This is a revised version of the section in the Economic Sustainability Reference Model on the Key Entities that need to be understood to build a sustainability strategy according to this model. [Written by Brian Lavoie.]
The economic environment surrounding any digital curation activity is complex, involving a host of factors interacting to shape the circumstances in which a particular activity must operate. Understanding this environment, and its implications for achieving sustainability, can be challenging. Fortunately, it is possible to distill the economic environment into a few key elements that are present in all digital curation contexts. A thorough understanding of these elements and their properties is a first and necessary step toward building a successful sustainability strategy.
Digital assets are the raison d’être of a digital curation activity. They can take a wide range of forms: research data sets, e-prints, executable software, web sites, and so on. To qualify as a digital asset, two criteria must be met: first (and obviously), the object in question must be digital, and second, it must be judged to have a value that will persist over some period of time. The second criterion is crucial, and must be considered carefully by digital curators in the context of any curation activity: is the digital object in question truly a digital asset? The question is not trivial: not all digital objects are digital assets!
Each class of digital assets exhibits a variety of properties that to a greater or lesser extent have an appreciable impact on the nature of the curation activity itself. This can be readily seen from the perspective of the technical aspects of curation: the techniques and workflows needed to curate a collection of research data sets are likely to be very different from those needed to curate a collection of executable software. In the same way, the properties of digital assets impact the strategies needed to support the curation activity from an economic standpoint. As in the technical sphere, different kinds of digital assets will have special properties that impact sustainability in unique ways. However, digital assets also share certain core properties that must be taken into consideration when organizing a curation activity.
- Digital assets are durable yet depreciable: If kept in proper condition, digital assets can continue to release value for scholarship, private enterprise, education, and entertainment over extended periods of time. However, the clause “if kept in proper condition” is crucial. In order to maintain digital assets’ value over time, those responsible for their stewardship will need to expend resources on an ongoing basis to support their curation. Digital assets are depreciable in the sense that if not properly maintained, they will tend to succumb to technological obsolescence or bit rot, and in this way, their ability to release value to users will be reduced and eventually eliminated.
The implication of this property for economic sustainability is clear: in assessing the costs of a proposed curation activity, planners must look beyond the current costs of gathering digital assets into a collection and consider something akin to “total cost of ownership”. What will be the costs of maintaining the collection in a usable condition for an extended period of time? And what mechanisms can be put in place to support a sustained flow of resources to the curation activity? From the outset, curation planners must discard notions of one-time chunks of funding, and think instead in terms of ongoing flows of funding. [see BRTF Report, p. 25-26; also see Chapter XX of the reference model regarding the economic lifecycle]
- Digital assets can be curated by one, but used by many: A digital asset residing on a public server is, at least in principal, accessible to any Web-enabled user. In this sense, the user base for a curated digital asset is potentially far larger than that of a “physical” item – e.g., a print book. Moreover, a digital asset can be used simultaneously by many users, again in contrast to a physical item, where use by one generally precludes use by another (again, consider a print book). Expressed In economic terms, the use of digital materials is non-rival in consumption.
The important implication of this property for economic sustainability is that although the potential user community – in other words, those that benefit from curation – is potentially quite large, the incentive to contribute resources to the curation activity by any one user is generally weak. Once a digital asset is curated by one organization, it is at least in theory available for use by all; there is no need for each organization to curate a local copy of the digital asset, as long as a single curated copy is available on the Web. In contrast, there is a far stronger need to curate local copies of print materials, since a particular print copy can only serve a limited pool of primarily local users. Since one organization can curate a digital asset on behalf of all, it tends to be in the interest of any one organization to have someone else incur the trouble and expense of curation, while still enjoying free access to the digital assets. This is what economists call the free-rider problem, which can make collection of sufficient resources among beneficiaries to sustain the curation activity a challenge. To overcome this problem, it may be necessary to exclude some beneficiaries from access to the digital assets if they do not contribute toward curation. If the planners for a curation activity are not prepared to exclude non-contributors, or for some reason it is impossible to do so, this may create significant problems for long-term sustainability; the incentives for contribution to the activity will likely be weak. [see BRTF Report, p. 26-28]
Long-term accessibility to digital assets is achieved through the curation process, which is the set of activities involved in maintaining digital assets in a usable form for an extended period of time. It is important to distinguish between the process of digital curation, and the outputs of digital curation. In particular, the value of digital curation is experienced through the value-creating capacity of the curated digital assets themselves. Put another way, we generally do not value curation for the sake of curation; we value it because of the uses to which we can put curated digital assets. This establishes an important property of the curation process that is crucial when considering economic sustainability.
