New Economic Models for Administering Cultural Intellectual Property

Section II

by David Bearman , Editor, Archives and Museum Informatics, Pittsburgh, USA

---

A paper presented at the Digital Knowledge Conference, Toronto, Ontario, February 7, 1996. Also presented at EVA Florence, Italy, February 9, 1996 with J. Trant.

---

Contents

Section I
• Introduction
• The Current Situation
  • A. Emerging Digital Services and Products
  • B. Dysfunctions of the existing system for licensing cultural intellectual property rights
  • C. Requirements for a New System of Rights Management
  Section II
• Analysis of the Economics of a New System, as reflected in MESL
• Issues in the design of a future system for cultural intellectual property management

Section III
• Models for Future Cultural Intellectual Property Management Systems
• Unknowns
• Conclusions
• Appendix A. Proposed study
• Appendix B. Interview Framework
• Footnotes

II. Analysis of the Economics of a New System, as reflected in MESL

In the spring of 1994, at a meeting of museum professional in New York City, the idea of a cooperative to manage museum licensing of digital rights to educational institutions was advanced. By the fall, a prototype with a three year life was funded by the Getty Art History Information Program and cultural content holding institutions (museums, archives and libraries) and universities were invited to submit applications to participate in what was called the Museum Educational Site Licensing Project (MESL). The organizing meeting of MESL was held in early 1995 and by the fall term of 1995, the seven universities involved had all received and made available to faculty and students, the initial distribution of museum and library data.4 As it enters its second year of operation, the MESL prototype of museum to higher education site licensing project is a model; in which many, although not all, aspects of the economics and technical requirements of a new system of distribution of cultural intellectual property can be tested.

Before examining the issues and discussing the light which MESL shines on the possible organization of a future system, we should acknowledge the respects in which MESL does not serve well as a model of the future. MESL institutions, are not typical of museums or universities in the United States as a whole. They were consciously selected for the degree to which they were prepared to participate in the experiment and therefore represent larger, better endowed and more technologically sophisticated institutions. In addition, the participating museum institutions are cultural history and art museums, and include no science centers or natural history museums. The educational institutions are all major universities and include no four-year colleges, two year colleges, or institutes of advanced education to say nothing of school districts or independent K-12 schools.

Even though MESL institutions are not typical in many respects, a study of the costs and benefits to these institutions of participation in a site licensing scheme can nonetheless provide useful indicators of what the costs and benefits would be across a broader range of institutions. Fortunately for those thinking about evolving such a future system, the Museum Educational Site Licensing project needs to collect much of the data required by these measures in order to account for itself. It is hoped that with that data we can make a case for the concept of site licensing museum information for education purposes, and possibly demonstrate its benefits for commercial licensing schemes as well. Baseline cost measures which need to be collected for content providers, in the MESL case museums and the one library (the Library of Congress), but potentially for archives, libraries, historic and archaeological sites, research institutes, parks and for collections within universities being made available to other educational institutions, include:

Degree of documentation of the collections, within universities being made available to other educational instituions as measured by the percentage of objects having various categories of documentation; of photography; and of digital images.

• Suitability of documentary products for digital licensing as measured by the percentage of objects having in depth documentation and/or interpretation; photography with color bars and scale indicators; high resolution digital images; and clear rights
• Content investment, as measured by the estimated cost, per object, of documentation; photography (ave. number of images per object); and digital images (ave. number of digital images per photograph).
• Infrastructure investments, as measured by the annualized cost of collections management systems (incl. personnel), the annualized cost of network access (incl. personnel).
• Digitization commitment as estimated from expenditures of the latest three fiscal years
• Rights and reproductions management costs, as measured by the number of requests for rights and permissions from educational institutions, the average processing time for such requests, and the income (if any) from such requests.

In addition, the time and money spent in acquiring rights from other museums, libraries and cultural institutions for museum publications and exhibits. Order fulfillment costs including the costs of photography and documentation for items requested for reproduction. These costs might be presumed to be the same for analog and digital supply but this ought to be examined especially when counting in the marginal costs to the provider of infrastructure investments required solely for delivery of digital images.

