Preserving digital scholarship 271. Kelly Russell, Ellis Weinberger, and Andy Stone The Cedars Project ALPSP 1999

Transcription

1 Preserving digital scholarship 271 Learned Publishing (1999)12, Introduction In his groundbreaking book Being Digital Nicholas Negroponte makes the statement: Computing is not about computers anymore. It is about living. 1 Although Negroponte s book is full of such aphorisms, there is a ring of truth to this bald statement that we ignore at our peril. As networked technologies have developed we have come to rely on them more and more often the extent to which we rely on technology is unknown even to ourselves. Nowhere is this more true than in research and scholarship where information technology has flourished due to the ease with which many scholarly resources adopt new media. Databases, electronic journals, and the hypertext links of the world wide web have become standard fare in academia. For the last decade research and scholarship has been frantically trying to grasp each new technological panacea. Scholarly publis hers have had to keep up with the growing trend toward networking and online publishing and the increasing demands made by their scholarly communities and, consequently, the marketplace itself. Likewise libraries, as information service providers, have had to deliver more (and more varied) information products to researchers in a variety of new and exciting ways. The necessary levels of urgent creativity and immediate resources to manage simply the production and delivery of new digital products is potentially vast. On top of which there is the necessary time (and additional resources) needed to preserve these materials in order to provide continuing access for scholars of the future. Continuing access to digital materials requires a long-term view of a short term marketplace to ensure preservation of digital products into the next millennium, the future is now. Preserving digital scholarship: the future is now Kelly Russell, Ellis Weinberger, and Andy Stone The Cedars Project ALPSP 1999 ABSTRACT: The Cedars project in the UK has been set up to explore some of the issues associated with archiving of digital materials and to make recommendations to libraries about preserving digital materials. Cedars, which is led by the Consortium of University Research Libraries (CURL), is funded as part of the Electronic Libraries Programme (elib) by the Joint Information Systems Committee of the Higher Education Funding Councils.

2 272 Article by Kelly Russell et al. we may at present be in a very precarious situation This article will explore some of the complex and pressing issues associated with preserving materials within the context of scholarship and research in the digital age. Based on the experience thus far in the Cedars project, 2 it will examine some of the technical issues associated with digital preservation as well as the (often more complex) issues associated with the management and organization of digital archives, such as accessibility, access control, and legal considerations. Background The long-term storage and preservation of scholarly materials has long been part of the remit of academic libraries. For print or more traditional resources, libraries and publishers knew and understood the role libraries played in providing access to these materials. A robust system of legal deposit ensured that copies of all publis hed material was deposited with specific deposit libraries for the purposes of posterity. However, with electronic materials, consideration of archival issues has hovered in the background or even been neglected entirely despite the fact that the sheer speed with which new technologies change and mutate means that their preservation is far more urgent than for their print predecessors. The dynamic nature of the technological marketplace demands a short-term view, which for the purposes of long-term preservation and access is more than a little problematic. The Cedars project in the UK has been funded as part of the JISC Electronic Libraries Programme to explore some of the issues associated with archiving of digital materials and to make re- commendations to libraries about preserving digital materials. In the UK, libraries have been involved in the development of the digital library for a number of years, and many of the early pilot projects led to partnerships with the publishing industry. Over recent years, libraries have been increasingly involved in digital library activities. Although much of the activity may have happened anyway in the fullness of time, there has also been an influx of new funding (both in the UK and abroad) for the creation and application of new technology in libraries. Most notably in the UK, the Electronic Libraries (elib) project has seen the establishment of over 70 different projects and studies in the UK in digital libraries (this work has involved over 100 different higher education institutions directly and a large number of commercial and non-commercial publishers). Although many libraries are engaged in their own preservation activities (both for old and new material), in a largely print environment, libraries relied on a system of legal deposit to ensure that copies of everything published in the UK were available in perpetuity. The British Library, the national libraries of Wales and Scotland, the universities of Oxford and Cambridge, and Trinity College Dublin take deposit of copyright materials and ensure preservation and longterm access to material. In fact, sometimes publishers also rely on these copyright libraries to ensure access to material they have discarded because it is no longer of commercial interest for them. In the UK there is currently no legislation for deposit of copyright digital material. Although many publishers are currently keeping archives of their digital material, most would admit that their interest in such archives is inextricably (and understandably) linked to its commercial potential, which is generally considered to last for no more than 10 years (depending on the subject area). The British Library is working closely with publishers and other interested parties to draw up new legislation for the deposit of non-print material which will hopefully be implemented in the future. Deposit in this sense would address the long-term archiving and preservation of materials. Nevertheless we may at present be in a very precarious situation. With no safety net legislation for retaining digital content and the current lack of understanding (or even acknowledgement) of the issues surrounding long-term preservation of digital material we are in danger of losing a substantial part of our scholarly heritage. Without some precautionary steps now, the 1980s and 1990s may well represent an information black hole for scholars of the future.

