[NOTE: This section is based on interviews with the following staff members at the Metropolitan Museum of Art on July 14-15, 2008: Doralynn Pines, Associate Director for Administration; Barbara Bridgers, General Manager for Imaging and Photography; Andrew Gessner, Chief Librarian of the Image Library; Peggy Hebard, Senior Financial Manager for Images and Publications; Billy Kwan, Associate Museum Librarian in the Image Library; Shyam Oberoi, Manager of Met Images; Julie Zeftel, Museum Librarian in the Image Library.]
In March 2007, the Metropolitan Museum of Art announced a “pioneering initiative to provide digital images to scholars at no charge.”1 What background research, infrastructure enhancements, financial analysis, and internal discussions led to this decision?
The Metropolitan Museum was one of the first major museums to recognize and embrace the potential of electronic management of collections information. Working with Gallery Systems, the commercial vendor of The Museum System (TMS) software,2 the museum created a fully automated inventory of objects in the textile collection that was launched simultaneously with the 1995 opening of the Antonio Ratti Textile Center. The records, many of which were accompanied by images, provided the staff and public with virtual access to all the textiles, including those rarely on view due to their fragility. With this momentum, the remaining curatorial departments were brought online one by one as separate TMS databases. While this aided management of the individual curatorial collections, the goal of a museum-wide database was unfulfilled. Rather than attempting to merge all the rich but non-standardized information from the separate curatorial TMS databases, the Met ultimately created one additional database and mapped into it only basic descriptive information from the sixteen TMS databases. This centralized collections database represents the collections information that the respective curatorial departments have approved for public access.
At the same time the museum was investing in collections management, it was also developing its capacity for digital imaging, thereby transforming the capture, management, and storage of object, event, education, installation, construction, and renovation photography. For over twelve years the museum has employed digital imaging consultants to steer planning and equipment purchase and to train and support staff.3 This investment in outside expertise has helped alert the staff to industry trends and developments that may have an impact on imaging operations.
Outfitting each of twelve photographers with a digital studio requires a capital expenditure of $150,000–$165,000, but museum staff members are convinced about the return on investment. Using a film camera formerly took up to four days to photograph a three-dimensional object. Today, using a digital camera, the same work can be shot in less than one day.4
It is not uncommon for less than 20 percent of any museum collection to be photographed, although museum professionals agree that object photography is a critical means of documenting and publishing the collection. The photographers at the Metropolitan Museum are now producing six to ten photographs of three-dimensional objects per day and an even greater number of photographs of two-dimensional works. This dramatic rise in efficiency helps balance the cost of digital equipment and results in increased photographic documentation of the museum’s treasures.
The price of film and processing for analog photography continues to rise, but direct digital capture eliminates most of those expenses. Furthermore, some pre-press costs associated with publication are reduced or eliminated when working digitally.5
Simultaneous with the conversion of analog to digital capture, staff initiated a retrospective scanning operation in the image studio. Selection criteria prioritized photography of works being included in the Collection Database and the highly acclaimed Timeline of the History of Art portions of the museum’s website.6 Photography for these works had the added advantage of descriptive captions recently written and/or vetted by the curatorial staff. Scanning was also undertaken on photography slated for publication in upcoming collection and special exhibition catalogs. The most recently produced color transparencies were favored over older photography in hopes of avoiding the need for extensive digital touch-up of film marred by particulates and scratches. In between color scanning projects, slow but steady progress has been made on converting the one-hundred-year-old archive of black-and-white negatives to digital format. The 35mm film is not being scanned at this time.
The digital imaging initiative benefits the museum in multiple ways. The number of digital images available for the website, publications, and internal use has been dramatically increased. It ensures access to images on film that celebrate the museum’s own history, a story covering almost 140 years and told by the visual documentation of people, events, gallery installations, special exhibitions, building construction and renovation, and educational programs. Digital surrogates reduce the handling of negatives and transparencies that have been moved to climate-controlled cold storage, thus increasing the longevity of these unique film masters.
Internal discussions about the Met Images project began long in advance of the official launch in fall 2007. Planning involved a team of staff members from Information Systems and Technology, the Photograph Studio, the Image Library, and curatorial departments working together to define and implement an enterprise-wide system for managing digital images. Shyam Oberoi, formerly manager of the Met Images project, describes the goals as twofold:
Initially, a third goal had been identified— increasing revenue streams for licensing of museum images. However, museum administration provided early feedback that this did not occur, urging that the staff team focus less on revenue generation and more on the value of preservation of, and access to, the digital assets being created throughout the institution.
