This chapter explains the DOI business principles, business/ organizational model and the DOI approach to standardization and the role of the "maintenance agency for the DOI that IDF fulfils.
10.1.1 Cost recovery as the basis of DOI deployment
The implementation of the DOI system will add value, but necessarily incur some costs. The three principle areas of cost are:
Once assigned, DOIs are free in use; whenever a DOI is encountered, it can be "clicked on" and used. The costs of running the system are borne by the registrant.
Our aim is that the system should become self-financing; costs for running the system should be recouped from those who benefit from the system. Cost recovery from registrants, not end-users, is a practical measure: the tasks should be delegated to appropriate organizations who can offer appropriate expertise, economies of scale, synergy with existing operations, marketing presence, etc. (in the case of number registration, these organizations are referred to as "Registration Agencies" (RAs) -- see Chapter 11 The appointment & role of Registration Agencies).
10.1.2 The creation of Registration Agencies
When DOIs were launched, no third party DOI Registration Agencies existed. Early users of DOIs interacted directly with the International DOI Foundation; the IDF was therefore the only Registration Agency. It was recognized that in order to be widely deployed, the system needed to be widely distributed, and the model chosen (discussed in more detail below) was to appoint third party registration agencies -- essentially holders of a franchise of the DOI system. The IDF then becomes more like a wholesaler and the Registration Agencies become retailers. Registration Agencies, rather then IDF, will become the face of the DOI system as far as the end customer id concerned.
Registration Agencies are now being appointed, in a gradual process of migration to this franchise model. It is therefore necessary to appreciate that the policies and procedures related to DOI allocation are moving from IDF to this network or federation of Registration Agencies. An aim of this section of the Handbook is to provide more information on the consequences of this transition for users of the system.
When a critical mass of registration agencies have been appointed the fees charged by the IDF to existing "direct" prefix holders will be reviewed, as will the rules and procedures applicable to such prefix holders. The IDF will offer limited pricing flexibility and DOI functionality if prefixes and DOIs are registered directly with IDF, and will recommend that DOIs are obtained from approved Registration Agencies (rather than direct from the Foundation) to take advantage of pricing schemes and related applications designed to suit the needs of the particular intellectual property community concerned.
At the outset of the DOI development, a very simple model was introduced whereby a prefix assignment is purchased for a one-off fee. A fee was introduced not to cover actual costs, but to recognize the fact that some charging for DOIs would be the intention. IDF used a simple initial economic model: a charge of $1000 for allocation of a prefix (a one-off charge) allowing unlimited number of DOIs to be constructed using that prefix. The current charge is one-off and entitles the registrant to an infinite number of suffixes; there is no annual fee, though we reserve the right to vary this at a future date; there is no limitation placed on the number of DOI prefixes that any organization may choose to apply for. It was recognized at the outset that this fee structure was a starting point but would be insufficiently flexible for the long term.
DOIs allocated using these prefixes purchased directly from IDF are registered without structured metadata: they are "zero Application profile" (see Chapter 5 Metadata). The disadvantage of using the direct $1000 route is that there is no metadata support and no social infrastructure support of the type, which can be given by a Registration Agency such as CrossRef.
We are now in a process of migration to a wide variety of potential business models, using third part registration agencies, in recognition of the fact that such a simple model is not a "one size fits all" solution. The direct prefix purchase route is still an option, but our intention is that eventually all future DOIs will be registered through one of many Registration Agencies, each of which will use one or more defined DOI Application Profiles (see Chapter 5 Metadata), and each of which is empowered to offer much more flexible pricing structures. The pricing structures and business models of the Registration Agencies will not be determined by the IDF; each RA will be autonomous as to its business model. Business models for these agencies could include, but not be limited to, cost recovery via direct charging based on prefix allocation, numbers of DOIs allocated, numbers of DOIs resolved, volume discounts, usage discounts, stepped charges, or any mix of these; indirect charging via cross subsidy from other value added services, agreed links, etc. The IDF will place minimal constraints on the business models offered by RAs, and enter into discussion on practical implementation of any of these (see also Chapter 11 Metadata and Chapter 13 Operating procedures).
