Hybrid Libraries

What is a hybrid library?

A hybrid library is not just a traditional library (only containing paper-based resources) or just a virtual library (only containing electronic resources), but somewhere between the two. It is a library which brings together a range of different information sources, printed and electronic, local and remote, in a seamless way.

So, as we are all used to dealing with resources of various media, does this mean that we are already working in hybrid libraries? Well, in one sense, ‘yes’. We are all already working in a hybrid information environment. But the term ‘hybrid library’ is normally used to convey the idea of greater integration of the different media as well co-existence. As Chris Rusbridge puts it, the hybrid library should be “designed to bring a range of technologies from different sources together in the context of a working library, and also to begin to explore integrated systems and services in both the electronic and print environments.”[1] This means that hybrid library development is not working towards a single, easily definable technical goal but is rather an ongoing process of trying to achieve greater integration.

The point of this is to encourage users to look at the best information source for their needs regardless of its format. At present, people wanting to use resources in different formats have to go to different places or use different equipment or switch between different proprietary interfaces. A lot of time is wasted doing this and many users are put off completely. But within the hybrid library access to a range of resources should be made seamless, so that people use the resources which best suit their needs rather than those which are simply more easily available.

And clumps?

A clump is an aggregation of catalogues (or other databases). The point of a clump is to enable large-scale resource discovery - so that users can search for resources they want across a number of different catalogues.[2]

Clumps fall into two categories: physical clumps and virtual clumps. The first of these is a union catalogue which brings together the data from a number of different sources and holds it in one place. A well known example of a physical clump is the COPAC database, which is the joint catalogue of the CURL libraries.[3] This database is held and managed at Manchester Computing. COPAC currently contains records from 11 institutions which it has uploaded onto a single database available on the Web.

On the other hand, virtual clumps are not centrally managed databases at all. Rather, they are distributed databases which can be aggregated for the purposes of a search. This aggregation may be ‘fixed’, so that users can always search a pre-determined group of catalogues; or it may be ‘dynamic’, so that users can themselves form a clump of particular catalogues (or parts of them) at a particular time. The latter is in many respects the ideal of clumping and would create major benefits for users.

The key technology involved in the formation of virtual clumps is Z39.50. Z39.50 is a network protocol which allows bibliographic databases to talk to each other using a predefined set of rules. A query is sent by a user to a Z39.50 client (‘origin’) which can then be forwarded to a Z39.50 server (‘target’). The target database is then searched and the results returned via Z39.50 to the user. Several different databases may be searched simultaneously and a single results set returned to the user. One of the main advantages of Z39.50 is that it allows the user to carry out searches simultaneously on a number of target databases using a familiar, local user interface.

What has eLib got to do with it?

eLib is currently funding a number of projects which are developing hybrid libraries and clumps. These are part of eLib Phase 3 which has as its watchwords ‘implementation’ and ‘integration’. The eLib projects are listed in Tables 1 and 2. Most are consortia of different HE institutions and other partners who are co-operating to develop models and exemplars of the hybrid library and clumps. Most of the projects began in January 1998 and are due to end in 2000.

Of course, eLib projects are not the only people working in this area. In the UK, there are a number of EU and other projects with similar or related aims. Outside the UK, an even larger number of people are working on hybrid libraries and clumps, even if they don’t call them that.

So what are the projects doing?

What are the clumps projects doing?

The eLib clumps projects are all aiming to set up working clumps, using Z39.50, which will operate amongst the partner institutions. To begin with, this involves a number of technical stages. Firstly, a server needs to be installed to which users can send their queries. Some projects are installing more than one server in different locations. Music Libraries Online and RIDING are using software on their servers produced by Fretwell-Downing Informatics who are involved as technical consultants in the projects. FDI are also producing a user interface (now available in prototype) which the different projects will be able to tailor to their needs and deliver on the Web. CAIRNS and M25 Link are currently investigating the software options, and are looking at products, such as Europogate and GEOWEB, in addition to the FDI software.

Then, of course, all of the participating libraries have got to ensure that they have Z39.50 compliant catalogues which can receive requests and send back results. All of the projects consist of participating institutions which between them have lots of different library systems.[4] One interesting part of the projects will be looking at how the different systems cope (or can be made to cope) with Z39.50.

Z39.50 is in many respects a complex series of rules about how databases can talk to each other. One very important aspect of the projects is agreeing on a particular set of rules (a ‘profile’) which all the participants can use. This will enable their systems to interoperate.

The precise arrangements the projects decide to make will differ. Music Libraries Online, for example, is looking into a range of issues which are unique to music resources, which may require specialised treatment. Other projects may find that agreeing on a profile may not be possible, at least in the short term. An alternative arrangement, planned for example by RIDING, would be to configure the central server so that it can deal with a variety of profiles and so communicate with catalogues configured in different ways. But whatever they decide, all of the projects will have to go through a lengthy series of tests to achieve interoperability.

It is envisaged that within clumps systems will talk to each other on different ‘levels’. Before users search a clump at the level of individual records, they will want to know whether it is worth their while searching these databases at all. For this reason, the clumps projects are working on the idea of ‘collection level descriptions’ (CLDs). These will provide ‘forward knowledge’ of collections of individual items to users, so that they will be able to decide whether to search them for particular sorts of items or not. It is hoped that this development will help to minimise the number of fruitless clumps searches being made which could potentially use up valuable system time or network capacity. The format of CLDs is currently being worked out. CAIRNS is using an existing CLD scheme, the Scottish RCO (Research Collections Online) conspectus database.