Once Met Images was approved, work began on selecting the appropriate digital asset management system (DAMS). Interwoven’s MediaBin8 was ultimately chosen as the system that could:
Staff determined that MediaBin would be the repository for images, including data about the images and rights information; and object information, including artist name, nationality, life dates, object basic description, title, date, materials, and dimensions.
Certain work-arounds to MediaBin’s data structure were required to support the complex data relationships inherent in TMS, such as repeatable fields and whole/part relationships. TMS object information was ultimately exported into a data file that contains a non- relational, flattened record for each museum object. Nightly uploads from the data file to MediaBin were scheduled to capture edits to existing records and addition of new acquisitions. Loading the digital images into MediaBin was also complex. The photography studio had approximately four thousand CDs and DVDs containing two hundred thousand images. Accompanying spreadsheets provided the link between the images and the objects, but the task was laborious because the data lacked consistency. After the data and image files were loaded in MediaBin, a script was run to establish the association between images and records from the TMS extract data file. This simplified explanation belies the months of planning, learning, data clean-up, and collaboration that led to the operational launch of MediaBin at the Metropolitan Museum in fall 2007.
As the staff at the Metropolitan Museum planned the centralized storage and management of its growing collection of digital images, they were also considering new opportunities for licensing images. Exploring ways to derive more income from commercial licensing led staff to examine the options offered by third-party image distributors. One successful model was the photographic agency of the Réunion des musées nationaux9 (RMN) that has an online image base of nearly 450,000 images of works of art from French regional and national museums and other European museums available for both educational and commercial licensing. Colleagues at the Victoria & Albert Museum also met with Met staff to talk about their growing image licensing initiative. After considerable deliberation and study, the Metropolitan Museum of Art decided to outsource commercial licensing, announcing an agreement with Art Resource10 in January 2007. Subsequently, additional distributors have been added: Scholars Resource,11 Scala,12 and RMN. The images and information are now exported from MediaBin and sent several times per year to the distributors.
Traditionally, museums charge less to supply an image (and the permission to reproduce it) for scholarly publication than for commercial publication or product development. The Metropolitan Museum had different rates for commercial and non-commercial licensing, and the unofficial policy was to supply fee-free images to Metropolitan curators writing for non- Metropolitan publications, to professional colleagues at other institutions, and to former Met colleagues. Museum staff wanted to formalize this practice by making fee-free images more widely available for scholarly publication.
Doralynn Pines, Associate Director for Administration at the Metropolitan Museum, describes some of the factors influencing this decision:
It is one thing to decide to provide fee-free images for scholarly publication, and quite another to commit staff time, and therefore dollars, to delivering those images. Clearly, the Met needed a partner in this venture and turned to ARTstor.13 The museum had been one of the early contributors to the ARTstor Digital Library when its AMICO records were released in 2005. It seemed natural for the museum to turn to ARTstor to build a delivery mechanism for the Metropolitan’s publication-quality images for use in scholarly publications. ARTstor readily embraced the idea and the partnership was launched. An ongoing stream of high-resolution images would thus be made available for use in the K-12 schools, colleges, universities, and museums that license ARTstor, and images that could be used in publications were made available for both users and non-users of the ARTstor Digital Library. Scholars would be well served by the ability to obtain publication-quality images, without fees, that could be downloaded immediately.
Working closely with the staff at the Metropolitan, ARTstor began to build Images for Academic Publishing (IAP) to meet the museum’s specifications.14 The project comprised the preparation of image assets and corresponding metadata, inclusion of these assets in the ARTstor Digital Library, and the development of a new protocol for user download of publication-quality images.
