National, Sectoral and Regional Grids: options for cross-sectoral links

Derek Law

Director of Information Services & Systems, King’s College London

After a period of some uncertainty we have now arrived in a settled political climate where we may expect the settled policies of a Labour government to be in force for perhaps a decade. Several elements are worth remarking on as collectively they influence the argument for cross-sectoral links. The first of these is the shift to regionalism. Scotland, Wales and London are to have assemblies while England is to have nine regional development agencies with their own budgets and powers. Perhaps not coincidentally, Higher Education is also moving from a national to a local focus. JANET, the Joint Academic Network, is being extended by the development of Metropolitan Area Networks which offer cross-sectoral opportunities in a way that JANET never has. Meanwhile the implementation of the post-Follett Anderson Report with significant funding has a budget line for cross-sectoral co-operation.

Other national initiatives also have the potential for significant cross-sectoral impacts. The National Grid for Learning will link all schools and many school users will also be public library and network users. The University for Industry is also a key government initiative. It will not run its own courses but will buy in courses from existing providers. This again implies a cross-sectoral basis for action. Twenty million pounds have been set aside for training librarians from the New Opportunities Fund and a further fifty million for content which is seen as related to both National Grid and University for Industry needs. In short government policy assumes implicitly that cross-sectoral co-operation can be made to underpin its regional and educational ambitions.

A number of networks exist actually or potentially at national level. At present these are managed separately, but it could be argued that rather than operate half a dozen nondescript networks we could organize these together at national level and create a truly world-class cross-sectoral network. As will be argued below this might be organised on a regional or metropolitan area network basis creating a necklace of local resources.

We have the potential of a number of national networks with intermediate to poor interoperability, each adequate to its own function but with the whole performing sub-optimally, when for the same national investment we could have a network which would be the envy of the world. However this requires a hitherto unsuspected capacity for interdepartmental planning.

The National Networks. A number of national networks exist actually or potentially. Higher Education has the JANET and SuperJANET network linking all Higher Education establishments. The NHS has NHSnet linking hospitals, but with no links to HE, even in teaching hospitals. A recent report proposed a People’s Network for Public Libraries while another proposed the National Grid for Learning to link schools. The Fryer Report on lifelong learning and the proposed University for Industry implicitly assume network access, although its nature is unspecified. And of course the government has its own network. It seems at least arguable that unless the issue is considered at the highest level we could create half a dozen uninspired but fit for purpose national networks - or we could combine the available resources to create a world-class structure.

National Policy on Networks. It used to be stated by the Office of Arts and Libraries that our national information policy was not to have a policy. A similar state is emerging with network planning although a few straws in the wind might be seen as encouraging. The Fryer Report alone gives encouragement. It assumes a universal need which implies access from and to every community. It is difficult to see how this need can be met except through networks. From a different perspective the report on the Superhighway Initiative describes the need “to organise the groupings of all schools into neighbourhood clusters” and to “encourage and support links between the schools and other local partners such as,..libraries and colleges”. There is a clear perception of cross-sectoral linkages as the way ahead. It may then be seen that regionally based politics and MANs provide a very elegant realisation of a political ambition.

The Growth of the Web. It is all too easy to forget just what a novelty the Internet is – and more particularly the World Wide Web. The first web browser was created only in 1994. It is better to think of it as a global experiment rather than a mature service. Paul Evan Peters, the late Director of the Coalition for Networked Information called it the largest mass migration in user history, and the figures bear this out.. It achieved fifty million users in a mere four years, whereas radio took thirty-eight years to reach that figure and television took thirteen years. A further year of growth has take the web to seventy million users.

And yet for all its attractiveness, all net users are familiar with the joke that WWW stands for World Wide Wait or wryly recognise the truth of Dilbert’s comment that all the time saved in the last fifty years by automation has been negated by the time spent sitting in front of computers waiting for web pages to load. The fat that a source is available on the Internet does not make it a resource which can be offered as a service to users.

The problem is further compounded by the apparent ease of use of the web to anyone, without the need for training or intermediaries. The satisfied inept are a growing group who fail to recognise that the stunning breadth of information available on the Internet conceals the shallowness of what is offered.

