Digimap: Mapping via the World Wide Web

by Peter Burnhill, Director of EDINA and Co-Director of the Digimap Project

Stop Press

In March, the JISC and Ordnance Survey agreed that EDINA should transform the Digimap eLib Project into a national online mapping service for staff and students throughout UK higher education. EDINA and MIDAS will collaborate to provide user support.

Digimap is a system that provides Internet access to Ordnance Survey mapping and digital map data. The 'demonstrator' version of Digimap, which was developed at the Data Library as part of the JISC eLib Programme, is in its second trial year at the Universities of Reading, Oxford, Newcastle, Glasgow, Edinburgh and Aberdeen. The librarians and other support staff at those universities have done much to make the project a success.

The Digimap Project included a number of accompanying evaluation studies that monitored the experiences of 400 users during the first full academic session, 1997/8. These record the wide variety of purposes to which maps of Great Britain are put in academic research and teaching: perhaps the most surprising fact to emerge is that over 80% of the users in 1997/8 were not geographers. The principal use by staff and students has been to generate maps to view on screen or to send to the user's local printer, although about 20% of the use has been to download digital map data into desktop geographical information systems.

How it works

We have recently started testing a very simple client interface that allows access to digital mapping via standard web browsers, in line with our policy of maximising accessibility. Our trial sites report that this version requires little support and we anticipate that it should pass the 80:20 rule - up to 80% of tasks can be done using the standard interface.

"It's perfect for students who just want an A4 map to put in their project or assignment. It's so easy to print from it that they find it wonderful (and I must say I agree with them)."
Linda Kelly, Newcastle University Library

We intend to use Java applets to have greater functionality, and data, delivered to the client desktop. We began the trials using a Java applet, which was innovative at the time and certainly ensured the feasibility of the project: in all, there were some 35,000 maps delivered over the Internet using the Java applet. It was, however, demanding of the client workstation and was not suitable for the novice user.

Past, Present and Future

The origins of the Digimap Project lie in initial discussions in 1992 at the British Cartographic Society (BCS) Map Curators' Group, on the problems and opportunities facing university map libraries, as Ordnance Survey large-scale maps became regarded as expensive or were available as digital map data only. We presented follow-up papers in 1994 and 1996. In 1994 we reported on a feasibility study and proposed a solution based on the online data library model, comparable to that offered by the UKBORDERS™ service. By 1996 we had embarked on the Digimap Project and were able to place the problem within the wider 'post-Follett' context of the challenges and opportunities posed by the electronic library, and the demand for direct access by end-users to networked electronic resources.

In parallel, the JISC and the Ordnance Survey had begun another round of negotiations for UK-wide academic access to the Ordnance Survey digital map data. The results from our evaluation studies, and our experience in handling the Ordnance Survey map data, helped give a focus to those negotiations. In November 1998, the JISC put out Circular 12/98, the response to which indicated sufficient support for a national agreement and progress to the next stage - planning for a national online service over the next five years.

We can now report that the JISC has accepted our proposal, recommended by the Committee on Electronic Information (CEI), to transform the Digimap 'demonstrator' into a system capable of sustaining a national service. The proposal calls for the launch of an online service early in 2000. The online service for staff and students in all subscribing HEIs will be preceded by a set-up period in 1999. This will be used for preparation by those providing the national service and for a programme of training and awareness for support staff in the universities which have indicated that they wish to subscribe. We also hope to provide some services to field-testers and specialist users of digital map data during 1999.

by Terry Bucknell, Faculty Team Librarian, University of Leeds

When this unsuspecting engineering librarian saw the message posted to lis-scitech on 28th August 1998 announcing that BIDS had not been chosen by the JISC to continue to host Ei Compendex, his heart sank. As other messages were posted to the list, it became clear that I wasn't alone. Ei Compendex is the most comprehensive interdisciplinary engineering information database in the world. The prospect of inducting hundreds of new engineering students in September/October and then retraining them in December filled me with dread. Then there were the lecturers who told their students to 'search BIDS' as if it were a single database. Hastily I had to cancel our order for BIDS Ei Compendex user guides and amend our training materials to include the rejoinder, 'and once you've mastered the interface, it will change'.

As D-Day, 6th December, approached it became clear that negotiations between Eduserv Chest and Ei were protracted, and had delayed the release of the EDINA service until 7 January 1999. It came as quite a relief when it was announced that Ei Compendex would continue at BIDS - at first until the end of December and then until 30th June 1999. The shift to 7 January left us plenty of time to decide how and when to effect the switch-over. The news that EDINA would accept logins from ATHENS shared access accounts until the end of June was good news too, although our users appear to have experienced little difficulty in setting up personal ATHENS accounts for other EDINA databases. We were already considering bulk uploading for the next academic year, so it was always likely that our users would have another new username and password for Eduserv Chest databases for 1999/2000.