The customer should be interested in "what does the retailer charge". An RA will provide a service -- e.g. CrossRef. One of the things they will do for that service is allocate a DOI -- and the metadata (or help with it, or specify it, or...). But it's not the ONLY thing they do. So you can't look at the charges of an RA and say "that's what a DOI costs". Crossref don't charge a single fee but it works out at anything up to 60 cents per DOI allocated.
DOI RAs may find it beneficial to develop new DOI applications for their customers, or to the same market segment, in order to widen the potential for use and income stream from their DOI activities. In some other sectors, products created as a spin off from basic registration activities provide the funding to cross-subsidize and create a low price for registration itself -- a "positive feedback loop".
The one-prefix/one fee model is insufficiently flexible not only for business reasons but also for technical reasons. We want to encourage the appropriate use of DOI prefixes without undue financial penalty. For example, to encourage multiple prefixes within a single organisation (this may prove administratively convenient especially in large organizations); or at a different level of granularity (e.g. prefixes allocated to imprints, record labels, image libraries, magazines, journals, etc). A fixed fee per prefix does not do this.
While the migration to these more sophisticated business models is under way, and market development is being undertaken by Registration Agencies, the IDF will continue to offer only a basic and relatively inflexible prefix-based fee structure. We receive many requests for flexibility in prefix allocation or costs. We will attempt to deal with these requests sympathetically but must point out that we have limited options in designing an equitable pricing scheme to suit every need before the appointment of specific registration agencies.
Can DOIs be made available at no charge? Yes.
(a) IDF is willing to allocate a DOI prefix free of charge to organizations for limited experimental non-commercial uses. Please contact us if you wish to apply for this.
(b) The business model includes two separate steps: a business relationship between IDF and an RA (the "franchise fee"); and a business relationship between an RA and a DOI registrant (the "registration fee"). The two are not directly connected; this enables the RA to offer to registrants any business model whatever, which suits its needs. This could include assigning DOIs without charge. Hence DOIs can be used in both commercial and non-commercial settings, interoperably. However, the franchise fee in such an example cannot be zero; this would immediately undercut any commercial use, and it would not provide any financial support for the operation of the system itself. Like any other piece of infrastructure, an identifier system (especially one which adds much value like metadata and resolution) must be paid for eventually by someone. So an organization could, if it wished, assign DOIs freely (registration fee zero to registrants) and subsidize this added-value service by paying the franchise fee to IDF.
10.3.1 The starting point
The IDF organisation was set up on a similar model to the World Wide Web Consortium (W3C). The funding is from Members, and there is no direct relation to the operational running costs of the DOI system. The member-based IDF currently subsidizes all the operational running of the system specific to the DOI implementation of Handle technology. The members of IDF, as with members of W3C, pay a membership fee to support development of the system as a pre- competitive standards activity, which when widely implemented will enable costs savings or new business opportunities in the community (an analogy: allowing the tide to enter the yacht harbor, enabling each yacht to float free but with no advantage to any one over the others). The IDF is run by its Members, via an elected Board and appointed Director.
The Foundation was created in 1998 and supports the needs of the intellectual property community in the digital environment, by the development and promotion of the Digital Object Identifier system as a common infrastructure for content management. The Foundation is controlled by a Board elected by the members of the Foundation, with an appointed full-time Director who is responsible for coordinating and planning its activities. Through the elected Board, the activities of the Foundation are ultimately controlled by its members. Membership is open to all organizations with an interest in electronic publishing and related enabling technologies. The aim in the long term is to reduce reliance of the DOI system on this membership fee, enabling the fee to be reduced, abolished, or redirected (as the Membership decrees), and seek income directly related to DOI system usage in order to cover costs.
The IDF currently contracts with various technical providers e.g. CNRI and will similarly contract with any other organizations to which operational tasks are delegated. IDF will continue to control the relationship with the global resolution provider, on behalf of all RAs. RA's are free to subcontract or partner with others to deliver part of their services.
The International DOI Foundation, Inc, is a non-stock membership corporation organized and existing under and by virtue of the General Corporation Law of the State of Delaware, USA. The Corporation is a "not-for-profit" organization, i.e. prohibited from activities not permitted to be carried on by a corporation exempt from US federal income tax under Section 501(c)(6) of the Internal Revenue Code of 1986 et seq.
