Philip Hunter asked for an assessment of the achievements and legacy of the eLib Programme 5 years on. It is a strange experience trying to summarise a huge enterprise, covering more than 5 years involving hundreds of people, and costing in excess of £20M, in a couple of thousand words. I immediately abandoned any thought of comprehensiveness, any formalised evaluation, or even any serious attempt at history. Instead this is a highly personal commentary on some of the highlights for me as Programme Director of the Electronic Libraries Programme (known as eLib) with a reflection on some of the less overtly successful things – again highlights where 20:20 hindsight showed we could have done better. I will mention only a small proportion of the 70+ projects involved, with apologies to, but no intention of slighting those omitted. Great work was done in very many different ways throughout the eLib programme.
Despite not wishing to start with a history, it may be useful to remind ourselves that eLib was not intended to be a digital library research programme, in the high technology sense of the NSF Digital Libraries Initiative. Indeed it was not described as a research programme at all – JISC still tends not to acknowledge that its activities include a number of significant applied research and development programmes.
The FIGIT programme (as eLib was first called) was an exploration of ways IT could assist libraries, particularly in the context of the abolition of the binary divide and the creation of the 1992 universities. I gather the outline of the programme was devised by Lynne Brindley, Derek Law, David Cook and others in a restaurant called the “Rusty Bucket”, near Washington DC while they were visiting the US preparing their contribution to the “Follett Report,” of which the eLib Programme was one major outcome.
In many ways the eLib programme was seen as distinct from the projects. We had the notion that the programme could succeed even if all the projects failed – and vice versa in both cases! This is because we were really aiming at a sea-change, a cultural shift: to see libraries taking up these new technologies in an enthusiastic way. From this point of view, I believe the programme has been a major success (as have many of the projects). A wide range of supporting technology is now deployed in libraries as a matter of course, a considerable part of it with its origins clearly in eLib.
One of the big issues from the start of the programme was that each project had to have a finite funded life of up to 3 years. Services that might ensue beyond that point had to be self-sustaining i.e. had to have an “exit strategy” that did not depend on continued injection of JISC funding. Why 3 years? Not because that is the time needed to make a cultural change or to establish an infant service so that the community will pay for it. We now know that (ignoring the huge investments of dot.coms, many of which could eat the entire eLib programme budget for breakfast) it is nearly impossible to take an idea through prototype into real implementation to sustainability in just 3 years! In practice, we managed to find ways of extending support in some cases for further transitional periods. Perhaps the justification was merely bureaucratic: simply because JISC operated within the 3 year financial forward planning cycle operated by HEFCE.
Back to the Electronic Libraries Programme - what were some of my favourite projects (I won’t say best; this is definitely a subjective list)? The project of greatest personal satisfaction for me is CEDARS, the digital preservation project from Phase 3. Ensuring the long term existence of digital materials was not an element of the Follett report, and this seemed a significant gap when I started thinking about applying for the job of Programme Director. Others were also aware of the importance of this area, most particularly FIGIT's Chair, Lynne Brindley, now CEO of the British Library. We still have CEDARS as an exploratory project piloting ideas rather than a full-blown digital preservation service; this is another example of the difficulty of taking even widely supported ideas through research into service. It is also true that the technical problems in this area are not yet solved, but also that the real problems are organisational, political and financial rather than technical. Nevertheless, CEDARS has made real progress on both national and international fronts.
A second favourite is in the area of the re-use of current copyright material in digital form for course readings. The area was variously known as On Demand Publishing, Electronic Reserve Collections, etc. The second generation project here is HERON, a great project battling against very difficult conditions to provide a valuable service to HEIs and to publishers. Membership is growing rapidly, as is usage amongst members. This is a genuine case where, after much hard work and many drafts of business models, a business strategy may have been identified. The important thing here is to devise sustainable ways that copyright material can be used with clear, known and reasonable costs, and at short notice.
More successes? From the Electronic Journals area, a journal which aimed from the start to get value from the technology rather than just using it to carry images of printed pages – this was Internet Archaeology. Archaeology was clearly a discipline desperately awaiting the emergence of a medium like the electronic journal, which could do justice to its many-faceted multi-disciplinarity. Sociological Research Online may not have quite fulfilled its promise of bringing multimedia into the forefront of the sociology publication arena, but the journal worked, became respectable and survived without further cash from JISC beyond its initial funded period – as comparatively few other projects did. Between these and other Electronic Journals projects, and the Pilot Site Licence Initiative of the funding councils (and its successor, the National Electronic Site Licence Initiative, or NESLI), we can claim a considerable part of the credit for the rapid scholarly move towards wide-spread use of electronic journals.
In Phase 2 of eLib, we had a programme area dedicated to electronic pre-prints. While all the projects were successful in their way, WoPEc and Cogprints, plus the latter's JISC/NSF successor, the Open Citation project can genuinely claim to be forerunners of the Open Archives Initiative. This may turn out to be a crucial development in providing access to a much wider range of “grey” materials, and possibly helping to counter the debilitating effect of the scholarly journals price spiral.