Initially, I wasn't too impressed with 'New Ei Compendex' (much like New Labour...) but an e-mail to the EDINA helpdesk revealed that most of the concerns I had were already being addressed by EDINA. In fact, I thought that the reply was so useful that I asked the EDINA helpdesk to forward it to lis-scitech. Some improvements have already gone live: records are now presented most recent first, and there is now an on-screen tally of marked records.

Here at Leeds we have directed users to BIDS Ei Compendex for the time being. We will promote the EDINA implementation with the next release at Easter when single-step marking of records and user profiling should be available. The EDINA Ei Compendex posters and flyers are certainly eye-catching, and we are confident that pointing our readers to the choice of services between April and August will result in a smooth switch-over: final-year students will be able to stick with BIDS; everyone else will have months to get used to the new interface.

As Leeds also subscribes to EDINA's Inspec service, the prospect of future cross-searching between Ei Compendex and Inspec is of real interest to us. In a few months, EDINA has turned from 'that datahost that I don't need to bother about', to the single most important online information source for engineers. The hype of the DNER for Engineering is rapidly becoming reality.

by Roddy McLeod, Senior Faculty Librarian, Heriot-Watt University

Up until the early 1990s, we allocated a very reasonable budget for mediated online searching at Heriot-Watt University Library, and we were able to offer searches at no cost to our users. As might be expected, the service was well-used, and each year we would make several hundred searches. One of the most popular databases was Inspec. Even though we subscribed to the printed Science Abstracts A, B and C, the allure of relatively easy searching through several million references was too strong for most of our researchers to resist, and the only disadvantage was the call on the time of the subject librarians, who had to make all the searches and then audit the resulting invoices from online hosts.

Along with many other institutions, financial cutbacks eventually meant that our online budget all but disappeared, and we were forced to recover the cost of most searches directly from readers. In practice, this meant that those with access to sizable research grants could afford to run searches, including some academic staff and a small number of fortunate postgraduates, but others suffered, in particular undergraduates. Even though we took advantage of educational discounts offered by some of the commercial hosts, the number of searches reduced considerably. Some side effects of the reduction were that the subject librarians increasingly became 'stale', finding it hard to keep up to date with advances in online search techniques, and little promotion of the service was undertaken as it was seen to be like dangling a carrot in front of people who couldn't afford the direct costs. However, a number of online alerts (saved searches) continued to be run, with in some cases the results being sent directly to the researchers by email, and in fact this amounted to our main use of the database.

During this period we considered purchasing Inspec Ondisc, the CD-ROM version of the database, but our lack of a campus CD network made this a less than perfect option. Then, finally, Eduserv Chest negotiated a dataset agreement for Inspec through five different authorised service providers. We chose EDINA as supplier for a number of reasons, not least because of the Ovid interface, and potential synergy and cross-searching in the future with EDINA Ei Compendex+. Now the database is available to all members of Heriot-Watt University at no direct cost to the user, and once again we look forward to renewed and substantial usage of the database. Networked access to Inspec brings us new challenges in that it is another database to promote, and although the subject librarians will no longer need to undertake searches themselves, they will find themselves well placed to introduce it to staff and students. Networked access to Inspec has a price, of course, but one of the first things we will do is transfer all current online alerts in the database to saved searches on EDINA Inspec, thus saving nearly ¥2,000 per annum of our researchers' hard-earned funds.

by Lorcan Dempsey, UK Office for Library and Information Networking, University of Bath

The Joint Information Systems Committee (JISC) provides a range of information services for the UK higher education community. Collectively these are the basis of a national electronic resource which is highly valued and a testament to the utility of collective action. In its Collections Policy, the JISC has now outlined how this collective resource will be taken forward. It has also turned its attention to the next phase of development, the creation of the Distributed National Electronic Resource. This is an idea which is currently being elaborated through various discussions. In what follows I offer a personal view of some possible DNER directions.

First, it is interesting to compare the current web environment with the types of services that might emerge through the DNER. A large part of what makes the web so seductive is that it gives point and click access directly to information content. A user can discover, locate, request, and have delivered documents using simple, integrated, point and click actions. One might suggest that the web provides simple 'content infrastructure', infrastructure which supports exchange and use of information content. This is not to say that it is an ideal information environment, but it is accustoming our users to easy access.