Most of the clumps projects are regional groupings based on existing regional co-operative arrangements. CAIRNS is based on the Scottish Confederation of University and Research Libraries (SCURL), M25 Link is based on the M25 Consortium of Higher Education Libraries and RIDING is based on the Yorkshire and Humberside Universities Association (YHUA). They see clumping as a natural extension of their existing co-operative arrangements. On the other hand, Music Libraries Online is the only subject-based (rather than regional) clump. But it is also based on a network of well established institutional relationships. These institutional relationships are at the heart of clumping developments. All of the projects have it within their remit to investigate the policy implications of enhanced institutional co-operation implied by clumps. Making large-scale resource discovery easier will inevitably increase demand to look at the resources discovered. How will the various partner institutions deal with this? They may allow more physical access, or improve inter-lending. There are even possibilities of co-ordinated acquisitions. This question is especially interesting as several clumps involve public library sites.

What are the hybrid library projects doing?

The hybrid library projects’ activities are more difficult than the clumps to pin down. There are a number of important similarities between the hybrid projects but they are all approaching the problem of integration from slightly different angles and all have a different set of emphases. Some, for example, are concentrating on technical issues, others are focusing on user issues.[5]

The HyLiFe project is looking at the hybrid library from a user perspective. It is concentrating on interface design for a variety of client groups. MALIBU is investigating the needs of humanities scholars in particular. It is working towards creating a number of specialised hybrid library prototypes in specific subject areas. HeadLine is working in the area of Business and Economics to create a hybrid library environment involving a wide range of different resources. BUILDER is looking at similar issues but from an institutional perspective, and is also considering issues of digitisation management. Agora is developing a Hybrid Library Management System which will provide access to a range of discovery, location, request and delivery services for users and will facilitate the effective management of the hybrid library for information professionals.

Whilst it perhaps too early to talk about particular technologies which are being used by the different projects, it is clear that a number of key themes are emerging. Firstly, there is authentication. All of the projects are looking at this issue to a greater or lesser extent. Just as many libraries allow only authorised users to set foot in their buildings, so electronic libraries must control access to their virtual spaces. In the hybrid library, both are important. Investigations are being carried out by a number of the projects on how to achieve more streamlined and less intrusive authentication processes which will allow users access to a variety of resources. Working with services such as ATHENS is obviously crucial here.

Another key theme is interconnectivity. Projects are investigating the technologies and tools which allow access to different databases to be interlinked. Z39.50 is important here, as are a number of other technologies. The Agora approach to this is based on the MODELS Information Architecture (MIA) produced as a result of the MODELS workshops.[6] Other projects have been informed by this approach. Most of the projects are working with external data providers on this issue.

The whole idea of improving the environment within which the user carries out the information seeking process is also important. There is a move towards a greater ‘personalisation’ of this environment. A number of the projects are investigating ways in which different views of the ‘information landscape’ (more buzz words) can be shown to different users or groups of users. This involves managing metadata about sources and also authorisation data about users in such a way that they can be made to interact; so that users can be matched with the kind of information sources they may need.

All the projects are also interested in cultural and skills issues, both from the point of view of users and information professionals. All of them are carrying out evaluation activities to gauge views about the hybrid library and its different elements. It is hoped that these studies will contribute to a better understanding of the current position and how it is likely to change in the immediate future.

What will the impact be?

Hybrid library and clumps projects are looking at issues which will have a real impact on libraries and information professionals in the near future (if they are not already). They will have an impact on the services that libraries deliver and also on the kind of skills information professionals need to develop in order to manage these services. They also imply the development of partnerships amongst HE institutions, and also between HE institutions and commercial partners, such as data providers. Library and information services should be active in developing and implementing these services; it is only then that we will be able to shape them in the way that is best for us and our users.

Acknowledgements

Thanks to the eLib hybrid library and clumps project managers for giving me their comments on drafts of this article.

[1]

Chris Rusbridge “Towards the Hybrid Library” D-Lib Magazine July/August 1998. URL: http://mirrored.ukoln.ac.uk/lis-journals/dlib/dlib/dlib/july98/rusbridge/07rusbridge.html . See also Jon Knight “The Hybrid Library: Books and Bytes” Ariadne 11, September 1997. URL: http://www.ariadne.ac.uk/issue11/knight/

[2] More details about clumps can be found on UKOLN’s MODELS Web site, URL: http://www.ukoln.ac.uk/dlis/models/clumps . The term ‘clump’ was actually coined at the MODELS3 workshop, URL: http://www.ukoln.ac.uk/dlis/models/models3

[3] COPAC stands for CURL Online Public Access Catalogue; the OPAC of the Consortium of University Research Libraries, URL: http://copac.ac.uk/copac

[4] See http://www.m25lib.ac.uk/M25Link/clumpssuppliers.html

[5] See Stephen Pinfield et al “Realizing the Hybrid Library” D-Lib Magazine October 1998. URL: http://mirrored.ukoln.ac.uk/lis-journals/dlib/dlib/dlib/october98/10pinfield.html.

[6] MODELS (MOving to Distributed Environments for Library Services). See Lorcan Dempsey et al “Managing access to a distributed library resource: report of the fifth MODELS workshop” Program, 32, 3, July 1998, pp. 265-281.

Hybrid Library Projects

Clumps Projects

Author details