3 d d d Preserving digital scholarship 273 What is digital preservation? Digital preservation, for the purposes of this paper, is defined simply as the storage, maintenance, and accessibility of digital material over the long term. Digital material refers to any material renderable by a computer and includes both that which is digitized (reformatted to digital) as well as those resources that are born digital. Long term in this context should be taken to mean long enough to be concerned with the impact of changing technologies a timescale of decades or even centuries. There has been a great deal of confusion around the term digital preservation mainly because of early projects which (perhaps inadvertently) equated the process of digitization with preservation. 3 In general digital preservation involves a number of organized tasks associated with a variety of technical approaches or strategies for ensuring that digital resources are not only stored but also adequately maintained and consistently usable over time. It is also worth clarifying that accessibility in this context means that access is technically possible and not that access is legally permitted these are two different issues and should not to be confused. As will be made clear, although digital preservation ensures the material is kept technically accessible, preservation of digital materials is not simply a technical issue; there are legal, economic, and organizational factors to consider. Indeed, despite the new media, many traditional skills and knowledge need to be applied, e.g. the selection and evaluation of material for archiving or copyright and access management. Digital resources represent a continuum 4 One of the distinguishing features of digital objects, as opposed to print resources, is that their lifespan represents a cycle or continuum. Where the life of a book from creation through to preservation represents a chronology, digital materials are better understood as a continuum. Digital materials often represent complex interdependencies between the different stages of a resource s life from creation, through acquisition, management, use, and finally preservation. Decisions taken at the time of creation will influence what options will be necessary for preserving that object into the future. Likewise at the time of preservation, decisions about how the resource will be preserved will impact on how it can be used in future. Although this may also be true to some extent for non-digital or analogue resources, for digital materials these interdependencies are critical. Effective preservation of digital materials will only be guaranteed by a clear understanding of the checks and balances resulting from different actions and decisions taken about a resource throughout its lifecycle these often involve different stakeholders, motivated by different interests, at different times. Let us take an electronic journal as a simple example: A commercial publisher creates the electronic journal (available as HTML on the web) for the purposes of generating revenue for as long as the market demands the material. The publisher is therefore (and often justifiably) not interested in preservation of the resource over the long term a perspective which might give rise to decisions at the time of creation which actually create difficulties later on when the object is to be preserved. A library comes later on in the information chain and is interested in using the resource and possibly in maintaining access to it over a number of years. The library has no control over decisions made by the publis her when the resource was created about what technology was employed, etc., but it must deal with those decisions when it comes to the preservation of the material (e.g. should the hypertext links be preserved?). Scholars of the future will be interested in using the journal (we cannot predict how or why) and will have had no control over how the resource was preserved despite the fact that the preservation strategy adopted by the library will have implications for the use the scholar can make of the digital object in the future (e.g. hypertext links are no longer valid). This is a simple example to illustrate a complex web of cause and effect decisions digital materials are better understood as a continuum

4 274 Article by Kelly Russell et al. What is meant by a digital preservation strategy? but one that aims to illustrate why digital resources present great challenges to publishers, libraries, and users alike. Digital preservation strategies As suggested above, digital preservation involves a number of organized tasks to ensure it is effective. The fluid nature of new technology demands a robust technical approach or strategy. What is meant by a digital preservation strategy? Put simply, a digital preservation strategy is a particular technical approach to the preservation of digital materials. Broadly speaking there are three main technical approaches to preserving digital materials. Technology preservation, technology emulation, and data migration. The first two focus on the technology itself. In order to preserve the functionality of any digital resource there must be preservation in some form, of the technical environment which created and ran it. Data migration strategies focus on the need to maintain the digital files in a format which is accessible using current