Archives & Museum Informatics

Conferences

Publications

Seminars

Consulting

Research

Search archimuse.com

Share

last updated:
November 30, 2017 12:56 PM

MESL: About MESL

New Economic Models for Administering Cultural Intellectual Property

Section III

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

IV. Models for Future Cultural Intellectual Property Management Systems

There are several models of what a future licensing environment for cultural intellectual property might look like. One model, being advanced by Corbis, Kodak Picture Exchange, and similar commercial interests, is the agency resale model. Another, being advanced by Picture Network International (the Seymour System) and Academic Pressþ planned Image Directory might be thought of as the brokering model. A third, suggested in outline by the MESL project and by DACS in the UK, is the cooperative of rights owners. Obviously, the current system, in which the rights holders license rights on a product by product basis, could be regarded as a fourth model but I have already said enough about its limitations previously and will not discuss it here.

In effect, these models produce very different results for rights holders and end-users although they look very similar at the point of technical implementation. each requires:

Let me discuss issues raised by each in turn, beginning with the agency resale model. In this model, rights holders contract with an agency to re-license their rights. The rights holdersþ fees may be paid as a fixed advance or as a royalty based on use or as a percentage of income, but the agency makes the bulk of the profits in return for building the central library and providing the value-added access services and absorbing the re-licensing costs.

In the brokerage model, the rights holders set their individual prices and collect income based on use, less the percentage for administrative fees charged by the broker. They determine what types of uses are allowable and do not grant re-licensing rights. Digital brokerage can work in either a centralized or distributed mode, with more or less of the costs of the license negotiation and fulfillment borne by the broker or the rights holder.

In the cooperative model, the rights holders own the licensing and distribution mechanism wlaws of most countries as a special kind of entity (a Rights and Reproduction Organization) exempt from anti-trust regulation, price fixing and monopoly restrictions. The RRO absorbs its administrative costs as an overhead and distribute profits among rightsholders according to a formula to which they agree. Its power derives from its position as the sole source of rights for a given kind of intellectual property.

In the product licensing model which is today's approach, each rights holder negotiates with each product producers separately and reaches agreements with them. While the rights holder retains all the income and control, there is less of a market for rights than anticipated since the burden of shopping for the correct images falls to buyers. When compared to any of the three collective models, this approach raises many of the same issues: what are the economies of scale and is there a point at which there are diseconomics of scale?

The agency and the cooperative incur relatively heavy expenses signing up rights holders and consumers. As a consequence they will favor suppliers and users of large quantities of data over those with small collections or single uses. Of course the agency may incur the added costs of paying for rights up front, but this is not necessarily part of the model. The brokerage operates on an item by item level and incurs relatively high expenses tracking transactions. These expenses may also be incurred by agencies which pay royalties.

The agency approach may produce dis-economies of scale at a certain level of participation because of the present limitations of computing and networking technologies since it is, by definition, a central system. The brokering model implies a central directory with distributed holding of the licensable content. The cooperative can employ both central and distributed approaches in response to technology constraints. The advantages to the centralized provision of rights and data are that both terms and data delivery can be standardized while the distributed model will generally be implemented without such control, and indeed because central control is considered unacceptable.

The three models are also quite different in their ability to deal with competition. The agency model, having typically paid up front for rights, needs to sell them and if necessary will do so by reducing the fees and trying to increase market demand. Its basic strategy is to create a mass market, if necessary by price based competition. The brokering model is able to accept any amount of competition as long as the rights holders do not offer the items they place in brokerage at more favorable rates elsewhere. The competition is between items in the brokered inventory. The cooperative is sensitive to any availability of individual items from other sources. The classical RRO doesnþt have any such competition and the MESL model, if extended to include more institutions but still dependent on site licensing would be vulnerable to item licensing competitors (including the museums themselves acting outside the collective) because potential site licensees might decide to acquire items on an as needed basis from other sources rather than subscribing to a full library at a fixed annual fee.

One of the uncertainties dealt with in any economic projections today has to do with the size of the market for cultural intellectual property and the degree to which cultural institutions will participate in any of the proposed models. While it is virtually impossible to predict the market share that would be controlled by any of these potential future systems, a model could incorporate an estimate of size of the licensable pool of data based on the percentage of institutions participating. At some point the snowball effect of critical mass will cause a large percentage of organizations to join a working scheme.

In previous articles Iþve argued that cultural intellectual property will only succeed in earning money for museums, archives and libraries if they deliver a quality of content beyond that which the commercial market is making for general consumer info-tainment. One of the unanswered questions of the economic study is what such added quality would cost cultural content providers. The specific answer will have to wait until further analysis of which qualities would best differentiate the product from commercial offerings and add to its value as perceived by consumers.

Each of the new models departs from the existing system by providing a central directory from which the contents of a large number of cultural heritage institutions are available rather than requiring potential licensees to contact each institution separately. Each model involves slightly different levels of fixed costs however for the maintenance of the central directory since each implements the system in a different way. Under the agency approach, the central directory CO-exists with all the digital contents. In the cooperative as implemented under MESL, many copies of the full directory and digital contents re implemented (although a future MESL model is unlikely to perpetuate this redundancy for hundreds more institutions). In the brokerage model the directory is free standing, and the contents are resident in each content providing institution.

