Exploring New Models for Administering Intellectual Property: The Museum Educational Site Licensing (MESL) Project
A paper prepared for the 33rd annual Data Processing Clinic, University of Illinois at Urbana-Champaign, March 25, 1996.
The museum and educational communities have seen the potential for digital imaging and network technologies to make cultural heritage information more broadly accessible. However, the integration of museum digital content in to higher education has been hampered by a lack of progress on the definition and administration of intellectual property rights. By their nature, imaging systems require a complex balancing of the interests of numerous rights holders in protecting their intellectual property and the desires of image users to use images in their study, teaching, and research. A common understanding of rights, permissions, and restrictions and a shared framework for administering rights reflecting broadly accepted terms and conditions for the use of materials would ease the burden of honoring intellectual property rights and further the educational use of digital materials.
The Museum Educational Site Licensing Project (MESL) brings representative U.S. museums, colleges, and universities together to explore these issues. Their goal is to define the terms and conditions for educational use of museums' digital images and information on campus-wide networks. During this two-year experiment (launched in 1995 by the Getty Art History Information Program in conjunction with MUSE Educational Media) a select group of educational and collecting institutions are collaborating in good faith to study the capture, distribution, and educational use of digital images and their associated texts.
The partners in the MESL project are developing and testing administrative, technical, and legal mechanisms that could eventually make it possible to deliver large quantities of high-quality museum images and information to all educational institutions. Participants are developing a model site licensing agreement, exploring models for the collective administration of intellectual property rights, and studying the economics of image creation and network distribution. The project has also provided a vehicle for exploring and promoting the educational benefits of digital access to museum collections through campus networks.
his paper reports on the first eighteen months of activity in the MESL project (January 1995 - June 1996) offering both some preliminary impressions of the participants' experiences to date, and an assessment of the issues the project faces in its second year of activity.
Uncertainty regarding intellectual property rights has been a barrier in the creation of networked information resources for some time. This problem is exacerbated when visual resources are concerned for a number of reasons. First, the rights in digital images are often multi-layered and complex. Simply determining who holds the rights in a particular work is often difficult. This problem is exacerbated when the digitization of existing visual resource collections is contemplated, because these collections have been constructed over time, and often lack detailed information about the sources of images. Secondly, existing rights administration systems are inefficient at best. We are without a comprehensive service that offers rights to museum images. As a result, a disproportionate amount of time and effort is expended in the information location and rights-negotiation process. Thirdly, the legal framework has yet to respond to the changes in technology. A solution to these conundrums is unlikely to come from the legislative arena, as the law is by nature responsive and conservative; a consensus on these issues, which is satisfactory to law-makers and lobby-groups alike, is unlikely to emerge in the short term.
Basic picture research has always been one of the primary research challenges in disciplines that depend on the visual as a primary information source. Simply locating works of art, that may have passed from private collection to private collection, or be held, but uncatalogued in a public collection is a specialist task, requiring much ingenuity, and not a small amount of serendipity or sometimes, blind luck. Much specialist knowledge is required to negotiate the vast number of information sources to identify the particular images that are relevant to a specific line of inquiry. A large proportion of research is based on the construction of the corpus of a particular creator - the catalogue raisonnée. There is still no union catalogue, or finding aid which indexes available images. Much picture research involves separately contacting numerous institutional collections, and requires much expert knowledge to identify appropriate visual resources. Many sources go unnoticed, and a disproportionate amount of time and effort is expended in this information location stage.
Once an image has been located, gaining the rights to use it in a publication, or multi-media project is an equally complex task. Much of the confusion regarding intellectual property rights in visual images arises from the many ways that they are created, and the many sources for images in educational institutions' collections. Digitizing and using digital images for educational purposes requires an analysis of all the rights connected with each image. Determining the rights inherent in an image requires an understanding of the source of the image, the content portrayed, and the nature of the image (whether it is an original visual image or a reproduction).
