What is the relationship between the hybrid library and the DNER (the Distributed National Electronic Resource)? This paper discusses that question and suggests a number of ways in which DNER strategy and thinking can be informed by hybrid library developments. ‘Suggests’ is the word, since there is currently an investigation underway that is dealing with this question which is still to report. This is being coordinated by Stephen Pinfield, one of the authors of this article. This article does not discuss in detail the progress and outcomes of the hybrid library projects – there is quite a lot published on the projects, and there are several companion articles in this issue of Ariadne. The hybrid library projects supported by eLib have also yet to be completed and formally evaluated. It is only when these activities are finished that we will be able to answer the question more confidently. In the meantime, a number of issues can still be discussed. First, a description of the DNER.
What is the DNER?
The DNER (Distributed National Electronic Resource) is a managed information environment which provides secure and convenient access to a range of information services and resources. As such, ‘DNER’ is a generic term used to describe the wide range of information-related activities and services supported by the Joint Information Systems Committee (JISC).(1) The DNER concept has been formulated to enable a coherent strategy to be developed for these varying activities and services.
Broadly, the DNER can be said to have two main components:
- Content: the collections of information resources themselves, selected for quality and relevance
- Environment: the online space within which these collections are accessed which is managed to allow a coherent user experience, and which interworks with other resources and initiatives.
These two main components are supported by other important elements: advisory services (such as those providing technical advice), preservation services (for digital preservation) and development programmes. In this article we do not cover this full range, but focus on collections and the information environment. Readers interested in the full range of DNER activity are referred to the DNER web site.
The draft Communications Strategy of the DNER sets three long term objectives for the DNER:
- To provide the world’s high quality information to staff and students in higher and further education at any time and anywhere.
- To be the leading innovator in UK education in the field of digital information provision of information and take an influential role in developing the UK’s learning and research agenda.
- To take a leading role in stimulating development both here and overseas.
The first of these looks not only at making high quality information resources available but also doing so in a way that fully exploits the available technology (distributing content as widely as possible). It is not just about collecting content but also putting technical structures and licensing arrangements in place to allow access “any time and anywhere”. The second objective sees national information providers as innovators taking a proactive role in learning and teaching and research in a way which improves the quality of these activities. The third objective emphasises the importance of ongoing ‘development’. ‘Development’ in the sense of technological development; and also ‘development’ in the sense of the widening remit of the DNER to include a broader range of content (for example, by encouraging institutions to locate their own quality content within or adjacent to the DNER).
Like all long-term objectives, these are ambitious. But they are an indicator of the direction in which the DNER is going and also the importance attached by JISC to DNER development.
Finally, it should be clear that these objectives will not be achieved in isolation of the wider information and user community. The JISC needs to work with, and through, individual institutions. It also needs to work with others to ensure that approaches taken are not idiosyncractic or rapidly superseded. In particular, the ‘information environment’ and the preservation agenda need to be a collaborative activity. For this reason, the JISC is actively implementing collaborative arrangements with other national information and learning initiatives and organisations, and with international bodies such as the National Science Foundation in the US.
Why the DNER?
There are a number of factors which came together and made the creation of the DNER an appropriate strategy:
1. Incremental development
JISC has been supporting the delivery of information resources for some time and this has been a valued activity. However collections and services have grown incrementally and as a result of opportunistic acquisitions. They have been delivered from different places in different ways with different support structures. As collections have grown, a more sustainable approach is required. The DNER is an opportunity to co-ordinate these developments and give them strategic direction.
2. The need for more consultation and market research
Many activities and acquisitions in the past have been determined by the centre. A consciousness has grown that there is a need for more effective consultation and market research mechanisms to be in place so that the needs of Higher and Further Education Institutions (HEIs/FEIs) can be assessed more systematically.
3. The incoherent ‘information brandscape’
Instead of information landscape what we have at the moment is more like an ‘information brandscape’ (a term coined by Lorcan Dempsey to describe an Internet environment where multiple individual websites compete for the user’s attention, encouraged by the proliferation of ‘portals’ and ‘one-stop-shops’). Currently the user’s experience of JISC services is not a coherent one. Rather, they are presented with a series of branded products and hosts all competing for brand recognition and offering themselves as individual opportunities. The DNER is an opportunity to begin to find ways to address this issue to support a more coherent user experience.
4. The need for systemic coherence
The development of the DNER involves the design and further development of a system of provision, not of a single service. Services need to cohere within that system, to work together to deliver the bigger goal. This means that roles need to be clear. This is at technical, service and policy levels. This is not just an internal DNER issue. The DNER needs to work with JISC’s middleware services for example, and also with wider development trends.