Internet meltdown? It is increasingly appropriate to question whether the Internet is the most appropriate general vehicle for information provision. It is a truism that for most of Europe the United States becomes a virtual country in the afternoon and ceases to exist in terms of connectivity.

As network charges grow – and are for the first time being introduced on JANET for international connections – the question of whether unrestricted access to international sources is a sensible policy must arise. In this context it is interesting to consider the Australian university model. Network charges have existed there for years, but for the first time differential charges are being made for services, with hits on non-Australian websites being charged at double the rate of Australian sites. This is a recognition that unrestricted international access is not a good use of resources and that a combination of local resources, mirror sites and caches can be more effective.

It has already been suggested that use of the Internet is wasteful of time. Waits for screens to load are legendary. Its resources are grotesquely inadequately classified. Search engines are very poor, with relatively sophisticated search questions producing absurdly large numbers of hits – often over a million. Poor connectivity and network topology means that many web sites are unavailable for long periods of time. The internet is wonderful when it works, but for large numbers of resources it is almost unusable. Very little thought has been given to alternative models in the rush to connect to what is good. The irony in this is that we have an excellent model for an alternative strategy in the classic library. It is a paradox of networks that electronic resources may make it worth reconsidering returning to a holdings strategy rather than an access strategy. As the cost of filestore drops and becomes competitive with bandwidth costs, it may be proper to acquire and hold information locally.

The Library paradigm. Historically, organisations have not lined up their members or users at the start of the year, offered them all say a hundred pounds and suggested that the individual then be responsible for their own information needs, for which the organisation abdicates all responsibility. More typically they set up a library. The organisation identifies its information needs and matching resources. It acquires these and collects them in one place called a library where they are managed, classified and preserved by information professionals. Arrangements are then set up to gain access to the information which cannot be held locally and is called document delivery. The same model works perfectly well in an electronic environment. Either an Intranet or a Metropolitan Area Network becomes the analogue of the library and information is acquired, mirrored or cached and held locally. This improves connectivity and allows issues of classification and preservation to be addressed and controlled by local professionals. This can be illustrated in tabular form by comparing old and new activities.

The Higher Education Experience

If Higher Education has learned one lesson about networked content it is that is the ownership costs not the acquisition costs which are key. Early decisions are required on where the data is to be stored and maintained and by whom; the implications of these decisions for network traffic and topology; the commitment to service, support and preservation by the chosen datacentres; the nature and compatibility of the search engines; methods for registering users and managing (potentially) millions of passwords.

These are very substantial issues which will make or break a programme and very few organisations are capable of mounting and sustaining a national or even regional service. Conversely, the commercial organisations which might do these will typically have poor network connectivity and/or will not be able to give credible and/or affordable guarantees on long term preservation and the refreshing of data.

Alliances. The most obvious benefit to come from cross-sectoral activity lies in the alliances which are forged. These can be manifested in many ways. The most imprecise but arguably most important is influence. A larger group has almost by definition more influence. Further, one which is designed to assist in delivering the objectives of a third party – in this case government – should expect a warm reception.

Network planning will also benefit from alliances. Whether this lies in bulk purchase of network bandwidth and services or in mobilising the complementary skills in the various sectors the much under-regarded area of network topology will be improved. This is another area where influence will also be of value in dealing with regulatory implications.

Metropolitan Area Networks have the potential to be a potent force in regional development. They have been designed by higher education staff with higher education in mind. It is time for a richer vein of the community to be involved and to discuss the reach of the network and the services, whether common or group-specific which can be mounted.

Mirroring and caching are technical but important issues and yet thee are no standards for these activities. As the cost of filestore reduces local mirrors and caches may be a better means of offering access to resources than a general routing to the international bandwidth of the Internet. Working together to define heavily used Internet sites and commonly required material will help to share costs.

Much of the success of networking in the United States is due to a liberal telecommunications environment in which “free” local telephone calls are the norm. Lobbying on regulatory issues and to licensing agencies will be essential if concepts such as public good are to be maintained and expanded. Alliances within the profession but also with other service providers will be necessary if we wish to change the current model.