Visual images can be original works themselves, they can be reproductions of other copyright works, or if a reproduction includes original elements, they can be both. Often, a digital image is many 'generations' removed from the original work that it reproduces. For example, a digital image may have been scanned from a slide, that was copied from a published book, that printed a photographic transparency, that reproduced an original work of art. Each stage of reproduction in this chain may involve an additional layer of rights. The rights in each of these images may be held by different rightsholders; obtaining rights to one does not automatically grant rights to use another. Existing Visual Resource Collections are comprised of many types of images, each with particular rights, or layers of rights associated. Digitizing such a collection requires a commitment to the identification of rightsholders and the negotiation of rights to convert an image into digital form.
The chart that follows offers a schematized (and simplified) representation of the sources for digital images.
An Original Visual Work can be defined as a work of art or an original work of authorship (or part of a work) fixed in a visual medium. Original Visual Images may be in digital or analog form. Examples of original visual images include graphic, photographic, sculptural and architectural works, as well as stills from motion pictures or other audio-visual works. The rights in an Original Visual Image are defined in Section 106 of the US Copyright Act, as the right to reproduce the work, to prepare derivative works based on it, to distribute copies of the work, to perform the work, and to display it in public.
A Reproduction can be defined as a copy, in digital or analog form, of an Original Visual Image. The most common forms of reproductions are photographic, including for example, prints, 35 mm slides and color transparencies. If a reproduction is legally made (i.e. with the permission of the rightsholder in the original work) and includes copyrightable elements, it can be eligible for its own copyright protection, which must be considered in addition to the rights inherent in the Original Visual Image. The Original Visual Image shown in a Reproduction is often referred to as the 'underlying work'.
Many digital images reproduce other works. Digital images can be reproductions of either Original Visual Images, for example, when an original work is scanned directly, or of other Reproductions, for example when a scan is made from a transparency reproducing a work of art.
A Published Reproduction, is a Reproduction of an Original Visual Image, appearing in a publication. Examples of Published Reproductions include a plate in an exhibition catalogue that reproduces a work of art, or a Digital Image appearing in on a CD-ROM. Separate copyrights may exist in the Publication, the Reproduction and the Original Visual Image.
In some cases, such as copystand photography, a Published Reproduction may have been further reproduced, creating a Copy of a Published Reproduction. As these types of copies are often mechanical in nature, they may not be copyrightable in themselves. However, rights in the Original Visual Image, the Reproduction and the Publication must still be considered.
A digital image is a single, still image stored in binary code - bits and bytes. Examples of digital images include bit-mapped images (encoded as a series of bits and bytes each representing a particular pixel, or part of the image) and vector graphics (encoded as equations and/or algorithms representing lines and curves). A digital image can be an Original Visual Image, a Reproduction, a Published Reproduction or a Copy of a Published Reproduction; determining what the way a digital image was created will determine the rights associated with it.
A thumbnail image is a small scale reproduction of a digital image, often used in an on-line catalogue or image browsing display to enable identification of an Original Visual Image. Thumbnail images are of low resolution and quality (often averaging between 100 x 100 to 256 x 256 pixels) and are considered to be of limited commercial or reproductive value. While there are still rights associated with thumbnail images, they are often distributed more freely than higher quality images, as a visual reference to the original work.
Visual resource collections in educational institutions often number tens of thousands of images, generally photographic slides, which may be Original Visual Images, Reproductions, Published Reproductions or Copies of Published Reproductions. The images in visual resource collections have been acquired from a wide variety of sources, by purchase, donation, or through copy-photography, or original photography. Collections have been built over an extended period of time, and it is often impossible to trace the sources of images acquired by purchase in the past, or to identify if a work is indeed still available in order to negotiate rights. This complexity makes the conversion of existing visual resource collections into digital form problematic.
Even if it is possible to identify who holds the intellectual property rights to an image, locating that rightsholder may be a very difficult task, and negotiating the rights an arduous process. Within museums, rights administration procedures are now based on a print model of publication and distribution, and are focused on the single image. Each image often requires a separate request, with its own forms and permissions to negotiate. Museums are without a single fee scale, and the fees that are charged are also based on the print model. What may have seemed reasonable for a high quality art book, containing at the most 50 images, seems unreasonable for a multi-media publication containing 10 times that many. In addition, each museum has defined its own terms and conditions under which an image can be used. As a result, a content user has to negotiate (and renegotiate) with many separate institutions in order to build up an archive of usable content. This redundancy adds a level of overhead to the rights acquisition process which impedes the use of large numbers of images, and may serve as a deterrent to the negotiation of rights to use images.