5. The need to engage with learning and teaching
There is a perception that much of JISC activity up until now has been geared up for research rather than for learning and teaching. The DNER will evolve to ensure there is a balance of provision, and that resources are more actively used in learning. The DNER team will work closely with the LTSN , BECTa and others to ensure that resources support learning and teaching in sensible ways. It is important to remember that JISC services support both Further and Higher Education, and must develop accordingly.
One of the interesting things about the factors outlined above is that many of them are also issues that the eLib hybrid library projects were set up to address, albeit on a more local level. The projects have been engaging with some of these issues for a while and therefore have some useful lessons to pass on. This applies in particular to the aspect of the DNER as an environment – after all, one of the aspirations of the hybrid libraries has been to provide a local ‘information environment’ which brings together resources and services in helpful ways.
The DNER environment
One way of thinking about resources within a DNER context which has been found useful is as follows;
The DNER coordinates informational activity in three broad areas:
- The creation of a strategic national resource of educational and learning materials: this consists of collections of material made available on a national level via networked systems designed to deliver them to the end user. These may be acquired through licensing arrangements, or by funding their creation. These include a range of bibliographic, research, image and other data, typically made available through JISC services such as the data centres (see http://www.jisc.ac.uk/dner/collections/ for further information about content collections). This activity is motivated by the recognition that there is a wide range of resources where the DNER can add value through such central activity, whether in terms of achieving favourable acquisition deals, achieving economies of scale in hosting and delivery, or eliminating unhelpful redundancy across institutions. Such activity complements institutional provision.
- The creation of a framework for Community resources: institutions and individuals within institutions also manage or create resources that may be articulated as part of a wider ‘community’ resource. The DNER provides services which address this requirement, adding value to institutional or individual resources by placing them alongside other resources in some wider framework of provision. Services which assist here may be of several types. Some services assist in the discovery of community resources (COPAC and the Archives Hub, for example, which point at collections within institutions). Some services manage materials deposited with them by researchers and others (the Arts and Humanities Data Service or the Data Archive for example). It is anticipated that this area will grow in importance as the value of accessing resources created by researchers, teachers and institution within a consistent framework becomes clearer (consider for example, image databases, e-print archives, learning materials, prospectuses, digitised special collections, and so on).
- The creation of a resource discovery framework for Aa global resource: The DNER provides services (for example the Resource Discovery Network) which provides access to quality controlled resources on the public Internet. It also supports the acquisition of other discovery services, abstracting and indexing services, for example.
In addition, the DNER recognizes that its services need to be visible and accessible alongside and with other resources. For example:
- Institutional resources: resources licensed, purchased or created by institutions and provided through their own service infrastructures. These might include e-prints, bibliographic resources, research data, or learning materials. In some cases, as mentioned above, these may be available for users outside their institution; in others, not.
- Personal resources: Increasingly, users have an expectation to be able to personalise their information spaces, and may have local collections (bookmarks, other tools). Users are also increasingly creators of resources, and may want support to ‘publish’ or ‘deposit’ them in sensible ways.
From an end user perspective, it is not important precisely where resources or services are located in this model. The end user wants to be able to see resources wherever they are located. The challenge associated with the DNER is then to weave rich information resources of all types into the fabric of people’s working and learning lives – to make them accessible when, where and how it is most most useful. This is a challenge which is jointly faced by institutions and by the DNER, and underlines the need to achieve useful complementarities and ways of working together.
The DNER aims to become a managed strategic resource which complements high quality institutional and community resources. It flows from this that the DNER strategy should not only be including acquisitions at the centre but also encouraging and supporting institutions to disclose their own information resources in an organised and effective way.
The collections themselves consist of a wide range of resource types: scholarly journals, monographs, textbooks, abstracts, manuscripts, maps, music scores, still images, geospatial images, vector and numeric data, moving picture collections and sound collections. These collections need to be acquired as part of a coherent collection development policy which serves community needs. The important question yet to be answered by the DNER is how is this wide range of resources to be brought together in a coherent whole? How can they be used in an integrated way?
This underlines the need to further develop the second main DNER component: the information environment. An information environment might be described as a set of network information services which support secure and convenient access to distributed collections. Hybrid library goals, as much as DNER goals, benefit from a shared information environment: it is not sensible for each institution to individually address all the technical and service challenges of digital information environments.
DNER architecture and hybrid libraries
This question of coherence and integration can begin to be addressed by mapping out a technical architecture for the DNER. This is already starting to happen and will be done more formally as part of a technical review taking place at present, and due to report in early 2000. In broad terms, the architecture is being seen in terms of four broad areas:
- Presentation: Most users will access DNER (and other) services through a web-based ‘front-door’. Currently most individual resources provide their own front door. The plan in the DNER is to look at providing a ‘presentation’ layer, which provides simple web-based collections of links to resources . This will develop in various ways in due course, building on fusion and other services.