Consider, in contrast, access to the journal literature. A potential reader has to discover relevant articles, locate where they are, request them from some source, and have them delivered. There are two areas where integration is lacking. The first is within operation, where one may have to interact with different systems to complete the operation satisfactorily, or where systems support the operation in different ways. For example, to discover items, one interacts individually with bibliographic databases on CD-ROM, at the datacentres, and elsewhere. To search several, one looks at each, behind its different interface and logon procedure. When it comes to locating where a desired item is, one may have to check a catalogue, a union list, or some other resource; and so on. The second area is creating the links between operations which support automation of end-to-end processes. An example of process automation might be where data is passed between a search service, a document delivery service and an accounting system, without the need for repeated manual intervention. Neither of these forms of integration is well supported, and this is a barrier to use. The existing model of service in this area is of standalone network services, which are integrated by user effort. There is limited 'content infrastructure' ¥ which would support point and click working across various services, or linking together services to automate processes. To discover, locate, request and have delivered items requires interaction with many systems with different access and use characteristics. Services do exist which provide some form of aggregation, but none exists which will provide the full range of what a user might require. This example focuses on the journal literature; readers will be able to supply examples from their own experience of other ways in which better integration would support better use of resources and better use of their time. The current 'clunkiness' of use is a feature of the immaturity of our information environments: they are largely, after all, still under construction.

It is agreed that this is not a sustainable approach to the provision and use of rich learning and research materials. For this reason, the JISC is promoting the provision of an additional layer of service which weaves these resources into a fabric of integrated use. Similar services are already the subject of some development effort: for example, this is a part of the focus of the eLib clump and hybrid library projects. They aim to provide common points of access to other services, hiding some of the difference from users, or creating links between services. So, for example, the Agora project hopes to integrate the end-to-end process of discovery, location, request and delivery of documents, and to integrate access to different services within each operation. We might call such higher level services 'brokers'. A broker provides consistent access to other services. Other examples within the higher education area are the Arts and Humanities Data Service Gateway, and the cross searching facility provided by ROADS for subject gateways. The gain with such federating approaches is that a library, or learning environment, can plug this higher level service into their offerings, saving the time and effort of their users. This does not mean that users cannot continue to interact with individual services as and when they wish, subject to whatever privileges they have. It does mean that their information use can be supported in new ways, and that greater content infrastructure reduces some of the effort involved in the use and management of resources. Current brokers are early examples of services which we can expect to become more common. Indeed, brokerage services are likely to become significant parts of local information and learning environments.

I have discussed some issues arising from access to the journal literature, partly because of EDINA's potential role in this area, but there are other examples of how services might be woven into a wider fabric of use. For example, EDINA is discussing the integration of different types of resource in the engineering area with the Edinburgh Engineering Virtual Library (EEVL). This is an example of subject level integration, and one can imagine various ways in which this could develop. EDINA's involvement in geospatial data suggests other potential integrative directions.

The collections policy rightly argues that "integration is key". Such integration should improve the quality of research, learning and teaching in two broad ways, from which other benefits would flow. First, it should improve the effectiveness of user interaction with resources, saving their time, and making it easier to link helpful resources - or collections of integrated resources - directly into their working environments. Secondly, it should improve the discovery and use of information resources, releasing the value of the JISC investment. This emphasis is a sign of the growing centrality of network information. With the growth in variety, volume and volatility of digital resources, effective use depends not merely on pointing people to resources, but supporting selection and active use.

So what types of things would it be useful to do? Three areas come to mind:

Description. The collections and services which comprise the DNER need to be described, so that potential users can discover and use them, and so that libraries, subject gateways, or others can advertise the services they provide. Descriptions should be readable by human users, but also, increasingly, by machine users acting on their behalf. Several things need to be described. There is the 'content' of a resource to facilitate discovery, selection, and use. Then there is how it is provided as a network service: in order for a client or broker to access a network service it must know: the location of that service; the access protocol (which may be HTTP, a search and retrieve protocol, a directory protocol or something else); the request format, which may be defined by a query language; and the schema(s) relevant to the service, (or example the metadata format in use. Finally, the terms and conditions of use may also need to be described; these may be associated with particular resources and services, and vary according to user and use. Initially, such descriptions may be provided in a database (UKOLN has been experimenting with such an approach: see http://www.ukoln.ac.uk/metadata/roads for details). Gradually a more distributed framework might emerge, making use of directory services and registries. Such a resource would make the DNER a more tangible part of information use, more effectively advertising what is available.