- The value of the curation process (and the associated digital information service in which it is embedded) is derived from the value-from-use of the curated digital assets: A curation activity that cannot make a compelling case for the value of the curated digital assets that it manages will find it difficult to attract the funding and other resources needed to sustain itself over time. This ties in with the distinction noted above between digital objects and digital assets. A digital asset is a digital object with a perceived future value. The value of the curation process, therefore, derives from its ability to deliver the value of the digital asset. In economic terms the demand for digital curation services – i.e., the curation process – is a derived demand: in other words, it is derived from the demand for curated digital assets. [see BRTF Report, p. 24-25]
Naturally, many of the activities associated with the curation process are technical in nature, consisting of curation techniques and workflows invoked on the digital assets to ensure they persist in a suitable condition for use. For the purposes of economic analysis, however, it is not the technical aspects of the curation process which are key; instead, it is the decision-making process that overarches the day-to-day management of curated digital assets. The decision-making processes associated with digital curation also exhibit several properties relevant to consider in regard to economic sustainability.
- Curation decision-making is path-dependent: Digital materials pass through a sequence of stages, also called a lifecycle, with endpoints of creation and disposal. Decisions made at one stage of the lifecycle often shape the choices available to decision-makers at later stages. This point is illustrated in its starkest terms by the initial choice of whether or not to curate; if the decision is “no”, it is often impossible to revisit and change this decision at a future date, because by then the digital asset may be unavailable or have deteriorated to such an extent that effective curation becomes prohibitively expensive or even infeasible. [see BRTF Report, p. 28-30]
Stakeholders and the Stakeholder Ecosystem
The network of stakeholders surrounding a particular curation activity can be complex and difficult to characterize. Moreover, these stakeholders can represent an equally wide range of interests in regard to the long-term future of the assets in question. The organization of these stakeholders, and in particular, the distribution of curation roles across them, heavily impacts the prospects for achieving sustainability, and by extension, the shape of the sustainability strategy best suited for those circumstances.
In order to understand the basic contours of the stakeholder interests associated with a set of digital assets, it is useful to employ an organizing device called a stakeholder ecosystem. A stakeholder eco-system designates the key stakeholder roles in lifecycle digital management, and articulates important relationships between them as they relate to sustainability. The stakeholder ecosystem takes the form of a high-level abstraction that helps organize stakeholders into broad categories of interest. Figure XX provides a view of the types of stakeholders associated with a digital curation activity. It is important to recognize that the stakeholders illustrated in the Figure are roles, and not necessarily distinct entities. A single organization (or individual) can fulfill multiple roles simultaneously. As we will see, this can have important implications in terms of creating conditions that either encourage or discourage long-term sustainability.
Figure XX: Types of Stakeholders
The stakeholder taxonomy depicted in Figure XX is divided into three areas:
- Supply-side: stakeholder roles pertaining to those who create and/or own the digital asset.
o Creators: entities responsible for creating digital assets (e.g., scientist who creates a data set; artist who creates a digital work of art; company that develops a new computer game)
o Rights Holders: entities who currently own key intellectual property rights related to digital assets, such as the right to provide access, the right to preserve, etc. Oftentimes this will be the same entity that created the digital asset (ie the same stakeholder has both the Creator and Rights Holder roles), but not always. For example, a publishing company may hold the rights to an e-journal article authored by someone else; a social media site may own the collective contributions of its members, etc.
- Demand-side: stakeholder roles pertaining to those who benefit from availability of the digital asset.
o Current beneficiaries: those who currently have well-defined uses for the digital asset, and derive value from its ongoing availability.
o Future beneficiaries: those who could be expected to benefit from the digital asset in the future, or those whose uses of the digital asset are as yet unknown/undefined. The BRTF report made a point of noting that the interests of future beneficiaries are often left out of preservation/curation decision-making.