The Metropolitan Museum staff decided to express information about their objects using CDWA-Lite,15 an XML data schema developed as a joint effort between the J. Paul Getty Trust,16 RLG Programs/OCLC,17 and ARTstor for describing cultural works and their visual surrogates. CDWA-Lite, based on a small subset of fields from the Categories for the Description of Works of Art (CDWA), represents the minimal set of data fields deemed necessary for describing cultural works and their visual surrogates in preparation for resource discovery in online environments. CDWA- Lite is intentionally “light” to lower the barrier for cultural heritage institutions wishing to share content. The CDWA-Lite schema is designed to be used with the Open Archives Initiative Protocol for Metadata Harvesting (OAI- PMH)18 that facilitates the sharing and updating of information between the provider and the distributor. Once MediaBin was fully implemented, the Metropolitan Museum staff and ARTstor began sharing the information formatted according to CDWA-Lite and harvested in a server-to-server exchange.19 Depending on number and file size, the high-resolution images can be retrieved from an FTP server or sent by overnight mail on a high-density drive.
ARTstor’s Images for Academic Publishing was launched in March 2007 and functions as follows:
Initially, IAP was only available to scholars and curators at institutions that license ARTstor, but after several months it was extended to any scholar who contacts either ARTstor or the Image Library at the Metropolitan to obtain a password to access IAP images. As of September 2008, approximately 5,600 images had been contributed to IAP by the Metropolitan Museum. Although this process is under review, the current plan results in additional deliveries of one thousand to two thousand images every four months. The Metropolitan’s Image Library staff monitor the ARTstor-generated usage reports that contain the raw information about users and intended uses.23
During the first year of service, 645 images were downloaded from IAP for scholarly publications. Staff members note that the benefits of working with ARTstor include:
[NOTE: This section is based on a September 15, 2008, telephone interview with Ian Blatchford, Deputy Director, and email exchanges with Alan Seal, Head of Records and Collections Services, Victoria & Albert Museum.]
The task of implementing electronic recordkeeping at the Victoria & Albert Museum (V&A) for its collections, numbering nearly 4.6 million objects, represents ongoing work that began more than two decades ago. By the late 1990s, three systems were in place: the Collections Information System (CIS) for inventory control, cataloging, and accessioning museum objects; the Photo Cataloguing System for information about analog and digital photographs of objects, books, events, gallery installations, and staff; and the Image Arena, in which medium-resolution images were stored and made accessible to the other two systems. All three resources were used to extract the data and images that originally fed the collections area of the V&A website. By 2005, the museum began investigating digital asset management systems (DAMS) to replace the Photo Cataloguing system, and ultimately implemented their DAMS, called Vadar (V&A Digital Asset Repository), during 2006. A storage area network, also brought online in 2006, greatly increased storage capacity and now allows the museum to access the high- resolution master images online rather than offline from an ever-expanding collection of CD-ROMs. Since putting the masters onto their storage area network, they no longer save copies on CD-ROM, even for archival purposes. A back-up routine of the online files is in place both for recovery in the event of disk failure and for business continuity.
As the technical infrastructure was growing, staff were also considering the best ways to prepare and openly share data and images on the museum’s website and possibly on other nonprofit educational websites. They plan to support data harvesting giving aggregators a choice of Dublin Core, CDWA-Lite or PNDS24 file formats, all of which will be generated on the fly from the database outside the firewall. The V&A has already implemented a Universal Resource Indicator (URI) link in the V&A records so the distributor’s site can lead the user back to the V&A website to view the fuller catalog information. This approach will allow the museum to avoid the synchronization problems of preparing different record formats for each distributor. The open-source routines could dramatically simplify contributing both new records and edits to existing records. In addition, the V&A is on the international team of museums working in partnership with OCLC/RLG Programs on the Museum Data Exchange Project, which is testing data exchange processes using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH).25
New digital photography is being done in concert with the ongoing program of capital improvements at the museum. The renovation of the William and Judith Bollinger Jewellery Gallery, which opened in spring 2008, afforded the opportunity to fund photography of 3,500 objects. Likewise, 35,000 objects from the encyclopedic and global ceramics collection are being photographed during the current redevelopment of the Ceramics Galleries that will be opened in September 2009. This program of thorough photographic documentation is a direct by-product of the refurbishment of galleries. In each case, it anticipates the worldwide scholarly attention that is focused on the museum as collections, long off-view, are returned to the public eye in freshly interpreted, elegantly presented spaces.