Federation. I have suggested that an additional service layer needs to be provided. This is brokerage activity which is concerned with management of service integration through additional federating services. The DNER has been a centrally led initiative, and has been welcomed by the community. While it is clear that a higher level of integration is desirable, it is less clear what form this should take and over what services. Accordingly, it would be useful to identify usage scenarios which will motivate the construction of the DNER. A couple of familiar examples are given here; some investigation of desirable scenarios in geospatial or scientific domains would be interesting. I have discussed journal articles. A service which supported use of the 'DNER journal collection' would be interesting. The DNER contains discovery services (abstracting and indexing services): it would be useful to be able to search across these in different combinations in response to user interests. An article may be available from numerous locations: library collections, the British Library Document Supply Centre (BLDSC), JISC-supported digital collections, other digital collections, and so on. Significant content infrastructure would be required to match records against this variety of resources to produce locations, but some work here would be beneficial. The broker might also provide access to request and delivery services. Broker services might also have a subject focus. The Resource Discovery Network will be organised around faculty level hubs, but currently services concentrate on one collection type (internet resource descriptions). A subject hub which federated different collection types (internet resource descriptions, data sets, journal articles, mailing lists) would be interesting. EDINA and EEVL are moving towards this type of arrangement. These and other scenarios might be realised through the construction of broker services.

This would in turn depend on providers making their network services available through agreed service profiles which facilitate interworking. Although the ambition in the long term will be to develop brokers in such a way that they provide a plug and play infrastructure for the use and exploitation of content, it is likely that initial broker services will require quite a bit of customisation or bespoke development given the lag between standards developments and requirements, and the immaturity of the current infrastructure. Where possible and relevant, service providers should conform to agreed service profiles to ensure maximum participation in the DNER. A set of common profiles which cover common resource types could be defined, and conformance to them would influence the selection of resources and suppliers. Such agreements would also support broker services being developed by local institutions (the hybrid libraries or clumps for example) or other emerging services. I am suggesting that some JISC funded broker services would be a useful way forward. In due course, such central services may or may not be necessary as a variety of broker services become more common in different settings. EDINA, for example, has invested significant effort in developing 'content infrastructure' which might be built on to offer brokerage across sets of services, some of which it hosts, some elsewhere.

In addition, an integrated approach to the DNER will benefit from approaches to authentication and rights management which become 'infrastructural', in the sense that they become services which are predictably and consistently available across providers, rather than being developed on a per-provider basis.

Embedding. The hybrid library projects have attracted quite a bit of interest, largely, I suspect, because they respond to this perceived need to begin to integrate at a deeper level than adding links to resources on a web page somewhere. In some ways, the hybrid libraries are the other side of the coin to DNER developments. They are looking to develop institutional broker services, not limited of course to brokering access to JISC services, but potentially benefiting from DNER developments. A next stage might be to create several exemplar institutional environments where information, learning and other resources are brought together in a user's normal working environment, together with rich communications and other tools. Whereas other programmes create the pedagogical framework for learning applications, this could provide patterns for weaving learning and informational resources into integrated institutional services, a step which will be essential to support longer term learning agendas. Alongside such work, other strands of activity might be supported which provide examples of, or tool support for, richer local information and learning habitats.

This is a short article, and this constraint perhaps suggests a more schematic view than I intend. Nor should the difficulty of doing some of this be underestimated. A changing technical environment, unsolved challenges, a lack of experience with working in such distributed contexts create uncertainty. This suggests to me that it is important to put in place a planned approach, involving strategically important content and technology providers and using JISC investment as a lever for wider change. This is partly because it is crucial that the 'content infrastructure' put in place to manage the DNER is as closely aligned with developments elsewhere as possible, and, moreover, that the lessons learned in its construction are shared in targeted ways with other national initiatives and sectors. As in other areas, the JISC now has the opportunity to act as a catalyst and exemplar for wider change, which in turn benefits higher education. The JISC is right to emphasise integration as an aspiration. Without such integration networked information systems will poorly serve developing research and learning agendas, and fail to deliver their full potential.

UKOLN is funded by the JISC and the British Library Research and Innovation Centre, and is supported by the University of Bath where it is based. The author is writing in a personal capacity.