These cost measures for content providers should be offset against the benefits to content providers of participating in educational site licensing. Measuring these benefits requires gauging attitudes within the cultural institutions and programs, particularly the value attributed to increased accessibility of the information to local and remote visitors and researchers and the potential this creates for attracting new categories of users or providing existing users with new ways to exploit visual resource collections. Such baseline use measures collected by content providers could include:

• Increases in public use, as measured by the growth in electronic visits per annum (and possibly by increases in on-site visiting that follows from electronic contact)
• Increases in staff uses, as measured by number of museum staff accessing the data and number of uses, and reduced costs to museum staff of internal access to their own image and data resources
• Content availability, as measured by the number of objects w/images available to electronic visitors, available via fixed media electronic publications and available over in-house network to staff
• Content preparation, as measured by number of objects in MESL distribution published electronically by museum and number of objects in MESL distribution for which electronic rights were granted to non-MESL uses in latest year
• Content improvement, as measured by number of feedbacks on MESL objects fromeducational institutions. Income, as measured in reduced agency fees for photographic agency management and payments received foruse of digital images and data in educational institutions, and elsewhere if the service becomes commercial.

The other side of the equation is equally important and can also be studied within the context of the MESL project. Baseline costs and benefits measures for content using organizations, when compared to costs and benefits of a new system, will ultimately determine the value of the site license to the educational institutions and the amount they would be willing to pay for participating in a broader site licensing scheme.

Benefits measures would include:

• Market "Penetration", as measure by the % of courses using images, digital imagery from all sources, and MESL images; the % of students using images, digital imagery from all sources, and MESL images; and the % of faculty using images, digital imagery from all sources and MESL images.
• Use, as measured by the number of access to museum object data (p.a.) and the traffic in networked images downloaded from MESL
• Information Resource Enhancement as measured by the number of items in the "library" catalogue, including the number of digital representations of cultural objects (artefacts, specimens, slides) available on campus and the percentage increase in such documentation as a consequence of participation in MESL
• Opportunity for Distance Education, as measured by the use of digital resources from the homes of faculty and students, and the actual number of students enrolled in distance education programs using these materials.
• Impacts on their own cultural collections, as measured by the percentage of collection objects of campus made available digitally within the campus, and subsequently through site licensing to others.

Of course, these benefits have ancillary costs associated with them, which could need to be measured, including:

• Increased end-user system requirements, as measured in the costs of classroom projection, student workstations, and laboratory and library viewing or printing facilities
• Increased network requirements as measured by costs of increasing bandwidth, storage and speed within campus and the costs of supporting access beyond the campus to and from MESL
• Increased application support requirements as measured in costs of creating or maintaining methods for access to digital libraries of object documentation

In addition, the content-using institutions have established costs which might be increased or decreased by virtue of participation in a site licensing scheme. Some of these costs include:

• Rights acquisition costs, such as would be measured by current levels of requests for rights and permissions for educational uses of images, the present cost of processing requests for permission and their current average fee payment for such uses.
• Content acquisition costs other than rights, such as time and effort expended to acquire visual information by the University,
• The cost in time and money for potential users to acquire slides they require, including the cost of research by users to locate visual information
• Software licensing costs, such as site licenses for software tools with which to search, manipulate or present MESL content, and site licenses for similar or related ddatabases which are used in conjunction with MESL data in order to create the augmented information resource
• Estimated marginal cost of adding new objects to MESL distribution
• Income produced through access to MESL data services based on it, or software licensing to other using content providers facilities.

There are aspects of a full licensing system that could produce significant costs savings to educational institutions depending on how (or when) they switch over from a hybrid environment, which needs to support both present and future methods of serving image data. Areas of potential cost savings here include:

• Cost of Maintaining Slide Collections, such as the space, equipment and staff costs of current visual information management services
• Costs of Educational Use such as classroom equipment or data conversion costs incurred in providing copies of images for laboratory/reference use
• Cost of Infrastructure investments in delivery/presentation
• Cost of Liabilities, such as risk management, legal protection or intellectual property violation insurance provided by the educational institution

There are, of course, known costs of supporting the new environment as well, although again these might differ significantly from institution to institution based on different implementation strategies and architectures. Among those generally identified are:

