927 9 and 10). The collection of mounted prints grew and metadata was added gradually, as is evident by analyzing different handwritings and  shift that turns a museum into a 'research institution' is reasonable but limited in the case of the British Museum. Only carefully selected scholars got prints of the clay tablets to decipher. Therefore, access was limited to a selected group. Access to both the original objects and the photographs was restricted by the trustees.
Only the museum's exhibition itself was open to the public and all social classes (Ibid., 239).

Photographs mounted and described on cardboard in the Museum of Arts and
Crafts Hamburg, and in many other institutions, transformed heavy original museum objects into mobile objects, standardized in their size and dimensionality. Museum objects could be easily compared without using the originals. This standardization raises the problem of a loss of sense for the size and material of the depicted objects. To address this Weimar developed a formula to calculate the reduction or enlargement a depicted object has in its photographic reproduction (1906: 189).
A caption in publications refers to the size of the objects (see Figures 11 and 12).
An advantage of the mounted prints on cardboard is that the arts and crafts objects are protected from damage. In the museum, custodians and interested people got Kreiseler: Between Re-production and Re-presentation 21 access to museum objects that might be hardly accessible in the depots. The separation of each object, on one mounted print or index card, enables many visual orderings, which cannot be arranged in the physical space. In archives and depots, the key task of preserving museum objects dictate an order, separating different materials to optimise the longevity and preservation of cultural artifacts (Deutscher Museumsbund e.V., 2006). Mounted prints made museum objects visible while protecting the originals, enabling different access points for collection users. These multiple access points support exhibition curation as a museum practice, as well as research.  On the index card the information is separated through lines where the writer could Another difference is the ability to create tags, which can provide more connections between museum objects than ever before. However not all information on the original index cards is provided online; for instance further readings can be consulted only on the index cards or in the Museum's internal database.
The fifth principle is access to high quality images. As described, every shot made by Weimar was supposed to be perfect, here meaning highly aesthetically refined. He only had time to take one shot; a second would probably have taken him Kreiseler: Between Re-production and Re-presentation 27 the whole day. Therefore he worked precisely and his photographs were in use for many decades for internal classification systems and in publications. Today, one would say his work was sustainable when producing high resolution images. By comparing the policies of different online museum collections, it is apparent that they provide different image resolutions. Whereas the online collection of the Rijksmuseum offers the possibility to download a jpg image ' of 4500 × 4500 pixels on average' and 'free The collection of the Museum of Arts and Crafts Hamburg was primarily intended to be a role model for craftsmen and trainees. They were allowed to take the museum objects out of their cupboards and had access to sets of prototype role models (for example ornament engravings which show patterns, ornaments and therefore styles of earlier stylistic eras). Generally in the physical space of museums in the 19 th century not everyone had the permission to work directly with objects or make reproductions of them (Bader, 2013: 330-40). Furthermore limited time, dependence on physical museum space and staff costs limited the access to objects. Today, the online collection of a museum allows everyone access, at any time, and at low cost. It is nevertheless essential that the collected information and images are added with professional care and in cooperation between experts and interested users.
But in contrast to the old physical classification systems, once the collection is digitized approaches to the objects can still be changed, and curatorial questions come to the fore in online collections.

Conclusion
Taking Weimar's photographs of arts and crafts objects as an example of an attempt to create an image inventory, this paper has drawn a new line between old photographic art reproductions, classification systems, and current online collections of museums.
Kreiseler: Between Re-production and  First, Weimar and his aspiration in taking photographs was (re-)presented.
The precision of his work can be seen through the example of the art nouveau cup, as well as through his knowledge of materiality and image structure. Second, the history of index cards introduced the classification systems that were used in museums in the early 20 th century. In the Museum of Arts and Crafts Hamburg's current internal database, much of the metadata is based on earlier index cards.