These restrictive forces are in clear opposition to an increased demand for multi-media content in the educational community. Experiments in using new technologies are maturing to become news tools for providing increased access to research resources. Where in the past, for example, undergraduate art history students were unable to use institutional slide libraries for review after class, collections of digital images made available on-line enable consultation both on and off-site, at a time convenient to the student.
The creation of digital image collections in a systematic and uniform way offers a real benefit to the educational and research community, enabling the creation of teaching resource to support the curriculum, and providing a source of quality content to support the integration of new technology into teaching and research. However, the digital conversion of existing slide collections is not necessarily an ideal solution. Slides may be of uncertain age, or many generations removed from the original work, and therefore of limited quality. Rights in the original visual images and their reproductions may be uncertain, and use restricted. In addition, documentation may be incomplete. The ability to acquire quality digital images from a reliable source, accompanied by authoritative textual descriptions would be of significant value to the educational community.
The Museum Educational Site Licensing (MESL) Project was established to respond to the need for educational access to high-quality, rights-cleared museum images and accompanying texts. The project brings museums (as information providers) and universities (as information users) together to define the terms and conditions for the educational use of digital resources drawn from museum collections. It is exploring an alternative method for distributing digital content drawn from museum collections to the educational market.
Museums are in a position to offer the educational community a quality information package of text and image - in effect a multi-media description and analysis of the works of art in their collections. Under appropriate licensing terms, it would be possible to make this depth of knowledge about museum collections available for research and teaching. What is required is a contractual arrangement under which museums could supply content to educational institutions at predictable terms, and for a reasonable cost.
Bringing the information providers and the information users together to prototype a licensing agreement also offers a means to address the uncertainty of the legal framework. Rather than having to rely upon the courts to define the application of old laws to new technology, both parties can negotiate a mutually beneficial licensing agreement. The terms and conditions for this license are being established through experimentation, entered into in a good-faith spirit of cooperation, and that will result in a contractual agreement that meets the needs of both sides.
Museums and universities have been ideal partners in this experiment. In addition to the obvious attraction - that museums have content that universities want - there are other factors that have contributed to the success of the MESL experiment. Both share a common culture of teaching and learning. This common focus has enabled the definition of licensing terms that enable a full range of educational uses. In addition, museums and universities are both information users and information providers. Museums often engage in research that requires the consultation and use of images in other collections; universities hold collections of unique materials in their libraries' Special Collections, and in campus museums. This duality has enabled the negotiation process, as participants have been able to see issues from both sides.
The participants in the MESL project were selected in a competitive call for participation issued in the fall of 1994. Fourteen participating institutions, seven universities and seven museums or collecting institutions, were chosen to represent a broad range of sizes and governance structures. Technological experience was also highly ranked, as it was seen as an essential pre-condition for full project participation. Each institution has fielded an interdisciplinary project team: museum teams include members from the curatorial, registration, photo services, and administration departments, along with the museum library and research centers; university teams include faculty, instructional technology, library, campus computing and administration representatives. The project is managed by the Getty Art History Information Program, and advised by a Management Committee.
Between January of 1995, and June of 1997, MESL participants will focus their attention on defining the terms and conditions for the educational use of museum digital content, and exploring appropriate technical and administrative mechanisms for enabling the distribution of high quality information. This will require balancing of the requirements of rights holders and rights users, and addressing a number of technological and pedagogical challenges.
As information providers and rights holders, museums' paramount concern is maintaining the integrity of the original works of art, that they preserve and interpret in trust. A distribution system must, therefore, ensure the accuracy of the information distributed, and provide adequate protection from alteration or unauthorized copying. It must also acknowledge both the artist and the collection, and offer some sort of remuneration for the intellectual property created by the museum.