V. Unknowns

One of the reasons that the topic of electronic museum services is of critical interest is that there is a widespread expectation throughout modern industrialized societies that significant new money making opportunities reside in the exploitation of information resources. It is anticipated that over the next decade consumers will increase the amount of money they have been spending on the acquisition of images and information from all sources. In part, it is suggested that this increase will be a consequence of shifting consumer expenditures from material goods and in part it will occur because the merger of information and entertainment allows some money to slip across ®education/information þ boundary which would previously have been spent on entertainment through TV, cable and film. Whether such a shift in consumption patterns will actually occur, of course, a huge unknown.

Nor is it possible to guess whether the specific model adopted by a new licensing system would impact on the influx of new money, not the least because of uncertainty surrounding just how licensed images and information might be distributed to the end consumers. The models which are most often discussed - fixed media publication, Internet services, and narrowcasting - involve quite different distribution systems and have different economic consequences. In the fixed media business, the publishers acquire rights for products or titles and disseminate these in traditional ËeditionsÓ to end-consumers. The number of end-user consumers will be restricted up front as the size of the edition. They will be interested in acquiring large numbers of rights through a single system. In the Internet services model, end-users will largely be acquiring rights on a one off basis with the number of users exactly equal to the number of downloads. In the narrowcasting model, both push (publishes) and pull (online search) modes could exist, but the number of users will not be known except through ®audience estimatingþ methods. In any of these models the per image license fee needs to be very small, and probably will end p being less than the costs of auditing and collecting individual uses; only the Internet model provides an opportunity for as much as $1 or so per image.

Another open question in the economic equation is whether the cultural sector will compete with itself. I believe that the holdings of cultural institutions, compiled as they were for scholars, are highly redundant from the point of view of consumers. Elsewhere Iþve described this by saying that individual holdings are fungible. If so, two unintended and undesirable consequences are likely. First, the prices that cultural institutions charge for licensing individual items in their collections will be driven down since the ËsameÓ items (from the perspective of the average non-expert) are available from another source for less. Second, cultural institutions will not be able to afford to digitize most of their collections because once they ®skimþ their holdings for the most salable properties, there will not be any percentage in adding more. The concept of site licensing of whole collections was in part developed in order to defeat this economic barrier. It also has a major advantage for both the cultural content providers and users in that it encourages increased use of the resource becasuse increases in use reduce the unit costs (because annual fees are fixed without respect to use levels) while in other models increased use translated into greater expense.

In addition, the entire premise behind capturing digital data for re-use by the market is impeded by a tremendous unknown: to what extent are digital intellectual properties re-usable? The premise behind digitally capturing images, sounds, video and even text relating to cultural properties is that more than one user will be able to take the digital file and use it, or one user could use it in a variety of products. The problem with this concept is that we really know almost nothing about the costs of reconfiguring data for eventual publication or dissemination, whether in the traditional analog environment or the newer digital one. There are many reasons to doubt that any one method of capturing an audio-visual record of an artifact will be entirely appropriate to a range of different products. Should we provide a video file with rotation and close up views or would multiple still images with a calculation of the movement between each be preferable? Should we shoot very high resolution images or are any of these larger files even worth saving? Among the things we donþt really know are whether economies of scale could be achieved by any other means than in capture, for example in delivery or in post-production, through making specific choices about formats? We do know that different products will place a different value on the image(s), the tombstone data, the extensive curatorial files, or associated educational and interpretive materials. Of course the cultural institution wants to encourage, or even insist on, use of its tombstone data and background documentation. This argues against pricing them separately, but perhaps pricing should be structured in such a way as to penalize those who donþt use the cultural institutions interpretation?

Finally, the pricing of cultural intellectual property will be sensitive to whether it is being licensed for a long or short period, the types of uses to which it will be put, and its commercial value to the licensee. Public policy is also likely to enter the picture since the provision of cultural information is considered a social benefit, and most countries are anxious to promote their heritage and identity. In addition, pricing will be dependent on ability to pay, which will to some extent depend on whether the new system replaces the old or simply lives side by side. For example, if digital image acquisition from agencies, brokerages or cooperatives replaces the analog slide collections of universities, money will be released to participate in the digital market, but if the digital library lives side by side with current slide libraries, or even is managed by the slide librarian, then there will be fewer funds.

Conclusions:

The development of rational systems for exploiting intellectual property is a major challenge for cultural institutions. There are experiments underway that help to identify the critical intellectual, social and economic issues and which can be expected to help in the design of a better cultural properties licensing mechanism. This paper identifies many of the measures that will need to be collected in order to model a new economic system; at the same time it identifies numerous unknowns and many areas in which chance and timing will probably play a large part in success. As cultural institutions move cooperatively towards the definition of new structures for intellectual properties administration they will need to keep testing the assumption on which the strategies were based against the changing realities if the market.