• Cost of mounting or downloading site licensed data and the costs of serving that data on campus networks, including the estimated marginal cost of adding new MESL objects
• Costs of teaching with digital resources, mostly involving the costs of training, and supporting, faculty using the resources in their preparation and presentation of instruction. Separately one should aslo measure the estimated marginal cost of an additional course offered by an instructor with prior experience teaching using these resources
• Costs, presumably lower than analog costs, of capturing and transforming MESL data for campus distribution

III. Issues in the design of a future system for cultural intellectual property management

The MESL project has revealed a number of systemic problems that will need to be overcome on the way to an integrated, end-to-end, user responsive licensing service to educational institutions or open to mass or niche commercial markets.

The most critical of these problems has been well known to museum and archives specialists for years and is only exacerbated by the digital environment: there is a tremendous lack of uniformity of existing source data, both text and image. The absence of standards in the documentation of collections when compounded by the general chaos in imaging standards creates a situation in which it is extremely difficult to compile a usable local resource of documentation from diverse sources.

The second problem has to do with the current lack of a 'critical mass'. In the absence of an adequate amount of available data from which to teaching any given topic, educators need to have a 'pull' mechanism by which they can make specific requests to content providers in order to acquire data not yet digitized within the context of their uniform subscription. The complexity of maintaining such a mechanism for a large number of content consumers and providers is daunting, sop MESL has adopted a 'collegial', person-to-person, model.

A third major problem in the MESL project arises from the fact that its planners made a conscious implementation decision intended to produce redundancy in order to promote a variety of competing solutions to the need for a "directory" server. Currently each educational institution maintains a full collection of all licensed data on its campus and for mechanisms to deliver data to campus systems. This will allow MESL to examine costs and functionality of a variety of different implementations and select the best for a future system implementation.

Finally, the current implementation provides little place for value-added collections or interpretive packages, built on documentation of the individual artifacts in the database. In the first year of the MESL project it has become evident that such value-added products would be very desirable to some museum educators and directors, and may be critical for attracting elementary and secondary educators, but the cost analysis of the requirements for introducing these has not yet been conducted.

The lack of uniformity of current sources means that either providers or users will need to assume certain costs. If providers are to bear the costs of uniformity, each providing institution will need to assume the costs of uniform representation in new data capture, the cost of data reformulation, or the costs of constructing software to output its existing information in a uniform, fashion. If users are to bear the costs of uniformity, they will individually or collectively need to develop software that exploit a variety of different text and image formats. Experiments just being considered within MESL will explore how well the Document Type Definitions (DTDØs) developed by the Consortium for Interchange of Museum Information (CIMI) can serve for MESL data interchange. There is a general recognition that greater conformance with data standards adopted by CIMI, by the Museum Documentation Association of the UK (MDA) and by the Committee on Documentation of the International Council of Museums (CIDOC) would be desirable, but the reality, especially for a system which desires to add hundreds, perhaps thousands, of new content providers over the next decade is that there will continue to be a great lack of uniformity or unification of data to agreed standards will be a barrier to participation on the part of many potential contributors. To date there has been little analysis of the costs of achieving the degree of uniformity we have achieved, because the only data distribution to date was based on a newly agreed set of rules; meaningful cost data will come with experience in repeatedly providing data under fixed rules, however much those rules require data reformulation or recapture.

One troublesome element of the initial phase of implementing a scheme for educational use of museum contents is how to ensure that the contents of the site licensed library is useful or adequate for the educational purposes of the users. When the library is relatively small, as it is in the MESL prototype, potential users may be prevented from actually using the resource because it lacks images and data they would require. In theory, this ceases to be a problem when the site licensing agreement achieves participation from most providers, but in fact even then most of their potentially available content will not be digitized for many years. What constitutes a Âcritical massØ is as yet unknown, and we have yet to devise a mechanism for ensuring that the critical mass which becomes available is best suited for education purposes. MESL ended up accepting individual user requests to individual content providing institutions, but it is difficult to imagine this model working with hundreds of participating institutions.