technology and require regular migration from one technical environment to a newer one. Each strategy will be described briefly in turn but it is important to realize that even within each approach there are variations. Technology preservation If digital material relies on the technical environment used to create it in order to preserve the functionality and look and feel of the product, then the most obvious approach is to preserve the original technology. This is a museum style approach and probably only suitable as a short-term solution. Hardware and software from the object itself are maintained so that access can be guaranteed. However, this generally means that access is limited to a specific physical location (i.e. where the hardware/ software are kept) and the cost/space implications for storing this kit are probably beyond the realm of possibility on a large scale. Over a number of years the machines themselves will inevitably degrade, making this approach problematic as a long-term strategy. The Science Museum and the Computer Conservation Society in the UK are interested in the merits of this approach and are currently maintaining old computer systems which may prove valuable resources for scholars in future. For a library or publisher the sheer space and resources needed to maintain old systems would in all likelihood make this approach impossible. Technology emulation Another approach which also bases itself on the need to preserve the technological environment (and therefore original functionality) is emulation. Unlike the strategy described above, an emulation strategy seeks to preserve that environment not through the preservation of original hardware/ software but by using current technology to mimic the original environment. This might involve emulation of the original software or (more likely) emulation of the original hardware (in this case the original software and operating system are stored along with the digital object itself). Either way, the strategy relies on a detailed description of the original environment on which to base the emulation in future. The emulator itself is not necessarily stored in the archive (although it may be) it may be created at a later date when there is demand for the material. The detailed technical description (metadata) on which this strategy is often based is a key component of an emulation strategy and, as yet, there are no standard approaches available for descriptions of this kind (see below). Although controversial, many experts are beginning to believe that for truly long-term preservation emulation is the best solution. It accepts the necessary conundrum of preserving the original technical environment but ensures that material is not held hostage to obsolete technology. Instead it can take advantage of new technologies as they develop for emulation. Although we are unable to predict how future technologies will develop, we can be certain of some general principles they will be more effective, cheaper, and faster. Data migration Unlike the strategies above, data migration focuses on maintaining digital material in

5 d d Preserving digital scholarship 275 current formats. At present many libraries and publishers are involved in regular migrations for image files, e.g. moving images from one software version to a newer version. The attraction of this strategy is that material is maintained in an accessible format. The two strategies above both advocate storing the material as a bytestream in its original format and then making it accessible when necessary. Data migration means the material is maintained in the archive in a currently useable format. For a library this is an important advantage. There are significant disadvantages to this strategy. Firstly, migration can be complex and time consuming; and, secondly, the frequency with which it would need to be done (in accordance with fluctuations in the volatile technology marketplace) is unpredictable. Furthermore migration strategies must also assume some (however inconsequential) loss of data. When a resource is moved into a new technical environment (e.g. even simply moving from one version of Microsoft PowerPoint to another) some of the original data will be lost in the transition. Sometimes the loss is identifiable and made very clear, and worryingly sometimes it is not. It is this potentially unknown data loss that makes migration somewhat problematic. It is important to stress that data migration as it is described here is more complex than what is often called data refreshing. All preservation strategies must include regular data refreshing, which is the systematic transfer of stored material to newer and fresher media (e.g. from one magnetic tape to another). Refreshing does not imply ensuring the material is kept useable it is only the transfer of a bytestream from one medium to another. Migration focuses on keeping the material functional with new technology. The Cedars project is currently beginning to compare and test different strategies for preserving digital materials. As part of a new International Digital Libraries Programme funded jointly by the National Science Foundation in the US and the Joint Information Systems Committee in the UK, Cedars has also been granted funding for a project in emulation. This project is scheduled to begin in autumn 1999 and will run for three years. Describing