Much of the work on standards such as metadata is done collaboratively and has proved very beneficial to the profession at large. If it is to have value to all sectors of the profession there is again clear advantage in having the widest possible base of involvement in such standards work. The success of standards such as AACR2 and MARC demonstrates the importance of the basic and unexciting but critical area of standards.

MANs Needs. Of course Metropolitan Area Networks do not sit in isolation. They will have to display many of the attributes at present shown by libraries. They will need an acquisitions policy. For example the images of the Visible Human Project created by the National Institutes of health in Bethesda, Maryland, have recently been mounted at a mirror site at Glasgow University. Not only does this improve the service to users who no longer have to bring the images slowly back over the Atlantic, it reduces the cost of purchasing bandwidth which that implies. Storing data locally means too that there is much greater assurance on data quality and data authority. The MAN manager will know the source and quality of the data and can indeed set minimum standards for acceptable data. Less helpfully, there are no agreed standards or contracts for setting up mirror sites. It would be very helpful if national and/or international agencies could address this and prepare model terms and conditions for this activity. Caching strategies are rather better understood and the Higher Education community has put a great deal of effort into developing these. This knowledge will have to be disseminated to the MANs, who should benefit greatly. The last major area of concern for the MANs would be, if the approach described in this article is adopted, how to define and manage access to external data.

Opportunities. The opportunities which cross-sectoral activity offer libraries are neither particularly innovative nor startling. That does not make them important. Content creation and sharing is one of the largest. Preventing the duplication of digitisation and creating a national corpus of material which we can all share clearly offers maximum advantage to the various funding agencies interested in this area. Similarly staff training programmes while not sector independent can have a high degree of commonality in the IT area. Sharing services such as reference support is neither fanciful nor impossible if one looks at the sort of help desks rum by commercial companies. Another obvious example is in the area of document delivery. Retaining revenue within the local area is as attractive as the speed offered by local collections whose OPACs at least are on-line. The same basic material can also be repackaged for different purposes. For example a recording of a local concert can be used for recreation, for teaching composition or sold to the public. Finally we should consider entering the publishing arena. The wealth of non-copyright material which can be nade available electronically is substantial and ripe for exploitation into a number of markets.

Acting together for common needs. Libraries have a great deal of experience of co-operation. The still vibrant LIPs are a good example of this. Local Information Plans were created in an attempt to harness the cross-sectoral library resources of a region. These continue to be created and demonstrate the will to co-operate. Essentially these provided routes to material. MAN’s will develop, increase and make more available the resources of the region. As an added benefit this improved access to resources can be provided electronically without compromising the physical security of either the library collections or premises, long a worry of the larger libraries in any consortium.

Institutional Needs. It is important to stress that the benefits of the MAN can accrue to the entire parent organisation and not just the library service. It has been shown that in a larger metropolis, local higher education can comprise as much as 10% of the local economy. Universities will have many local relationships which would be improved by electronic connectivity. Student accommodation and student support has links to Local Authorities involving thousands of paper based transactions. Most universities have a large and relatively volatile estate which involves significant relationships with the local planning and environmental departments as well as the Fire Service. E-mail, electronic diaries and e-documents all offer the ability to facilitate links to schools, hospitals, local authorities and industry.

Cross-domain links are also increasingly important . Museums, galleries and archives bring different curatorial traditions and a quite astonishing richness of media content to the table. This is only very slowly being recognised. A wonderful example of this is the SCRAN project which mixes media from many sources to create packages aimed at a variety of audiences from schoolchildren to academic researchers

The SCRAN project is ideally suited to an era when the government is committed to “Education, education, education” and the continuum of learning is seen as critical to national development. In this environment using the MANs to link all education from schools to universities seems to have a certain inevitability

Conclusion. The emerging political climate in the United Kingdom favours regionalism. At the same time both the public and private sectors are making a massive investment in networks as an essential element of infrastructure in creating an information society and developing an information economy. The new political orthodoxy of “education, education, education” and cradle to the grave learning must encourage us to rethink our approaches to sharing and co-operation. Cross-sectoral investment seems inevitable and we should be prepared to press cross-sectoral services as part of that vision.