- Lifecycle management: stakeholder roles pertaining to those responsible for curating the digital asset, and for ensuring its ongoing inclusion in the cultural/scholarly record.
o Managing Agencies: entities responsible for managing/curating/preserving/providing access to digital assets. Essentially, this is the entity that actually has custody of the material and ensures that is continues to exist and is accessible for use.
o Representatives of the Public Interest: entities who advocate for the digital asset’s inclusion as part of society’s ongoing cultural or scholarly record. The BRTF report focuses on digital assets in which there is a clear long-term public interest. Generally speaking, these are materials that potentially would be included as part of society’s cultural record or scholarly record (acknowledging that these terms are only vaguely defined). Typically cultural heritage institutions (libraries, archives, museums) play a prominent role in advocating or actively intervening to ensure certain materials are in fact curated on behalf of the long-term public interest (this role is noted in the BRTF report). Various philanthropic organizations, funding agencies and informal volunteer networks would fall into this category as well.
o Resource providers [Investors above]: this term is used broadly for those who provide resources (i.e., funds, in-kind transfers, etc) to support the curation process. Provision of resources can occur in a variety of ways: for example, through payment of a price-for-services (e.g., where the Managing Agency is a third-party digital archiving service and the Resource Provider is a customer of this service); donation (e.g., where the Resource Provider is a corporate sponsor of a digital collection or of a curation organization); grant award; advertising (e.g., where the Resource Provider pays the Managing Agency for the placement of advertisements in association with the latter’s access services); and so on.
As Figure XX illustrates, many different categories of stakeholders are associated with a digital curation activity. But this in and of itself does not create an ecosystem. The stakeholder ecosystem is created by the pattern of relationships between the stakeholder roles that holds for a particular activity. This pattern of relationships can be quite complex, and is illustrated by Figure XX.
Figure XX: Stakeholder relationships
Relationships in the ecosystem define how the various stakeholder categories interact with one another. For example, in many digital curation contexts, Representatives of the Public Interest such as libraries and archives act on behalf of Future Beneficiaries by making sure that the latter’s interests are represented in today’s curation decision-making. Similarly, the Rights Holder and Managing Agency often have a relationship in that the former must grant an exclusive or non-exclusive right to curate to the latter. Investors also have a relationship with the Managing Agency, in that they supply resources sufficient for the Managing Agency to carry out its curation activities.
The pattern of relationships depicted in Figure XX is not necessarily complete – many other relationships between the various stakeholders could be imagined. However, it does suggest the potential complexity that might prevail in the pattern of relationships associated with a particular digital curation activity. It also points up the importance of orchestrating the interaction of all stakeholders in a sustainable digital curation activity. A sustainability strategy that neglects to account for the interests and participation of any of these stakeholder categories is likely to founder at some point in the economic lifecycle.
It is important to emphasize again that all of the stakeholders represented in the ecosystem (i.e., in Figures XX and XX) are roles, and not necessarily distinct entities. So a single person or organization could simultaneously fill multiple roles. Many of the important issues regarding incentives to preserve, allocation of long-term curation responsibilities, etc. hinge on whether various roles are combined within a single organization, or separated across distinct entities. The more that stakeholder roles are dispersed across distinct entities, the more negotiation and consensus must be achieved to reach sustainability goals, and the greater the probability that individual interests will conflict. This leads to an important property of stakeholders that has enormous implications for the prospects of achieving long-term economic sustainability:
- The distribution of curation roles across the network of stakeholders is critical for understanding both the factors encouraging sustainability in a particular digital curation context, as well as the risks that might prevent sustainability from being achieved.
The ecosystem diagram illustrates the major categories of stakeholder interest in curated digital assets. The distribution of stakeholder roles across distinct individuals or organizations is a key factor in influencing the nature of the economic risks associated with a given curation activity, and the kinds of economic remedies that are appropriate for addressing them. For example:
- The incentives to curate are strengthened when the Rights Holder for a particular digital asset is also a Current or Future Beneficiary. They are weakened when this is not the case.