In December 2006, the V&A announced that it would drop charges for the reproduction of images in scholarly books and magazines, a decision the Art Newspaper heralded as “a move which could transform art publishing?”26 V&A Deputy Director Ian Blatchford explained that, although the government encourages museums to provide this public access to collections, it is not a funded mandate. He described, instead, the internal factors leading to this policy:
The V&A decided to integrate the delivery of high-resolution images for scholarly publishing into the “Collections” area of its website. The “Collections” database includes thirty thousand works represented by more than fifty thousand images and is expected to grow by about twenty thousand images per year. In addition to simple and advanced searching options, the user finds explanations on the website of three ways to obtain images:
At the same time that the museum is foregoing licensing income from scholarly publishing, it is very actively marketing commercial use of images. These include filming at the V&A, licensing short educational films produced by the museum, photo-shoots, commissioned photography, and a customized high-quality print service.
During the first year of the fee-free image service there has been “no collapse in the finances of the Picture Library; indeed, the results of providing images without charge for scholarly publishing have been completely neutral with regard to the bottom line.” Only about twenty people have downloaded as many as two hundred images during the year, which is well within the terms and conditions of use established by the museum.
In the future, upgrades to the web application will enable the museum to track the download frequency of individual images and to study, over time, the way in which this growing body of images is used. “We care about branding; if people see our content online and realize the V&A is a great place to visit, we win.”
[NOTE: This case study is based on a document prepared by the Smithsonian Institution and sent to the author in October 2008, and a telephone conference on October 9, 2008, with Anne Van Camp, Archivist, Smithsonian Institution, and Katherine Spiess, Director, Central Digitization Office.]
In June 2008, the Smithsonian Institution27 released nine hundred public domain images in The Commons on Flickr.28 By November 2008, the number had grown to over two thousand. The images vary in file size, but the largest are suitable for publication. How and why did the Smithsonian Institution decide to become a member of Flickr, a popular and commercial photo-sharing website? Does The Commons provide a viable delivery mechanism of images for scholarly publication?
The Smithsonian Institution (SI) is a federated institution comprised of nineteen museums, nine research centers, and the National Zoo, all of which have photographic holdings in digital and analog formats. The estimated thirteen million images in the Smithsonian photographic collections have historic, artistic, and scientific import. They also document the history of photographic processes and techniques.
One strategic goal of the SI is to unite the collections virtually in a digital asset management system. A pan-institutional assessment of photographic holdings is needed to establish collection priorities and determine the resources needed to undertake a unified, large-scale digitization, cataloging, and access program. In the absence of such an assessment, the individual SI units have developed separate websites to present whatever fraction of their photographic collections have been digitized and cataloged in sufficient depth to support discovery and research.
Against this backdrop, the Smithsonian Photography Initiative (SPI) was established in 2001.29 It serves as a central, web-based programming unit designed to stimulate dialogue about the cultural impact of photography with new and existing Smithsonian audiences. Organized as a series of integrated programs, the website Click! photography changes everything30 invites the public to consider ways in which photography enables people to see, experience, and interact with the world. Although the Search Images feature of the website presents a relatively small percentage of art, science, culture, and history images available in each of the Smithsonian’s units, it does present the only pan-institutional image cross-section. SPI does not provide sales or licensing services; instead it redirects web visitors via links to the websites of the SI units, which each manage their own images and content. There is no consistent policy regarding licensing fees across the units of the Smithsonian Institution, and some work with third-party licensing agencies such as Corbis and Art Resource.
At the same time the Smithsonian Photography Initiative was launched, Smithsonian Images31 was created as a pilot web program with the goals of using credit cards for e- commerce and increasing visibility and access to its online digital images. Fees charged for non-commercial use were designated for recovery of distribution costs rather than to generate profit for the Institution. From its inception, Smithsonian Images has permitted free download of digital images at a non- publication resolution for educational, scholarly and personal use under the terms of “fair use.”
Early in 2007, the Smithsonian Institution’s Digitization Steering Committee issued a report with recommendations about the resources and infrastructure needed to create, manage, provide access to, and use the Institution’s digital assets to effectively meet the needs of real and virtual visitors. In April 2007, the Smithsonian met with Library of Congress staff to discuss technology infrastructure requirements to support digitization and various procedural considerations, including intellectual property rights. The Smithsonian Digital Media Use Committee was formed in July 2007 to create a new pan-institutional policy reflecting current technologies and SI’s commitment to providing broad access to digital assets in a manner consistent with its legal and stewardship responsibilities.