digital resources to ensure long-term preservation Managing information resources (whether digital or not) requires a comprehensive system of documentation and description: digital preservation and digital archives are no exception. As is implicit in the technical issues described above (particularly in the case of emulation), for a digital preservation strategy to work, digital objects must include a great deal of documentation which will need to be accessed, queried, and updated regularly. The descriptive information necessary to achieve this is referred to as metadata (or data about data ). One of the aims of the Cedars project is to develop a metadata set which can effectively be used in digital preservation, thereby contributing to the development of new standards for this type of data description. The Cedars project work focuses on a number of practical demonstrator projects based at the universities of Oxford, Cambridge, and Leeds which will all implement an agreed model 5 or architecture for managing a distributed digital archive. The Cedars model identified the following broad objectives: Ensuring that one can find the preserved digital object (resource discovery). Ensuring that one can successfully access relevant aspects of this object when it has been found (retrieval). The metadata necessary to archive digital material successfully must support a wide range of archival functions from resource discovery through to access control, document integrity, and, finally, preservation itself. One of the main challenges that the Cedars project has faced in relation to metadata is developing a complete set of metadata elements which reflect the multiple needs and functions that have to be addressed. These functions include acquisition, storage, searching, access management, delivery, and preservation. Although a great deal of work has been done on metadata and metadata necessary to archive digital material must support a wide range of archival functions

6 d d 276 Article by Kelly Russell et al. For digital materials, there is no preservation without retrieval developing schemes for describing resources, existing metadata schema do not adequately cover all aspects of a digital archive. For example, most schema focus on resource discovery (searching and retrieval) or were developed for a specific type of digital object such as digitized images. In addition to having a complete set of metadata, an appropriate framework within which to implement it will also be critical particularly if material is to be stored across a number of distributed archives. A number of formalisms already exist which could be used within the Cedars project to express the metadata element set. The project is currently monitoring the development of XML (extensible Mark-up Language) and investigating the use of this emerging standard for implementing metadata, as it is non-proprietary and flexible. For the sake of clarity the discussion below will focus on the following functions within a working archive: searching, retrieval or delivery, rights or access management, and preservation. Metadata for searching Being able to find the object in question within a digital archive is a fundamental requirement. Most materials include some description and documentation which is intended to facilitate search and retrieval. It may be possible to use some of this existing metadata to assist the generation of metadata for archiving purposes; indeed, it may be possible to develop crosswalks between different metadata standards so that mappings can be automated (e.g. an SGML header into a MARC record). By preserving the original metadata, we can avoid the pitfalls associated with converting pre-existing metadata schema into another format where richness of description is inevitably compromised. It is acknowledged that such crosswalks involve the loss of some elements: one schema may require deeper detail of certain attributes than another; alternatively, the rules, structuring, and formatting of certain fields may differ, which can result in difficulty in seamless two-way mapping. Metadata for retrieval and delivery Successful archiving will also rely on a detailed description of the technology needed to retrieve the object, i.e. to render it and make it usable. For digital materials, there is no preservation without retrieval. When material is simply stored as ones and zeros retrieval is a critical component of preservation. For example for TIFF images or PDF files the system will need to record and describe the specifications for specific software, operating system, and perhaps even hardware too. This type of information is critical and as yet very little work has been done to develop standards in this area. Cedars is particularly interested in developing a standard approach for such detailed technical descriptions. Metadata for rights management There is a multiplicity of rights issues raised by digital preservation. These include: Appropriate agreements for placing the digital object in a long-term storage repository as this will inevitably involve moving (and therefore migration) of the data from its original physical media. Publisher copyright statements as well as identifying the copyright holders (who may need to be approached for extra permissions), date information will be useful over the long term to identify when a work is no longer in copyright. Any organization taking deposit of materials for archiving will need to be able to assure depositors that access will be adequately managed to ensure the interests of copyright holders are protected. For digital materials this takes place on a number of different levels. Those owning rights to the intellectual content of the material will need to know that their material is secure and can only be accessed by authorized individuals. They will also need to be sure the material they deposit is not tampered with, i.e. its integrity is guaranteed. Digital materials also involve a number of other third party rights holders, such as the software and systems providers, who will need to agree to any technical strategies