The term 'geographic information', or GI, refers to digital information or data that is spatially referenced and is produced for use in computerised systems known as Geographical Information Systems (GIS). One form of output from a GIS may be Ordnance Survey maps, of varying scales. Another can be thematic maps showing, for example, statistical summary information related to areas across the country, defined by digitised boundaries. A contemporary example in medicine is the use of GIS in the management of a general practice by combining address and postcode patient information with population census, road network, and the location of various services (such as pharmacies), in order to help manage workload and planning. Addresses and postcodes can also be of value to historians. For example, in places such as Edinburgh New Town, where the street layout has not changed for some two hundred years, it is possible to fix the location of the residences of merchants at different times, and map these in relation to other known attributes such as income and family size.

GIS software handles spatially-related data as a series of transparent layers: data from a variety of sources can be overlaid, and viewed in a GIS. Because the layers are transparent, the composite visual effect of placing layers of data on top of each other is that of a map. But the real value of GIS for geographers and non-geographers alike is that, as well as visualisation, they allow the analysis of geographic information data based on common spatial relationships: the overall result is, in effect, greater than the sum of the parts.

Geographic information is, in other words, as much about socio-economic attributes as topography. In UKBORDERS™, EDINA provides UK higher education institutions with online access to a wide variety of high precision digital boundary data. The largest single user group is comprised of geographers, which is not surprising. But because the boundary data relates to a variety of geographies (census, administrative, electoral, postal, etc.) and includes historical data, the boundaries available are used by an increasingly wide variety of users from other disciplines.

The period covered is from the mid-19th century to the present, the 1980s onwards having the widest variety of data. Because there is an increasing tendency for statistical data to include a spatial reference, such as a postcode, it is possible to link them in a straightforward way to relevant digital boundary and map data and exploit the information by use of a GIS or mapping package.

The list of examples of use of geographic information is long, and growing, as is the range of boundary data provided in UKBORDERS™. With the increasing availability of affordable desktop mapping packages which allow non-geographers to explore spatial relationships in ways which were previously too time-consuming and difficult, the usefulness of geographic information to research and teaching in higher education will continue to grow. For more information see http://edina.ac.uk/ukborders/.

UK Serials Group (UKSG) Conference. Manchester (UMIST), 12-14 April 1999. EDINA at Stand 6.

Library + Information Show. Birmingham NEC, 8-10 June 1999. EDINA at Stand 539.

ARCLIB'99. Glasgow School of Art, 21-23 July 1999. Demonstration of Art Abstracts and Periodicals Contents Index.

CADE99. University of Teesside, 7-9 April 1999. Joint presentation with VADS and CTIAD on 'Creating and Using Digital Resources in the Visual Arts'.

UKOLUG meeting. University of Edinburgh, 23 June 1999. Main Library, Wolfson Suite. 'Engineering Resources: Providers and Consumers'.

EDINA, based at Edinburgh University Data Library, is a JISC-funded national datacentre. It offers the UK higher education and research community networked access to a library of data, information and research resources. All EDINA services are available free of charge to members of UK higher education institutions for academic use, although university subscription and end-user registration is required for some services.

EDINA services are:

• EDINA Art Abstracts
• EDINA BIOSIS
• EDINA CAB Abstracts
• EDINA Ei Compendex®
• ESPMD
• EDINA Inspec
• EDINA Palmer's Index to The Times
• EDINA PCI
• Ordnance Survey Strategi®
• EDINA SALSER
• EDINA UKBORDERS™

EDINA subscription and registration

Some EDINA services require the completion of a licence agreement before those services can be made available to users. Free 30-day trials are available for most of these services.

For most services, licence agreements must be obtained from Eduserv Chest (email chest@chest.ac.uk) and a subscription fee must be paid. Individual users must register locally at their library. If in doubt, check with the EDINA website.

For UKBORDERS™, there is no fee for academic institutions within the UK, but a licence agreement must be signed (email edina@ed.ac.uk) and individual users must sign an End User Licence.

For Ordnance Survey Strategi, each institution is required to hold a current and valid Ordnance Survey Educational Copyright Licence in addition to a subscription to EDINA. Contact EDINA in the first instance (email edina@ed.ac.uk).

SALSER is a completely free service, with no subscription fee. No licence or prior registration is required.

EDINA contacts
Helen Kerr, Claudia Gröpl and Stuart Macdonald (Helpdesk)
Margarete Tubby (User Support Manager)
Alison Bayley (Manager, EDINA National Services)
Peter Burnhill (Director of EDINA)
Tel: 0131 650 3302
Fax: 0131 650 3308
Email: edina@ed.ac.uk
URL: http://edina.ac.uk

Reference cards

Reference cards for most EDINA services are available for purchase at £12/100. They are also available free from the EDINA Web pages in PDF and PostScript formats.