- Allocation of responsibility for undertaking curation is clearer when the Rights Holder is the same as the Managing Agency. When this is not the case, it is less clear who should undertake the curation process, especially when ownership (i.e., the Rights Holder role) is split among many entities.
- The interests of Future Beneficiaries are often not represented in the curation decision-making process. Do Representatives of the Long-Term Public Interest adequately represent these interests and are they being taken into account in today’s curation decision-making?
- When Investors are distinct entities from Current and Future Beneficiaries, it is vital that a compelling value proposition is made for providing resources to support long-term curation.
In summary, the stakeholder ecosystem is derived from the layering of three components one on top of the other: first, the set of stakeholder categories depicted in Figure XX; second, the pattern of relationships between stakeholders depicted in Figure XX; and third, the distribution of stakeholder roles discussed above. The contours of the stakeholder ecosystem associated with a given digital curation activity are fundamental in determining the potential “traps” that might impede the activity from achieving long-term economic sustainability. By extension, then, the design of an effective sustainability strategy to mitigate these traps must rest on a thorough understanding of the underlying stakeholder ecosystem.
The impact of the stakeholder ecosystem on the economic risks faced by a digital curation activity can be illustrated with an example. ArXiv is a repository for e-prints in physics, mathematics, and other sciences. Access to the contents of arXiv is free and open to all. Over the years, arXiv has become a key mode of scholarly communication between scientists, who rely on it both as a means of exposing their work to colleagues around the world and eliciting comment and discussion. Cornell University currently serves as the Managing Agency for arXiv, and prior to 2010, was the sole Resource Provider as well. Although faculty and students at Cornell certainly benefit from use of arXiv, the vast majority of those who obtain value from arXiv – that is, the Current Beneficiaries – are unaffiliated with Cornell University.
A key feature of arXiv’s stakeholder ecosystem is that the Resource Provider is distinct from Current Beneficiaries. Those who benefit from using arXiv are not obligated to contribute toward its costs, and indeed have a weak incentive to do so, since they are not excluded from the benefits even if they contribute nothing. This creates a free-rider problem, since those who benefit from arXiv are free to do so without contributing resources to its ongoing operations. This situation was sustainable only as long as Cornell University was willing to fund arXiv’s approximately $400,000 annual budget on behalf of the general scientific community. This in turn introduces an economic risk into the curation activity: the Current Beneficiaries have little incentive to provide resources to support a curation activity undertaken in their interest, which calls into question the ability of the curation activity to marshal sufficient resources on an ongoing basis to continue its operations. One way to mitigate this risk is to induce or persuade Beneficiaries to contribute to arXiv – that is, to combine the roles of Resource Provider and Beneficiary in the same entities.
In 2010, Cornell made its first step toward doing just that, by issuing a call to other institutions whose affiliated faculty and students benefited from arXiv to make voluntary donations to support arXiv’s operations, with the requested donations scaled to each institution’s level of use (as measured by number of downloads). In describing this new funding strategy, Cornell noted that “[s]cholars worldwide depend upon the stable operation and continued development of arXiv. Sustainability is best assured by aligning revenue sources with the constituents that realize value from arXiv, and by reducing dependence upon Cornell University Library’s budget. We have decided to pursue a collaborative business model that will engage the institutions that benefit from arXiv.” In short, Cornell has realized the potential pitfalls of a stakeholder ecosystem where beneficiaries are completely distinct from Resource Providers. The donation program is a first step toward re-aligning the ecosystem such that the roles of Resource Provider and Beneficiary become merged within the same entities. It is significant to note, however, that Cornell considers the donation program to be only an interim solution until a long-term business plan can be worked out. An obvious weakness of the interim plan is that it is voluntary, with no “repercussions” if institutions choose not to participate. A long-term strategy might consider restricting arXiv access only to those institutions who contribute toward its upkeep. However, given its long history as a freely available resource, such a strategy is bound to encounter resistance.
This example demonstrates that knowledge of the stakeholder ecosystem underlying a particular digital curation activity helps identify potential economic risks impacting long-term sustainability, as well as possible remedies to overcome them.
[Update to add links