Go to National Library of Australia Home PageNational Library of Australia Online

SEARCH:

Keynote address presented by Michael Alexander, British Library Computing & Telecommunications, at the 2nd National Preservation Office Conference: Multimedia Preservation - Capturing the Rainbow, in Brisbane, 28-30 November 1995.
  • Introduction
  • The UK Experience
    • Digitisation of Microfilm
    • The Electronic Beowulf
    • Electronic Photographic Viewing
    • Excalibur PixTex/EFS
  • International Activities
  • Conclusion

Introduction

I would like to start by thanking Jan Lyall and the National Preservation Office for inviting me to participate in this extremely interesting conference. It is clear to me that not only are there many very interesting 'digital' projects underway in this country but also that the issues and concerns which this technology and its use raises for libraries are being addressed in a mature and thoughtful way, which is not always readily apparent to me in other parts of the World. Frankly I would like to see more of that intellectual debate taking place in the UK and in the European context than has I think occurred to date on the challenges we face. But I am confident that there will be more of that when the British National Preservation Office hosts its conference on this very theme next year. Also perhaps the UK meeting to discuss a response to the US paper on digital archiving, which Nancy Elkington mentioned on Tuesday, will be the start of that process.

Talking of challenges I must say that I find myself facing a few personal challenges in giving this talk on the last day of the conference. I find that my few meagre points have already been made repeatedly and more eloquently by many of the previous speakers. So, despite my earlier comments, I intend to eschew philosophising or even looking too far into the future. I will certainly not look as far as the year 3000, which is a spectre which Maggie Exon raised yesterday-quite rightly I think, particularly for those of us in institutions whose collections already extend back the other way over twice as long a time period. And if she didn't actually lay that 'ghost', she did, if I understood her aright somewhat calm the spirit by invoking time itself as a deciding factor in what survives and what does not. I am therefore simply going to enumerate some of the initiatives occurring in the UK and elsewhere and from these I will try to draw out some relevant strands for your consideration.

Today we live in 'The Digital Age'. I know this, because everywhere I turn, the television, the radio, newspapers and magazines, and now even Hollywood, tell me it is so. The statement seems to be taking on the status of a truism. However, I should say that for many of my curatorial colleagues within The British Library, who are continuing to battle with the challenge of material printed on nineteenth-century woodpulp or trying to collate and catalogue acres of manuscript material for which there is constant demand, this digital age does not seem readily apparent. Nor does it seem too apparent to many readers in the Round Reading Room in Bloomsbury. Yes, they certainly seem to like using our Online Public Access Catalogue or OPAC (anything is an improvement on microfiche catalogues) but many still spend their time poring over the guard books of the laid-down catalogue; and when they order their material it arrives as printed books and journals. Even those remote users of our Document Supply Service, many of them eminent research scientists and technologists around the World who use computers as constant tools, all of whom moreover would be more than happy to avail themselves of the facilities we are putting in place to deliver their documents to them electronically, would still seem to prefer to print off a copy which they can read and annotate at their own convenience. So how digital are we?

We have heard in the past two days that librarians are pursuing digital projects of their own and even those who are not are increasingly aware of the digital imperative which we are facing from the outside world. One straw in the wind is that the computer is being discovered by humanities scholars not just as an elaborate electric typewriter but as an important research tool. I think this will have a significant impact on libraries. Until recently the major use of computing in academic circles resided with the scientists and technologists-but they use it in their laboratories where they do their primary research. For them the library is very largely a source of published information in the form of journal articles. For the humanities researchers, the library is their laboratory; it is where they do their primary research. If they are now to come to the view that computers will add significant value to their research, they will expect the librarian and archivist to cooperate with them in making this possible and I will touch on this again later when I look at some of the British Library's projects. Humanities scholars can be paradoxical in this regard. They certainly share the concerns of conservators, curators and archivists in wishing to maintain to the highest standards the primary source material which is in their care. But the scholars also want access. This is the raw data on which so much of their life's work may depend. How is this paradox of demand versus conservation to be resolved?