7 Preserving digital scholarship 277 which will be applied to their materials which may underpin the content itself. The extent to which this latter set of rights will be fully understood when materials are accepted for archiving, particularly for complex digital materials, remains unclear. This is discussed further below. Despite the obvious complexity, digital material will rely on effective and standard descriptions or documentation providing details of the associated intellectual property rights. It is in this area that publishers and creators of digital material may prove helpful. Sometimes this information can be easily included with a resource at the time of creation, where chasing it down or recreating it when the material is deposited for archiving would be difficult and time consuming. The Cedars project is currently working with publishers and rights holders to try and understand how data creators could assist in the provision of adequate descrip- tions and documentation for resources at the time of creation. Metadata for preservation A functioning digital archive will also need to ensure that the process of preservation is both guaranteed and adequately described. Depositors will need assurance that the material they deposit in the archive will remain authentic and that it will be preserved through changing technology. Metadata will be necessary to describe the objects authenticity (such as its provenance ) in archival terms, its context and relationship to other objects, and any technical processes that have been applied for preservation purposes (e.g. migration from one format to another). As mentioned above, issues such as document integrity or fixity are critical and archives will need to provide a mechanism for detecting any unauthorized changes to a digital object. Such changes may not necessarily be malicious; for example, one can envisage additional work being performed on OCR text to improve its accuracy although it is expected that such work will be documented and described in detail. Although technical considerations and data description standards for preserving digital materials are important and challenging in themselves, there are a whole host of other non-technical issues which will need to be addressed. These include collection management issues for libraries, such as selection of resources for preservation; integration of digital preservation into existing policy and practice; legal considerations; control for copyright materials; co-operation between organizations and possible sharing of responsibility; and finally (but not least!) the costs. Collection management issues Digital preservation throws up a number of issues related to managing digital collections. For example, decisions about what material to select for long-term preservation continues to be an issue with which libraries grapple. Since digital objects do not in most senses care for themselves as well as books do, greater resources will need to be devoted to their care. Because maintenance of digital objects will be expensive, policies for acquiring digital objects must become more rigorous than policies have traditionally been for acquiring books. The technical complexity and ongoing maintenance (as illustrated above) will mean that it is not possible to store everything. More to the point, the speed with which digital resources are born will mean storage of everything is not desirable: filtering digital collections to select material specifically for long-term preservation will be necessary. The digital objects stored in a library must contribute to the subject strengths of the library. Due to the complex interdependencies described above to do with the lifecycle of a digital object, preservation must be considered at the time of acquisition. In a print environment, a decision to acquire can be an implicit decision to preserve. However, for non-digital materials libraries generally have the luxury of making a separate preservation decision later. For digital materials acquisition and preservation will need to be more closely linked because if a preservation decision is not made at the time of acquisition, it may be too late to consider preserving it later on. There are risks associated with making acquisition decisions based solely on the desire for immediate access to the material material to select for long-term preservation an issue with which libraries grapple