10.3.2 The future Operating Federation organisation model
A different sort of organisation typifies a mature or established standards deployment activity, e.g. ISBN, EAN, VISA. The EAN model is perhaps the closest to likely mature DOI system yet identified. In such a model, the operating entities which deploy the standard (in the case of DOI, we call these Registration Agencies (RAs)) find it advantageous to form a federation structure. The Operating Federation is run by the Agencies via an agreed structure and with appointed central staff.
The "Federal principle" specifies that users of the system make decisions at the lowest appropriate level; a governance layer ensures interoperability of lower levels; the governance layer implies "minimal constraint". From this it follows that each RA is free to determine its own business activities, constrained only by the agreed level of federal governance. This is a market economy model: in a market economy, anybody can trade with anybody, and they don't have to go to a market square to do it. What they do need, however, are a few practices everyone has to agree to, such as the currency used for trade, and the rules of fair trading.
In order to consider the development of such a structure for the DOI system, it's useful to project forward to a stable mature economic model, and then to work back to how this might be achieved.
10.3.3. Economic model for Operating Federation
This is analogous to e.g. the EAN (physical bar code) business model. The purpose of the Operating Federation is to maintain and control the few practices everyone has to agree to ensure the health of the system.
The business model of each RA is determined by the RA itself. The RA enters into agreements with its customers (DOI assigners); these may be radically different between RAs, depending on the needs of its customer communities, its own other business, the value-added services it offers, competitive pressures, etc. The agreements will however require conformance to the overall federal standards.
The RA enters into a contractual agreement with the Operating Federation, guaranteeing the conformance to minimal criteria covering conformance to technical, information management, and economic criteria. The "economic criteria" for being a member of the Federation is a payment to support central Federation governance: a "participation fee", or "franchise fee".
In the mature model, all RAs agree that certain high level or "central" functions are to be carried out not by one RA but by the central Operating Federation organisation on behalf of all. The costs of carrying out these agreed central functions result in an annual operating budget. The budget costs are apportioned across all RAs in an agreed fashion, resulting in the "participation fee" paid by each RA into the central Federation. The sum of the participation fees matches the budget costs. If the proposed central budget is agreed by the RAs (who govern the Operating Federation) to be too high, RA's must either agree to decrease the central functions, or modify their business model to generate the necessary increased participation fee. The participation fee can be viewed as the minimum cost necessary to participate in the system and gain access to the infrastructure, technology, existing brand value, franchise materials, etc.
The percentage of total central Operating Federation costs borne by RA can be determined by a fixed criterion (e.g. in the EAN system, by a formula based on potential market size), or dynamically by agreeing on a pre-set formula (e.g. for DOI it could be a function of number of prefixes, number of DOIs, and number of resolutions).
For illustrative purposes only, let us assume that central Operating Federation costs in a mature (long-term) model are $2M p.a. If there were 50 RA's, then average costs per RA = $40,000 (which seems not unreasonable -- each RA is bearing costs equivalent to employing a middle-order administrative person). In the early days of setting up RAs, this model alone cannot support the whole system. If there were for example only 2 RAs, it would not be feasible for them to apportion central costs of $2m between them (= $1M each). A more realistic course is for the Operating Federation to come into existence but for its costs to be derived from two sources:
As the number of RAs increases, the proportion of costs contributed from Operating Federation increases and from Membership declines. Consider a possible migration path (this is necessarily a simplification for illustrative purposes, assuming simply one fixed participation fee, but the same principles can apply to a more dynamic cost apportioning model and multiple year budgets).
These figures ignore the effect of volume charges for number of DOIs beyond the minimum; these charges will substantially increase the revenue derived from RAs and decrease the need for member-based subsidy. A more detailed financial model is used in practice to allow for these factors.
In practice, it would seem likely that a variation on this model would be feasible. Some areas of development costs may not be essential to the existing Operating Federation, but some potential RAs or members may want to see these developed: e.g. future functionality beyond that already provided. A structure can be conceived which supports both and where governance is via two "chambers", embracing both the current organisation model (for "Development") and the Operating Federation model. An organisation which is both a Member of the Foundation and an RA (participant in the Operating Federation) would receive some benefit (e.g. votes in both "chambers" but probably also some reduction in one or other fee).