The UK Experience

In the past three years two important initiatives in UK librarianship have occurred which have some bearing on this issue. In 1993 the Higher Education Joint Funding Council issued the report of a review, chaired by Sir Brian Follett, vice-chancellor of the University of Warwick, into the future of funding of university libraries. One of its most significant recommendations was for substantial investment in digital and networking technology within university libraries and this recommendation has been adopted by the Funding Council. This has resulted in the implementation of a program which was initially called the FIGIT program, standing for Follett Implementation Group on IT, but has now been more prosaically named the Electronic Library Program, or 'eLIB'. It does have guaranteed funding of some 20 million pounds over the next five years, so it has sufficient clout to support its future. This ambitious programme has among its major goals the establishment of an on-line journal service of 1 200 titles. I think Mike Lesk made reference to the fact that, by putting themselves into a caucus in this way, the university librarians can exercise a very impressive amount of pressure on the academic publishers to assist them in this task. A national image database service, an arts and humanities data service and a national archives server are also planned [1].

I have to say that the most significant driver for this activity is the perceived difficulty of providing in a traditional manner adequate library facilities at both undergraduate and postgraduate level to the ever-expanding numbers within UK higher education. The ongoing cost of providing, maintaining and filling library storage facilities in multiple sites across the UK is becoming insupportable. As Michael Lesk said yesterday, the relative cost benefit of storing digital information as opposed to maintaining and expanding library stacks is high. The aim of the 'eLIB' programme is to provide a much wider access to a range of digital learning materials, tailored or packaged where appropriate to individual course requirements. The burden of maintaining original source materials which underpin this kind of distributed information provision can then be shared more equitably across the whole Higher Education library sector. The whole program is being carried forward through a series of some 30 project initiatives organised by consortia of university libraries and external partners-for example, publishers and ourselves at the British Library.

Interestingly, in view of the hierarchy of material types which Maggie Exon showed us yesterday, there does not appear at this stage to be any significant concentration on authoring the new teaching materials in digital form, which in that list of hers would be near the bottom. Instead, the aim seems to be on creating a corpus of texts and images from within the existing collections-the exception being the on-line journal project-which would at a later stage be available to teachers and others to create the new teaching resources. Perhaps this might be taken as some confirmation that primary source materials still remain the library's chief concern, even in a digital environment.

Also in 1993 The British Library published a policy document entitled 'For Scholarship Research and Innovation'. It laid out the Library's strategic objectives for the next decade, taking account of its imminent move to a new central building in London's St Pancras and of the increasing impact which the use of digital and networking technologies for information provision is likely to have upon its key constituencies of researchers in academia and industry [2].

Now for an audience such as this I doubt that the British Library (BL) needs much introduction. It is one of the world's great treasure houses of information from every age and culture, containing over eighteen million printed or written volumes and substantial collections of non-book materials: photographs, stamps, recorded sound, prints and drawings. As I suggested earlier, the tasks of organising, conserving and making this material accessible to our users are continuous and continuing and the Library is always looking for new ways to support and enhance these activities. It is from this position that two and half years ago, the Library began an IT programme entitled 'Initiatives for Access' (IFA). This programme contains a series of projects across our collections; projects which are largely, but not exclusively, pilot format in appearance. The object of the programme has been to examine and assess the impact of new digital and networking technologies on some very varied materials, on the BL's operational practices and ultimately on the services which we can offer to our users.

I will briefly outline the more important of these projects to offer you a flavour of what this programme has been about. Fuller details can be obtained from the British Library's Web server, Portico [3].

Digitisation of Microfilm

The Library has a large collection of microfilm. Much of it is copies of original printed and manuscript material offered to readers as a surrogate research tool to reduce wear and tear on originals. This project has examined the use of digital copies either as an alternative or addition to the microfilm copies. Microfilm has been scanned using a Mekel microfilm scanner to create bi-tonal TIFF images at 300 dpi compressed using CCITT Group IV fax compression. This yields an average image size of 100 kilobytes. The project has concentrated on scanning microfilmed newspaper material; in particular material from the Library's Burney collection of 17th and 18th Century newspapers, concentrating on the French Revolution, a part of the collection which is always very heavily in demand by researchers.

One of the curiosities about the microfilming of the Burney collection is the rather idiosyncratic way that the microfilms have been put together. Not only is the microfilming rather old and poor in places, but the sequences are particularly bizarre. It is extremely difficult for users to follow any sequence either in terms of newspaper titles or dates, in order to find their way about the material. Yet it is a crucial piece of information for historians of the 17th and 18th centuries. Thus, there was a great need to produce a better image of this material and more appropriately indexed access to these long runs of newspapers.