8 278 Article by Kelly Russell et al. Will libraries be able to acquire and retain staff with this skills set and knowledge? rather than also considering the long-term view. Decisions about format for deposit and arrangements for licensing material will need to consider preservation issues at the time the material is acquired. For example, if a library agrees to a licence which is based on access only, it may find that when the license expires, there is little, if any, scope for preserving the material to provide long-term access for scholars of the future; the format agreed under the terms of the licence may not be appropriate for long-term preservation or, more seriously, access to the material is no longer possible (either technically or legally). A preoccupation with immediate access arrangements for digital resources without taking long-term access into account could seriously jeopardize a library s ability to preserve digital material. Arrangements for archiving material of commercial interest will need to be covered by existing legal agreements to ensure that in the long run preservation is guaranteed. At present many licences do not adequately cover longer-term access to materials. The purpose of the library and the needs of its users must be clearly defined before acquisition/preservation decisions can be made. The digital objects chosen should be approved by subject specialists who understand the scholarly value of the material as well as the costs involved in storing these objects. The materials should also be analysed by experts in the field of digital preservation, to ensure that preserving these objects is possible without the loss of critical content or functionality. In general, thinking of digital objects as manuscript collections with perpetual conservation needs is a useful perspective. Library policy issues If one point is made clear by initial work in digital preservation it is this: the library will need to dedicate substantial resources to ensure preservation of digital objects for posterity. The objects will need to be tested regularly for accessibility and file integrity, and migrated or refreshed to new hardware platforms whenever necessary. In 1998 the Research Libraries Group (RLG) commissioned a study, Digital preservation needs and requirements in RLG member institutions, 6 which suggested that libraries expected that over the next few years they would need people who have the skills and expertise to make informed decisions and perform the necessary technical tasks for preserving digital material. Outside bodies, such as organizations specializing in data transfer or migration, may be able to assist the library, but the results of the study suggest that even for this type of outsourcing, informed library staff would be needed to make effective use of these external organizations. Will libraries be able to acquire and retain staff with this skills set and knowledge? Unfortunately, as technical projects in libraries have already proven, the expert personnel needed to carry out technical work such as digital preservation will tend to have skills of value to commercial bodies, and hiring and retaining them may be problematic. Libraries must ensure that digital preservation does not reduce resources for traditional library activities. For the foreseeable future, scholars will need printed works as well as digital material. Digital preservation will need to sit alongside such existing library functions as preservation and conservation of print and manuscript material, and acquisitions and cataloguing. At present a substantial portion of digital library work is funded as project work, which, while it does provide a valuable oppor tunity for exploration and experimentation, makes integrating new areas of work into current practice more difficult. Existing policies, such as those for selection, acquisition, and preservation of traditional materials, will need to expand to include digital components. Intellectual property rights and access control The digital environment has given rise to a number of new and challenging issues for managing intellectual property rights. Although many of the rights issues libraries face for digital material are similar to the issues the library is already used to dealing with, there are increasingly issues which arise specifically for digital objects. Digital resources are both easier to copy and easier

9 Preserving digital scholarship 279 to change than traditional library materials. Fair dealing and the integrity of the files will have to be carefully monitored. The library must satisfy the technical security demands of the copyright owners. The computers and network which store the digital objects will need to be proven to be secure. Secure system and network tools are readily available, and the expertise to secure a network is becoming more common. Secure remote access can be achieved by various means, and the academic community can offer practical advice to libraries. The spread of electronic commerce may ease the fears of copyright owners about these issues. Creators of digital objects need to feel comfortable that their investment will not be lost due to digital preservation of these objects. Publishers may worry that allowing a library to preserve a digital object will cost them a sale, by giving free access to those few libraries who would otherwise be paying for access to the object. The creation of digital objects requires significantly more investment than books and often these digital resources are created for particular specialist scholars with networking in mind (i.e. a single copy will be purchased on