Although the Smithsonian Institution seeks to increase access for educational and research purposes, it cites the following reasons for not making images available free of charge for scholarly publishing:
In January 2008, the Library of Congress announced a new pilot project launched with Flickr that was designed to:
help address at least two major challenges: how to ensure better and better access to our collections, and how to ensure that we have the best possible information about those collections for the benefit of researchers and posterity…. 3,000 photos from two of our most popular collections are being made available on our new Flickr page, to include only images for which no copyright restrictions are known to exist…. We want people to tag, comment and make notes on the images, just like any other Flickr photo, which will benefit not only the community but also the collections themselves…. We’re also very excited that, as part of this pilot, Flickr has created a new publication model for publicly held photographic collections called The Commons. Flickr hopes, as do we, that the project will eventually capture the imagination and involvement of other public institutions, as well.32
Staff at the Smithsonian Institution were meeting with Library of Congress staff about digital issues during 2007 and early 2008. The early success of the LC/Flickr pilot helped convince Smithsonian units to launch their own Flickr project.33 They formed a core team that drew its members from the SI library, archives and museum communities, central programmatic units that support public programs, and from the Office of the Chief Information Office. In February 2008, the group issued a call to the Smithsonian community for digital images to be used in the Flickr project with a goal of providing approximately two thousand digital photographic images from a variety of collections throughout the Institution.
In deciding to move forward with participation in The Commons, the Smithsonian defined the following goals for the project:
On June 16, 2008, nine hundred images were uploaded from the Smithsonian Institution to The Commons; by October 2008, the number of SI images available in The Commons had doubled. All the images in The Commons can be viewed and downloaded at five different resolutions; the original image, contributed by the owning institution, determines the size of the largest file.34 In the case of the Smithsonian, there is no pan- institutional policy about what size the “original” image should be; each SI unit makes that determination independently. As a result, only a portion of the SI images in The Commons are, at this time, of adequate size to download for publication. Recognizing that Flickr is commercial, and therefore not a trusted website in many educational environments, SI has added the same images to the Smithsonian Photography Initiative website, which is educational, trusted, and branded with the Smithsonian Institution imprimatur. Each SI image appearing on a Flickr Commons page links back to the same image on the SPI website, Click.35
The Smithsonian Institution researches images contributed to The Commons, releasing those they believe have no known legal restrictions. This includes copyright and other legal restrictions, such as those required by the donor of the image or the underlying object. The images are flagged with the rights statement, “No known copyright restrictions,” and viewed on The Commons with two associated links. One link leads to the generic Flickr rights page that reminds users to conduct “an independent analysis of applicable law before proceeding with a particular new use.”36 The other link takes the user to the explicit rights statement for the image provided by the contributing institution. In the case of the SI images, the link takes the user to the “Copyright: Terms and Conditions” page of the Smithsonian Institution website to learn:
Anyone wishing to use any of these files or images for commercial use, publication, or any purpose other than fair use as defined by law, must request and receive prior written permission from the Smithsonian Institution. Permission for such use is granted on a case-by-case basis at the sole discretion of Smithsonian's Office of Product Development and Licensing. A usage fee may be assessed depending on the type and nature of the proposed use.37
At present, this leads to some confusion for users, as the statement suggests that some of the SI content on The Commons may be by protected by usage restrictions. SI is aware of this discrepancy and is working to develop new, more accurate language.
The Smithsonian Institution joined The Commons on Flickr to make its content more widely accessible. After the first four months, it found that Flickr definitely increased exposure of their images to more individuals. “Portraits of Scientists,” a set of images of nineteenth- and early twentieth-century scientists and inventors, has been on the Smithsonian Libraries’ website since 2003. In the first three months the set was on The Commons, the images received nearly as many visits as during the previous five years on the Smithsonian site. SI is not currently using The Commons to delivery fee-free, high-resolution images for scholarly publication, but may consider doing so in the future.
The Commons is a space for public institutions to share images of their collections. “Pro” membership in The Commons permits museums to upload an unlimited number of images as large as twenty megabytes in size. The Smithsonian Institution’s goal in providing images to The Commons is to expand access to its photographic collections, not explicitly to supply high-resolution images for scholarly publication. Other museums, however, may find that The Commons provides a cost-effective delivery mechanism of fee- free images for scholarly publication.