- Content/service delivery: this is the content itself which currently sits on the network and is accessible in different ways. The predominant approach is plain old HTTP, but this limits how the services can be used. They are available to human users, throwing away structure. The aspiration here is to make more services available through structured protocols such as Z39.50, LDAP, Open Archives Initiative. In that way, their data is more directly accessible and manipulable for others to use.This aspiration is broadly in line with the vision of the ‘semantic web’ espoused by the World Wide Web Consortium.
- Fusion services: these services will use various technologies to fuse content in different ways so that it can presented and re-presented to different user groups in different ways. The availability of content through standard protocols will facilitate the work of the fusion services. JISC is funding a number of services that will begin to achieve this, as an exploratory and exemplary process – leading to an examination of issues and a sharing of lessons. The RDN will develop fusion services in the engineering, medical and social science areas. These will provide services which cross-search resources in their area, which provide links between services (document search and document delivery, for example), which explore generalisable customization, and so on. The hope is that individual institutions will also produce fusion services – and, indeed, one way of looking at the hybrid library projects is as fusion services.
- Middleware: in JISC terminology, ‘middleware’ is used in a specialised way to refer to security and authentication services. This category might be expanded to include other services. The aim here is to split out common tasks into infrastructure where it makes sense to do so. Authentication has been handled with benefit in this way, and other areas which will be explored include authorization, user profiling, and collection description directory services.
This basic architecture is in many ways similar to that developed by eLib hybrid library projects. These projects have also investigated many of the processes and technologies that might be involved in creating a DNER technical architecture. Hybrid library projects have investigated issues such as authentication and middleware, presentation, fusion technologies, content creation, and personalisation. The projects therefore have important lessons to offer in the technical review of the DNER.
Many DNER activities may be seen as natural supports of hybrid library project activity. For example, a number of hybrid library projects have implemented some kind of cross-searching facility. Agora uses Z39.50, MALIBU uses HTTP. An explanation and demonstrator of the latter has been provided by BUILDER. One thing that has become apparent to projects is that effective provision of cross-searching services requires dialogue and co-operation between the data provider and library service. There have to be agreed ways of presenting the data which ensures that it remains accessible to users. Many of the projects have found that multiple individual approaches to data providers, each with slightly different questions, is not a sustainable or sensible long-term approach. It seems more sensible for this to be done on a national level. This is a role the DNER might play. Data providers are far more likely to co-operate with a single national service than an array of local services. The DNER can where necessary use the collective market power of further and higher education institutions to encourage data providers to make their information available using rich protocols, such as Z39.50.
Other technologies being used by the hybrid library projects may be of use in the DNER. One interesting point here is that there are also a number of commercial products now (coming) on the market which seem to offer hybrid-library-like functionality. These include products such as VDX (produced by Fretwell-Downing), MetaLib and SFX (produced by Ex Libris) and WebExpress (produced by OCLC). These products are not the same of course but they each offer possible ways of achieving aspirations of the hybrid library. It is possible that the DNER may want to make use of these products as part of its fusion services as an alternative to developing in-house solutions. Some local institutions are certainly looking to them to support service development. The question of the extent to which commercial providers should be used in developing the DNER infrastructure has yet to be decided. There are models of the large-scale services being developed in partnership with a commercial company. For instance, the DigiBib service developed in North Rhine Westphalia in Germany was developed by university library services alongside a company called Axion. The service provides a tightly co-ordinated state-wide digital library using a variety of protocols and includes a personalised billing service for document delivery.(2) DigiBib is of course a very different animal from the DNER but its success in working with a commercial provider is an interesting model.
National and local needs
There are then a number of important correspondences between the concept of hybrid library and the DNER. This raises an interesting question. Is the DNER a national hybrid library? The answer to this question depends to a large extent on the definition of hybrid library used (and there are several). Some people use the term hybrid library to describe the general library service which combines access to a range of media, printed and electronic. Others use the term more specifically to describe the integration of electronic services into a more coherent whole. Using the latter definition, the DNER in many respects fits the bill.
The most important difference between the DNER and hybrid library projects is the most obvious one. Whereas the hybrid library projects are based in individual institutions, the DNER is a national service. There are a number of major advantages to running a national service some of which have already been suggested. The national service can use its market power, there is also economy of effort and opportunity to concentrate expertise. But one thing is clear, a national service is not of itself enough. The national service cannot replace the provision in institutions – and nor is it meant to. Institutional portals and landscapes are still required.
