I'm pleased to post this on behalf of James Shulman.
Max Marmor, ARTstor
*****
Dear colleagues,
We are frequently asked about ARTstor's policy concerning "interoperability." Why did ARTstor develop its own software environment and tools, when a variety of image management and delivery packages already exist and when some are widely used among ARTstor's main audiences? Why does ARTstor limit the user's ability to download or export large-format image files into other software? Why does ARTstor not allow participating institutions to host ARTstor content locally in their preferred software? All of these are perfectly reasonable questions. We hope and trust our answers to these questions - which are available in the Frequently Asked Questions section of our website, but also discussed when we participate in conferences like VRA, ALA, ARLIS, DLF, CNI, NMC, and others - are also reasonable. I also think it's helpful to see these issues against the broader background of ARTstor's mission and the Mellon Foundation's intentions in creating ARTstor. And so I wanted to take a moment to describe this larger context briefly.
In creating ARTstor, The Andrew W. Mellon Foundation had three fundamental objectives:
First, the Mellon Foundation sought from the beginning to create a digital library that would include user-friendly tools supporting use of the content, and that would be helpful to a wide audience with varying needs within a reasonably short period of time. In assessing the needs of educational users, we found that there were some individuals and institutions in the community who had been the pioneers of the digital frontier for one, two, five or more years, who had learned lessons, gained experience, established practices and norms, and who were already in the process of developing their own image delivery systems or using third-party software. Some of these systems were home-grown, some were made available on an open-source basis, some were licensed by commercial vendors, and many of them did not - and do not - interoperate with each other. At the same time, we found a much vaster number of places who had no digital image delivery systems, and who indeed had little or no access to digital images at all. This range of users is reflected in the different size and types of educational institutions which are now among the approximately 200 institutions participating in ARTstor; these include community colleges (which frequently do not have access to a significant amount of content, let alone image delivery tools), and large research universities (some of which were developing their own systems and some of which were relying on third party sources) and all sorts of institutions in between.
To meet the pressing needs of many of our users, the Mellon Foundation believed - as ARTstor believes now - that the first step was to obtain - either by license or through our own development efforts - an image delivery software system. Over the next few years, the Mellon Foundation hoped to collaborate with a commercial firm that had built software. Eventually, however, we concluded for a range of reasons that it was necessary to retain a team of programmers who were charged to develop a software system as quickly as possible that was easy to use, had a broad range of functionality, and could be used by any educational institution that chose to do so.
The second objective of the Mellon Foundation was to get a significant amount of content to users quickly. A number of institutions had approached the Mellon Foundation for support to digitize their core teaching collections. They had varying reasons for doing so (such as deteriorating slides, concerns about discontinuation of slide projectors, demands of faculty and students, and an interest in making digital images available for student review). Rather than support redundant efforts at a number of institutions that would perhaps not benefit the community at large in a cost-effective, scalable way, the Foundation devoted very significant funds to build a resource that would meet those common demands of users for access to content. This not only had an impact on choices we made about software systems, but also has been reflected in our intellectual property policies: to meet the core and pressing teaching needs of a wide range of educational institutions, ARTstor in some instances has relied on the fair use doctrine to make this content available, rather than seeking licenses - and this is noted in ARTstor's intellectual property statement [http://www.artstor.org/info/about/statement.jsp]. Implicit in this decision was the recognition that if we had to exclusively seek licensed collections, we would not be able to meet the basic teaching needs of our users within any reasonable time period. Moreover, to allay any concerns of our users and to encourage broad uptake of the resource within the educational community, ARTstor also agreed to indemnify our users for any copyright liability stemming from the permitted educational and scholarly uses of those images.
The third objective of the Mellon Foundation was to create a sustainable resource. This objective was critical, and not only because the Mellon Foundation had made very significant investments in ARTstor rather than devoting those monies to other worthy causes. It was critical because the first two objectives - providing a solution that satisfied the broad needs of many of our users and that helped our users avoid the costs of having to digitize their own materials - could only be achieved if we succeeded in this third objective. In other words, if ARTstor provided content that was later withdrawn en masse, as has happened with other digital resources, or developed a solution to many users needs that eventually could not be sustained, or provided such a small amount of content that the vast majority of our users' basic needs were not met, then the Mellon Foundation would not have met any of its aims in undertaking this very sizeable effort.
At the heart of this last objective has been the belief that if we did not work with copyright owners and content providers, we would not develop a sustainable resource. This is true for a couple of reasons: First, the costs of digitizing content - particularly at high resolution - and of obtaining and updating cataloging for such content are very significant. While the Mellon Foundation has been willing to provide very significant funding (separate from the funding for the software) to create digital collections, we have recognized that it will be very important for the long-term success of ARTstor for museums and other content owners to embrace and, where possible, to contribute to ARTstor. Second, given the legal uncertainties surrounding fair use and other exceptions to copyright, ARTstor has believed that - especially as we make the library available abroad - we will only be able to manage risks responsibly and continue to make content available to users on an ongoing basis if we reach out to, and collaborate with, those who own and care for art objects and sites, as well as those who create or document art.