We have begun the transition to this form of model by the creation in 2001 of a new category of membership, that of "Registration Agency". Registration Agencies currently have 25% of the seats on the controlling board of the Foundation. It is the intention that this percentage be increased in proportion to the percentage of revenue derived from RA operating activities. We have evidence that this transition is under way. In 2000, we had no income from Registration Agencies (as of course we had no formal RAs). In 2001, we budgeted (and achieved) 4 RAs, and RA fees produced 14% of the Foundation's income. In 2002, our budget is for a further 4 RAs and the current forecast is that RA fees will provide 22% of the Foundation's income. Of course, the timing of when this figure will reach 100% is dependent not only on numbers but on volume: to take a simplified example, if our fixed costs were $2M and each RA produced the current minimum of $50K, we'd obviously need 40 RAs to recoup costs totally: but in fact, our first RA (CrossRef) is already producing about twice that amount of income due to the large volume of DOIs they assign (i.e. "20 CrossRefs" would mean a balanced budget). Undoubtedly there is a chicken-and-egg, or threshold, effect, as the success of DOIs in one area encourages new applications.
The organizational model outlined here provides a clear basis for the relationship between end customers (registrants) and Registration Agencies; and between RAs and the Operating Federation.
There is however another set of relationships, which needs to be considered, between the various Registration Agencies themselves. In the Operating Federation model as implemented in e.g. EAN or ISBN, each RA has a geographical basis. Although customers are free to choose which RA to use, in practice most may go to one familiar to them, a local language agency. In the digital world, it is not clear whether such a basis is appropriate. In favour of such an arrangement is the need for language-specific related materials and support (e.g. local language guideline materials, helpdesk systems, and potential specialized consultancy staff). Arguing against such an arrangement is the fact that in a digital world, geographical barriers are less important, and an arrangement focussed on content sector or content type may be more effective. In some major markets (e.g. the English language markets) it could be possible that the intellectual property sector approach will be favored; whereas in smaller language markets, a geographical (or at least linguistic) basis may be more appropriate. Initial RA appointments made by IDF include examples of each.
The IDF Registration Agency Working Group has been set up to deal with issues such as this, and agree on common principles which foster a climate in which working as a registration agency is attractive, yet any long term monopoly is avoided. For further information on the work of the RAWG and more detailed documentation, please refer to the IDF. RAWG participation is limited to members of the IDF.
The DOI System is like any other complex technical and social construct. Attributes of reliability and predictability can only be delivered, particularly in an automated environment, if the DOI System operates in conformance with technical and procedural standards.
However, standards cannot be established once and then forgotten. Particularly in a rapidly moving environment like the Internet, standards need continuous attention to ensure that they meet the real requirements of the market place -- otherwise they fall into disrepute and are rapidly discarded.
This leads to a dilemma -- how can anyone implement a technology in the face of so much uncertainty and change. The solution adopted by the IDF is to build a flexible and extensible framework of standards; the framework itself can remain unchanging, while specific market- driven developments can be incorporated and managed by extending the framework. For example, new DOI-APs can be readily developed to meet specific community needs (see Chapter 5 Metadata and Appendix 2); and new data types can be added, to allow continuing flexibility in resolution (see Chapter 6 Resolution). There is an increasing community of interest in the DOI -- Registration Agencies, Registrants, users, and the members of IDF. Each of these groups needs to have a voice in the development of DOI technical and procedural standards, to ensure that they are genuinely market driven. However, there must ultimately be one organization that "holds the ring" and decides what should or should not be developed into a standard for the DOI System, a strong "force of convergence".
This "force of convergence" for the DOI System is provided by the IDF, as Maintenance Agency both for those aspects of the DOI that are put through external standardization procedures (the DOI Syntax, for example -- see Appendix 1) and for those aspects of the DOI System that are considered more appropriate for purely internal standardization. This Handbook is an exemplar of the role of the IDF in promulgating a common and consistent approach across all Registration Agencies and users.
IDF fulfils slightly different roles in relation to external standards and internal standards. With respect to external standards, the role of the IDF as Maintenance Agency is laid down by the regulations of the external standards body. IDF maintains formal and informal alliances and strategic relationships with a number of standards bodies and other organizations; for a current list refer to the DOI web site. With respect to internal DOI standards, the IDF acts as final arbiter.