Another favourite of a very different kind was Digimap (which became the award-winning EDINA Digimap service for Ordinance Survey map access), plus its little known follow-on project Digimap.Plus. The latter is looking at the issue of changing landscape; that is, matching time-stamped maps over a long period (say 100+ years). Some of the users of modern Ordnance Survey data in the original Digimap project, however, were thinking on a broad timeline too – one I remember was an investigation of sight-lines in the location of Roman forts, and another into the location of food dumps for Cromwell’s New Model Army!
There were also a number of projects which somehow did not quite meet our most ambitious aims. These included document delivery projects, where both the software complexity and the business model were problems. Despite these, Lamda has developed into a sustainable service doing quite a respectable volume of business.
In contrast, the document delivery project InfoBike may have failed in its original terms, but it mutated: first into the BIDS JournalsOnline service, which then provided the core of the spin-off company ingenta, now a £100 million publicly quoted company! In this sense it was an extremely successful project, and one of which we can be justly proud, even if some parts of JISC are a little unsure of claiming such commercial success as an appropriate yardstick (unlike the NSF, which is proud to count successful spin-off companies as a performance indicator).
In digitisation and images we suffered from the indecision of the pioneer; although anything we did was valuable, nothing appeared of itself likely to justify the very high cost of digitisation. This was an area where consistent belief in a vision of the increasing value of contributions towards a critical mass of digitised material is needed. JSTOR sustained this vision; JISC is only now establishing the structures to support this. At the time and now, it appears to me very important that JISC, with its significant digital library budgets, should put a consistent amount of money into digitisation (moving legacy information into the digital arena) every year.
Two of the later programme areas have yet to be as influential as we hope, through no fault of those working in them. The so-called clumps projects have not been able to demonstrate that virtual union catalogues based on the Z39:50 protocol scale to provide reliable service on a national dimension. The M25 Link project will connect the catalogues of almost 40 institutions, but despite important developments in interoperability such as the Bath Profile and Collection Level Descriptions, this appears pretty much at or slightly above the upper limit. Perhaps their most important contributions will be seen to have been the promotion of co-operative agendas between libraries. Meanwhile, the appropriate models for a national-scale union catalogue are being investigated through the UKNUC study funded by JISC, RSLP and the BL.
The hybrid library projects are described elsewhere in this issue. Extraordinarily interesting in detail and exceptionally innovative in many aspects including their dissemination programmes, the ideas are only beginning to change the way university libraries operate. However, these projects are just entering their final reporting stages, and some have conferences and workshops presenting their results very soon.
It was a source of some sadness to me that some kind of natural modesty persuaded most of the programme participants that they could not or should not mix it on an international stage. Despite the 1980s Coloured Books problem (associated with hanging on to a defunct standard too long rather than being wrongly innovative) there is every reason to believe JISC and participants in its projects have much to offer. There are certainly exceptions, including CEDARS and its JISC/NSF funded companion, CAMiLEON both with a high international profile. Indeed, the whole programme jointly funded by JISC and NSF which paralleled the later stages of eLib has been an important stimulus to international co-operative activity.
Two other disappointments for me: one was the small number of projects involving library or computer science schools, together with the small amount of real hard digital library research done in UK academic circles (compared with the US where the NSF has funded some major developments on rather similar budgets). Most eLib projects were based in practitioner groups (I know there were reasons for and benefits from this!).
Finally in this list, the difficulty of overcoming the barriers between JISC committees was a disappointment. This was particularly so for example when the first phases of eLib had so raised the profile of the human and cultural issues (through the Training and Awareness area) that a separate JISC sub-committee was formed; from that point, it was harder to make progress on these important issues in the digital library area which had identified them!
This survey also ignores important non-project activities, like the very significant collaborative activities between the JISC and publishers, a wide variety of studies, various copyright and other guidelines, etc.
So what of the future? For the early part of eLib we did not have a grand national plan linking the projects to other developments such as datasets. Gradually, particularly through the MODELS workshops (a favourite I have held in reserve), and linked to the subject-based Internet gateways that were coming together in the Resource Discovery Network, plus the increasingly focused JISC collections of datasets, an idea began to emerge and coalesce. This was referred to as the Distributed National Electronic Resource (DNER), for want of a better name: it still is! To me, the idea of the DNER is simple: consistent access to the widest range of stuff.
To make the DNER happen requires a new team with a new approach. The construction task is led by Lorcan Dempsey (amongst a daunting array of other responsibilities); his ideas formed the basis of much of the design of the DNER. My 5 years as eLib Programme Director was the best job of my life (so far), and I sincerely hope Lorcan will feel his job as exciting and satisfying. We all owe him and his colleagues our full support.
Finally, in this brief tour which leaves so much more out than I would wish, a personal favourite has always been Ariadne, which has succeeded in entertaining while informing us for the past 5 years or so. Long may she continue to do so!
- The Follett Report: http://www.ukoln.ac.uk/services/papers/follett/report/
- The eLib Programme Web site, which gives access to individual project pages: http://www.ukoln.ac.uk/services/elib/
- CEDARS project: http://www.leeds.ac.uk/cedars/index.htm
- HERON: http://www.heron.ac.uk/
- Joint Information Systems Committee Web pages: http://www.jisc.ac.uk
- D-Lib Magazine briefing on eLib, December 1995:http://www.dlib.org/dlib/december95/briefings/12uk.html
- Article on Hybrid Libraries in D-Lib, July 1998: http://www.dlib.org/dlib/july98/rusbridge/07rusbridge.html