Like users in the educational community that ARTstor seeks to serve, content providers have their own legitimate needs and concerns. Some seek to supplement their essential revenues through commercial use of their digital images and are concerned about the potential impacts on such revenues should they lose control of their images, particularly if such images were to be made available through the internet. Others are worried about copyright issues - even in those cases in which institutions or individuals do not own the copyright to the underlying works in their collections, some are concerned that by sharing the images they could be perceived as not respecting third parties' copyrights. Still others have issues with the quality of the images of their works and how they will appear if they are displayed in digital media. These are just a few of the many concerns we have heard, and it's important to note that we have heard these concerns even from educational institutions providing content to us.
ARTstor has, by in large, been fortunate in our negotiations with content providers to date. We are now collaborating with many different contributors, including universities and schools, museums, research institutions, libraries, publishers, individual scholars, and others worldwide. We are pleased that, to date, such institutions have been very willing to make available images of their works to ARTstor and its users at no cost other than the cost of some assistance - when necessary - with digitizing or cataloging the materials in question. Because of these collaborative efforts, and the generosity of our "archive contributors," we expect to have a searchable database of about half a million images and related cataloging within the next couple of years.
While the negotiations for content have varied depending on the content provider, they have generally not been easy. In large part, we believe that this is due to the complexity and relative novelty of what ARTstor is seeking to do, including the scale of ARTstor and its objectives and the uncertainty that still exists regarding digital technologies and intellectual property rights. In many instances, our negotiations with content providers have literally taken years to complete, despite the pressure that ARTstor staff have felt to conclude the negotiations so as to meet the needs of our users. In almost all instances, these negotiations have required extended and complicated discussions. In others, they have required meetings with multiple departments or "layers" of an institution to obtain the necessary permissions. Even where the content providers did not own the rights to the content, these negotiations have often been very protracted because of the intellectual property concerns and uncertainties surrounding the uses of digital images.
A critical component - and perhaps the critical component - of our ultimate success in obtaining this content has been ARTstor's promise that it would take substantive steps intended to "safeguard" the interests of content owners. This has included making the images available through an access-restricted database, with terms and conditions of use, governing the uses of images. It has meant making promises that in the event of inappropriate uses we would work with our educational users to stop such violations of our agreements. It has meant agreeing that we would provide some monitoring of usage patterns, while of course respecting the privacy of our users. And it has also involved imposing limits on the ability of users to export images from that restricted database.
Given our fundamental objectives - to make a sustainable resource that tries to meet the needs of the majority of educational users and provides significant content to teachers, students, and scholars within a reasonably short timeframe so as to meet those pressing needs - we have built a protected environment. We do hope, over time, that content owners who hold underlying copyrights to content will become more comfortable with digital technologies and with purely non-commercial educational use of the images. We also hope that as content providers' concerns regarding a resource like ARTstor subside and the uncertainty surrounding copyright issues dissipates somewhat - both here and abroad - we will be able to relax these early restrictions. In the meantime, however, we believe that the approach we have taken is the only one that will achieve the objectives of ARTstor and the Mellon Foundation, as well as the needs of most of our users. And we believe it is the only approach that will help us and those we aim to serve achieve long-term solutions to common needs.
In the meantime, we continue to explore ways in which we can share content with users who have developed their own image delivery systems and who want to continue to use those systems, while simultaneously respecting the needs and interests of our content providers. We will continue to develop tools that will allow users to mix their own content and ARTstor content, even if that has to be, at least initially, within the ARTstor system. And we will continue to explore ways to partner with a range of institutions, even while recognizing that ARTstor could never have all content, nor be the solution for every user's needs. We are also working on ways to enable metasearching; in fact, Bill Ying (ARTstor's CTO) and I presented on this topic recently at the Digital Library Federation Fall Forum and will be doing so again in December at the Coalition for Networked Information annual conference. We welcome wide participation in this discussion and, of course, we are active participants - and have valued so much - the counsel of our friends in the VRA community.
We hope that this provides you with a better understanding of the reasons we have made choices in developing our software, in approaching "interoperability," and in restricting the export of high-resolution images from ARTstor. As we have noted elsewhere, we welcome your comments - and criticisms - as we continue to build this resource.
With best regards,
James Shulman
Executive Director, ARTstor
__________________________________________________________________
Mail submissions to [log in to unmask]
For information about joining ARLIS/NA see:
http://www.arlisna.org//membership.html
Send administrative matters (file requests, subscription requests, etc)
to [log in to unmask]
ARLIS-L Archives and subscription maintenance:
http://lsv.uky.edu/archives/arlis-l.html
Questions may be addressed to list owner (Kerri Scannell) at: [log in to unmask]