behalf of an entire institution or group of institutions). For such reasons, expected sales in the UK may only be a dozen copies. The library will need to have the legal expertise to negotiate with publishers and other owners of copyright. The terms and conditions which the publishers demand in order to allow preservation and access will have to be carefully analysed. These terms have to be worded so as to facilitate controlled access as well as allow for detailed technical examination and manipulation of the digital objects sufficient to enable all digital preservation activities. Digital materials are frequently built using a variety of software and hardware much of which will be copyright to another organization or third party. Libraries and publishers will need to determine whether the publisher has the right to permit the library to store the software and the content comprising the digital object, or whether the rights belong to a third party. Although licence agree- ments may allow backup copies to be made, as has been illustrated above, backup copies are not preservation. Preservation strategies (e.g. migration) are often a set of specific technical tasks which are more involved than simply copying. When the publishers receive permission to use software to create a digital object, the right to preserve or archive that software may not be implicit in that agreement, and an archive or library may need to seek separate permissions if the material is to be preserved. Costs As yet the costs of preservation are unknown but we assume very high. Unlike print materials, nothing in a digital format will be preserved by accident. Where an old manuscript can be discovered 200 years later up in an attic, opened, and read, digital files will require vigilance and vigilance will mean time, resources, and therefore costs. Figures are available for the costs of storing data (usually given in pounds per megabyte) but very little is known about the costs of maintaining a digital archive where the material can be accessed over the long term. Based on demonstrator projects, Cedars plans to produce some cost models for digital preservation later in the project. The Library and Information Commission has recently funded a study which will be done over the next year by the Arts and Humanities Data Service to assess the cost/benefits associated with digital preservation. Conclusions The preservation of and long-term access to digital material will be an area of concern for libraries and other organizations involved in the preservation of our scholarly and cultural heritage well into the next millennium. Currently we are only at the base of what will be a significant learning curve for libraries, publishers, and scholars, for whom this will be an increasing concern. Although the technical challenges are great, a whole host of other non-technical issues will need to be addressed regarding the management of digital collections and the legal framework in which preservation and long-term access must be situated. This involves a nothing in a digital format will be preserved by accident

10 280 Article by Kelly Russell et al. number of different stakeholders with sometimes vastly different interests and agenda. Currently there is a need for preliminary work to inform the development of policies and practices in libraries, and the Cedars project will produce guidance in this area for libraries. However, despite the value of short-term pilot projects, digital preservation is still an issue which will need to be fully integrated into existing collection management policies and access agreements. It is an issue which must now be addressed head on by all stakeholders involved in the creation, management, and accessibility of digital resources. Unless there is concerted action now to ensure the continuity of this material the future of scholarship will be jeopardized for preservation in a digital age, the future is now. References 1. Negroponte, Nicholas. Being Digital. London: Hodder and Stoughton, Cedars stands for CURL exemplars in digital archives (CURL: Consortium of University Research Libraries). 3. Early digitization projects were interested in exploring the preservation possibilities of using digital imaging to preserve rare or fragile materials by providing a means of alternative access. This is not what is meant in this instance by digital preservation. 4. The concept of a resource s lifecycle is frequently alluded to in the archives community. A more indepth analysis of this useful concept has been provided by the Arts and Humanities Data Service in a recent study entitled A strategic framework for managing and preserving digital collections, manage/framework.htm. 5. This model will be based on the Open Archival Information System (OAIS) developed by the International Standards Organization and the Committee on Space Data Systems: nasa.gov/nost/isoas/ref_model.html. 6. Hedstrom, M., Montgomery, S. Digital preservation needs and requirements in RLG member institutions, Research Libraries Group, preserv/digpres.html. Kelly Russell, Ellis Weinberger, and Andy Stone The Cedars Project Edward Boyle Library The University of Leeds Leeds LS2 9JT The URL, which is a pointer to the regularly updated Cedars bibliography, might prove a useful reference for readers wishing to follow up these topics. leeds.ac.uk/cedars/documents/cis05.htm RCP ad 2