APPENDIX A: Proposed study

The study is designed to provide data to assist in defining the best structure and pricing for a future not-for-profit entity to pool museum digital resources for use in education. The study should enable comparative analysis of different pricing scenarios. All scenarios are for annualized license, unmetered use based on the decision that increased use should be encouraged rather than inhibited and that per-transaction fees would impose an unnecessary overhead for this market. The study has four components: 1) establishing the baseline costs and perceived value of the existing mechanisms for distribution of museum images and data for educational use 2) estimating the projected costs and benefits of providing a museum content with methods reflecting MESL's experience and a variety of scenarios for participation 3) determining the overhead costs of administering a future service, including especially determining the recovery points for various fixed costs and the best volumes/service mixes for lowest marginal costs, and 4) measuring the market demand for a variety of different services at different price points. The study will identify fixed and variable costs and technology implications (risks) of each scenario. A market study will then identify price points and volumes required for viable operation. The product will be a model of an effective distribution system as well as the business plan for an organization that directors of current MESL institutions would accept. The study is designed to assist in determining variables in fees for participation in a future RRO and for site license pricing, so it will collect data coded according to at least the following dimensions:

Institution type
University
College/Other Post Secondary Institution
School District/System
Individual School
Museum
Institution size (Population served):
100,000
50,000 - 99,999
10,000 - 49,999
5,000 - 9,999
1,000 - 4,999
<1,000
Uses:
Personal use only
Teaching
Teaching and Library/Student Resource
Teaching, Library/Student Resource and Research/Research-based Publication
Support:
Unsupported (load own data and access tools)
Access tools provided for local server
Access tools and use of central server
The market analysis will include assessment of scenarios for income redistribution to providers including:
No overhead options:
1) All suppliers share equally
2) Suppliers share proportionate to number of records contributed
Costs in calculation options:
3) Suppliers share based on overall quantity/quality rank (weight given to documentation accompanying each object)
4) Suppliers share based on statistical sampling of use
Management Options:
1) Management role for suppliers
2) Management role for suppliers and subscribers

Appendix B: Interview Framework

Interviews could also determine whether individuals were changing their behaviors as a consequence of having access to such a licensed library. For example, whether they previously used visual images, and if so, how and from what sources. If interviewees used MESL, one could identify variables in satisfaction in MESL and non-MESL uses, any benefits they perceived in digital access (especially if not yet identified by us), and their expectations for the future (after seeing prototype/demo) Detailed interviews could reveal the difference between the costs of current and future methods at a detailed level of process and for different kinds of institutions. A matrix of these might look like:

University/College Public/Private School Museum Individual
Cost Savings:
picture research
acquisition
cataloging
labeling
storage
loan/refiling
restoration/replacement
reference facility
billing/rights administration
Value Added:
image quality (scale/color)
authoritative data
remote availability
ease of integration
searchability
local print
digital watermark
use tracking data
Opportunity Costs:
simultaneous use
student use
off-site preparation
potential commercial sale
increased image pool
comparison/analysis
new teaching methods
competitive advantage
New Institutional Costs:
Infrastructure enhancement
Marginal cost of storage
Admin. overhead of consortium
New Institutional Benefits:
Source of photography/digitization $
Basis for inter-institutional data exch.

Footnotes

1 Elsewhere I've discussed some of the new products and services which museums might develop, the problems they confront in selecting appropriate standards and the general question of why the world would want intellectual property developed by cultural institutions over that which is likely to be available from commercial interests. See: David Bearman, "Information Strategies and Structures for Electronic Museums", Information: The Hidden Resource, Museums and the Internet. Proceedings of the Seventh International Conference of the MDA, 1995 ed. by Anne Fahy and Dr. Wendy Sudbury (Cambridge,UK, Museum Documentation Association, 1995) p.5-22

2 For a variety of reasons publication of series and provision of consistent information through regularly updated information services will be more cost effective than single titles or a providing access to a relatively stable database. These reasons include the potential which repeated content delivery has for recouping some "first copy" costs, the benefits of normalizing production methods and integrating content creation into other everyday functions of the museum, and the benefits of reseals based on building customer loyalty.

3cf. Lyn Elliot Sherwood, "Moving from Experiment to Reality: Choices for Cultural Heritage Institutions and their Governments", fn. 11

4 Jennifer Trant, "The Getty AHIP Imaging Initiative: A Status Report", Archives and Museum Informatics vol.9 #3, 1995

5 This planned database has not yet established a fee structure nor proven that museums and archives will be willing to use a reference database publisher as an image broker, but it is expected to get underway in 1996.

6 The framework for analysis of this series of interviews is reproduced as Appendix A.

7 David Bearman, "Museum Strategies for Success on the Internet", Museum Collections and the Information Superhighway (London, Science Museum, 1995) p.15-27; also published in Spectra, vol.22 #4 p.18-24

Informatics: The interdisciplinary study of information content, representation, technology, and applications,
and the methods and strategies by which information is used in organizations, networks, cultures, and societies.