As information users, universities require easy access to a large body of high-quality material, from a central or coordinated source. Materials should be predictably high quality, and available under a reasonable fee structure, according to common terms and conditions, regardless of source. High administrative overheads and processing costs should be avoided.
An effective and efficient system of administering intellectual property rights is key to the development of educational use of museum digital content. Establishing such a system depends on the definition of a set of standard terms and conditions for the use of quantities of material, the development of equitable pricing models, and the creation of a framework within which it is possible to negotiate rights efficiently.
Traditionally rights are assigned by holding individuals or institutions for the specific use of a particular work. This kind of licensing, focused on the individual item, is difficult to adapt to the in-depth research requirements of higher-education, where access to a broad range of material is essential and depth of content may be as critical as access to a particular work. MESL is exploring a model whereby museums offer collections of material to universities under a single site-license. A range of digital images and information from museum collections is made available under the same terms for use by all members of a campus community.
Many pricing structures and systems being developed to support digital commerce are premised on a 'pay-per-bit' or 'pay-per-view model'. This type of transaction-based pricing did not fit well with the educational goals of the MESL project. It was feared that per-use charges would inhibit access and discourage the exploration of a new kind of information. In addition, participants did not have the monitoring systems in place that would be required to gather individual usage statistics. Finally, as we were introducing a new resource to the campus community it seemed impossible to predict usage levels, and therefore derive realistic pricing models. The pricing model that MESL is exploring is based on a subscription - a predictable fee paid for unlimited use of a defined information set.
There are a number of models for the administration of license terms and fees, each with its own pros and cons. These can be characterized as:
- a rights holder's collective
A collective body acts for rights holders, and representing their interests to copyright users. Often a standard set of terms and conditions with a single fee scale is applied regardless of the information supplier or user. Examples of this type of organization include the Copyright Clearance Center (CCC), that acts for publishers, and the American Society of Media Photographers (ASMP), that represents photographers.
- a brokerage
A third party administers rights for a fee, which is often charged as a percentage of the license fee negotiated for the use of intellectual property. The terms of each transaction may vary. Examples include Picture Network International (PNI) or the Kodak Picture Exchange (KPX).
- a rights reseller
A third party can acquire rights and then resells them with or without consulting the original rights holder. Examples include Corbis Media and stock photo agencies.
- a consortium
A membership organization, such as the Research Libraries group (RLG) can agree to exchange information for mutual benefit.
- a locator service
Not truly a rights administering body, a locator service acts as a finding agency, which passes requests through to rights holders, which define the terms and conditions of use, and negotiate licenses individually. An example includes Academic Press's planned Image Directory (ID) service.
Each of these administrative models has pros and cons, often trading simplicity in administrative structure for flexibility in licensing terms. MESL participants are examining these models to see which would best serve the needs of museums and universities, and best satisfy the requirements for integrated information location and intellectual property rights acquisition services.
The participants in MESL adopted an experimental methodology to explore the issues of licensing museum digital intellectual property for educational purposes. Over the course of the two-year project, the participating collecting institutions agreed to make a significant number of images from their collections available for educational use on the campus networks of the participating educational institutions. This allowed the project participants to gain real experience with the technical issues associated with the digital distribution of museum information, and to develop a framework of use within which to define and test the terms of a model site-license agreement.
As a basis for their collaboration, all participating MESL institutions signed a Cooperative Agreement. This document outlined the goals of the experiment, and defined the responsibilities of each participant. It also outlined the terms and conditions for the use of museum information on campus-networks - the first draft of the terms and conditions of a site license.
MESL institutions desired to fashion terms that would enable the broadest possible use of museum digital intellectual property within the educational context, but which protected the investment museums have made in its creation. MESL information may be distributed over the campus network for educational use only, including research, teaching and student projects. Any commercial use, or redistribution beyond the bounds of the campus, is not permitted.
In the next year, the Cooperative Agreement will be rewritten as a set of model site-license terms. These will address the use of museum information for educational purposes, but will not define the legal framework for a licensing body, or the technical framework for information collection and distribution.