Almost all of the end users of the DNER are based in institutions. It is institutions who buy and provide access to services (even if they do so through national agencies). Information resources available to a given user at any one institution made available through the DNER are only a proportion of the totality of resources available to that user. Institutions will also provide local networked resources (such as CD-ROMs) or may purchase access to web resources not part of the DNER. These will differ from institution to institution. Institutions will want to guide their users to the full range of resources available to them not just those provided as part of the DNER. Of course, the DNER can encourage institutions to co-locate locally produced data with the DNER but there will still be room for portals or hybrid library implementations at a local level.
This is an opportunity for the DNER. The DNER is not just yet another separate service provider but should also become a service enabler for HE and FE. It should be set up in such a way as to enable institutions to better provide nationally-supported resources in a more integrated way to its user local communities. Various approaches might be taken to this, here is an example of one. The Resource Discovery Network (RDN) services, such as SOSIG, BIOME and EEVL, are already available through a single interface. More recently this facility has been enabled so that it can be delivered through an institution’s own interface via a service known as RDN-i.(3) Users can then search nationally supported datasets through their own local interface. This is an example of how a national service can be an enabler – allowing local providers to more easily deliver information to their users. A similar approach might be used for other national services. For example, ATHENS (or its successor service) might be provided in such a way that local institutions can make use of it more easily for local needs. This would create the potential for a more unified local and national provision.
This leads again into the notion of an ‘information environment’ which provides a range of shared services (middleware, content, resource discovery, etc) which can be assembled to meet local needs. We hope that the DNER will enable more integrated use of information resources across HE and FE. It will do this by providing some integrating services itself. But, importantly, it will also work with providers and others to provide services so that they can be more easily integrated into local arrangements (hybrid library management systems, virtual learning environments, and so on). It will also work to identify further ‘infrastructural’ services – authorization and user profiling have been mentioned – to better support local integration also. Priorities will fall out of the technical review.
The way forward
In practical terms, a range of work is now underway which will support the strategic development of the DNER. There are investigations planned or going on in the following areas:
- Content mapping study: looking at procedures, practices, actors - completed
- Collections subject mapping exercise: looking at collections against subject benchmarks to identify gaps - being commissioned.
- Service delivery framework: a stakeholder analysis looking at pattern, governance and coherence of service providers - being commissioned.
- Technical architecture: aiming to identify broad architecture and technologies - commissioned from UKOLN, due to report in early 2001
- Presentation requirements: being specified.
These will all support the emerging DNER strategy will will take shape within the broader JISC Strategy. This will have the following components: collection development strategy, service delivery strategy, preservation strategy, development strategy, and communications strategy. Some of this work now exists in draft and is being consulted. Some awaits the outcomes of the work mentioned above. The aim is to have the completed strategy framework in place by early Summer 2001. This will in turn be supported by a range of policies which encapsulate decision making processes in important areas, and by several plans which outline particular service developments (draft plans for advancing services for FE and for providing distributed image services now exist). Documents will be shared on the DNER website as they become available.
This work is being carried out or co-ordinated by the new DNER Team based at King’s College London. This is how the team is shaping up:
Lorcan Dempsey is the Director of the Team and he is supported by three Assistant Directors. One for Collections and Communications (Alicia Wise), one for Development (vacant), one for Preservation (Neil Beagrie). Staff work closely with JISC committees, JISC services, and with other partners to advance work. They also work closely with projects in development programmes. (One reason for putting a development strategy in place is so that future programmes have a clearer framework within which to be taken forward).
In conclusion …
We have discussed some points of contact between DNER and hybrid libraries. We have noted how the DNER will be taken forward, noting particularly continued activities in the collections area, and the development of the concept of an information environment which supports secure and convenient access to DNER and other collections. The DNER team will be refining consultative mechanisms for its collections activity. We also need to think about how best to have a dialogue over development and integration issues. This dialogue needs to take place with several communities – with those responsible for developing hybrid and digital library services, as well as with those developing learning environments, and others. We welcome advice on how best to achieve a fruitful dialogue here.
And finally, the DNER team has a number of priorities but one thing should not be forgotten in the rush for progress. The DNER is an important service development for users but as yet it is lacking one thing - a memorable and marketable name. The terms ‘eLib’ and ‘hybrid library’ have been successful in inspiring the imagination of the community. We look forward to an inspiring name for the DNER also!
Stephen Pinfield is Academic Services Librarian at the University of Nottingham. Before this he managed the BUILDER hybrid library project at the University of Birmingham. He is acting as part-time Programme Consultant to the DNER Office, overseeing eLib Phase 3 projects. (email: Stephen.firstname.lastname@example.org).
Lorcan Dempsey is the JISC/DNER Director, based at King’s College London. He previously worked at UKOLN, a partner in the Agora hybrid library project and several other related activies. (email: Lorcan.Dempsey@kcl.ac.uk)