An expensive aspect of the MESL experiment was the decision to encourage each content using institution to establish its own complete system for access to the content library on its campus. The purpose was to gain experience both in the costs and complexities of this undertaking, but it was recognized early that most content users would not be able, or want, to maintain a complete library with all the search software, especially once the licensed body of material grew to include hundreds of thousands of objects. The experience of MESL, which collected all data from each contributor in one location and redistributed it to each subscriber was that such a redundant copy method of distribution is very expensive. Costs would have included media and mailing charges, mounting time and machine reading time on the receiving end, even if the process had been smooth. tape preparation/distribution.

Before a full-fledged system for site licensing can be implemented, a cost analysis of various options for providing a directory server and on demand downloading capability must be assessed. The relative costs of site licensing under such a scheme must be keep competitive with the costs of acquiring the digital resource from a commercial supplier, such as the Academic Press Image Directory, and paying a per item fee. Potentially the value-added by in-depth textual information might make the site licensing option more desirable too. Alternatively, museums and archives will have to resist making their content available to educational institutions except through site licensing in order to create the demand for site licenses.

Even with central directory services, the expansion of the site licensing approach to smaller colleges and to secondary and elementary schools will mean that many potential users would prefer access to pre-edited collections of resources, curricular packages or interpretive guides, rather than to individual images without any added-value. It seems unlikely that such value-added resources can be created simply from pre-existing textual resources, such as museum education packets, exhibition catalogs, catalogues raisonee, etc., but certainly some types of integrative, multi-item, interpretive packages do already exist which could be provided. Cost analyses will need to be conducted to determine if providing such information requires additional investment by the network or if it could be included as a site licensing feature giving the option of entering into such licenses more value to the content users. The benefits of such packages in terms of off-setting the costs of training every faculty member to make their own teaching packages are enormous and museum educators are always seeking larger audiences for their now somewhat limited use publications.

It would be difficult to determine which benefits of an end-to-end delivery and rights clearance system are most desirable to users without interviewing different participants in the overall licensing system to determine what aspects of the scheme are potentially most attractive to them. For example, administrators at museums, libraries and archives (supplier institutions) have shown less interest in the potential to tap new sources of rights licensing income as by the opportunity to extend educational activities beyond the groups reached by the education department and the establishment of a mechanism to direct resources brought in by licensing to make better visual and textual documentation of the collection. Administrators of educational institutions (the consumers) seem most excited by the opportunity to distinguish their institutions by the forward looking activity of acquiring digital resources, and by the opportunity to position their organizations to take advantage of distance learning rather than by any potential cost savings from current visual resources collections.

Visual resources library staff seem to view the digital image as just another format of slides, which will co-exist with current image collections. In this they see the benefits as being that they can draw from a larger universe of images, but I fear they may undervalue the textual information that comes with the images (both object documentation and contextual and interpretive data) and under-estimate the long-term threat that widespread availability of digital images with secured rights poses to existing slide collections. Because of the legal risks, I would anticipate that existing slide collections would be largely abandoned as sources of images for digital delivery.

Expanded use is a slippery but important benefit. At most universities today, faculty typically donØt use images because image research is too expensive and complicated, so most of the issues of teaching with images donØt arise. To the majority of faculty the benefits of the scheme would be to make possible the integration of visual resources into teaching. When they try, however, faculty will realize that many other infrastructure, training and pedagogical issues are involved in successfully exploiting images delivered over the campus network in classroom teaching and out-of-class assignments. Students, whether graduate, under-graduates or elementary and secondary students, have little experience using images as source materials or researching images. they rarely exploit images as part of their formal papers. To them the benefits of these new resources may reside in the raw material they provide to multi-media expression, which is increasingly going to be the form that their preferred mode of presentation takes.

Learning about benefits, through open-ended interviews, will be essential to the design of adequate and cost-effective services based on these resources. It will help us to understand the types of uses of a licensed image library, which includes both where the image was used (in a lecture, in a publication, in an exams, in a dormitory) and the purposes (to illustrate a point, to review classroom presentations, to create a new image, to test a technical assumption). Ideally interviews would help define the estimated dollar value of using the image library, which, when combined with types of uses, could provide a measure of the overall value of having the site license to the institution holding it.

To Section III