Recently, two other sources of material have also been added to this project. One is project INCIPIT. This is a project, funded by the European Union, which involves the British Library and the national libraries of Holland, Belgium, Portugal and Italy. The project is going to develop a CD-ROM of incunable records, based on the Incunabula Short Title Catalogue (ISTC) which the British Library started in 1980. To each bibliographic record will be added scanned images of pages which are considered to enhance and expand the text record, such as colophons, printers' marks et cetera. This CD-ROM is due to be published in the summer of 1996.

The other project is rather curious: it concerns the scanning and indexing of the Batley News, a local English newspaper. A local consortium of educators, librarians and also the Society of Newspaper Publishers, which is the organisation which represents publishers of local newspapers in England and Wales, is interested in scanning the back numbers of this newspaper and using it as a tool within the community-both inter-community groups and schools-to enable them to increase their local history knowledge and to be exposed to IT through taking part in indexing exercises of this material.

At this point, I am sure the cataloguers and indexers amongst you will be sharpening your knives and I am not going to claim that the end result is going to be a particularly marvellous set of newspaper indexes. It is not the sort of local initiative that one normally associates with the British Library, but we decided to involve ourselves in this because it does not need a great deal of input from us. All we need to do is scan our microfilm collection of the Batley News; it does give us the possible opportunity, I think, to strengthen our involvement with both the public library and the secondary school sectors in the UK, where we perhaps have been less active than we might have been.

The Electronic Beowulf

The Electronic Beowulf is possibly the most well known of the BL's IFA projects. Beowulf is an Old English poem, extant in only one manuscript from the eleventh century. Because of its uniqueness, its literary and linguistic value to scholars is immense. To add to our problems the manuscript was damaged by fire in the early eighteenth century and during its re-binding further parts of the document were obscured. Because of this fragility and uniqueness it is not possible to allow even the most bona fide of scholars to subject the manuscript to any regular detailed examination. The aim of The Electronic Beowulf has been to provide a surrogate which scholars would find more than adequate for their needs. The project has been conducted in cooperation with two of the world's leading Beowulf scholars; Professor Kevin Kiernan from the University of Kentucky and Professor Paul Szarmach from the State University of New York. Images were captured at a variety of resolutions in 24-bit colour using a Kontron ProgRes digital camera. Average file size is between 20-30 megabytes per image, although with JPEG compression this can be reduced to 1-2 megabytes per image.

In addition to a straight facsimile quality scan of each folio a variety of 'forensic' scans were carried out. Using small fibre optic cables and ultra-violet light it has been possible to expose parts of the manuscript previously hidden either through age, fire-damage or the effect of the post-fire re-binding.

This is an example of the sort of cooperation between libraries and humanities scholars which can be fruitful: substantial value has been added to the original document. Not only are the facsimile images far superior to anything that scholars currently have to work with, but the forensic scans have contributed considerable amounts of information concerning the form and production of the manuscript. The drawback of this approach, apart from the high cost of buying the equipment to carry out the work, is that the task itself is very resource intensive. In order to carry out the project successfully there needed to be a high input from curatorial, conservation and photographic staff within the British Library. Therefore, in order to carry out this kind of a project in the future, I think the British Library or any such library that wants to undertake it needs to be certain that it is going to produce a substantial enrichment to the body of scholastic knowledge before it can afford to undertake the work.

Electronic Photographic Viewing

I mentioned earlier that the BL has a large amount of material whose primary information is contained in visual rather than textual format; this includes photographs, illuminated manuscripts, oriental miniature paintings, postage stamps and a variety of other prints or drawings.

Two approaches were taken with this material. One has been to capture a variety of such material digitally using the Kodak Photo-CD technique and viewing has been enabled through a database retrieval system which has been developed for the BL by iBase Image Systems in Ilkeley, Yorkshire. In addition to the use of Photo-CD a number of items regarded as visual 'treasures' by the Library had been previously photographed on high-quality ektachrome film for the production of printed facsimiles. Included are such documents as The Lindisfarne Gospels, The Sforza Book of Hours, The Leonardo Notebooks and The Diamond Sutra. These 5x4 transparencies have been scanned on a Scitex scanner to produce 24 bit TIFF colour images, roughly 34-40 megabytes per image size. The images of these 'treasures' will eventually be on display through public browsing kiosks in our new building, alongside the original documents in their display cases. They will form a 'Turning the Pages' facility which will enable visitors to the Treasures Gallery to view the complete document in digital form instead of only the single page opened for that day.

Excalibur PixTex/EFS

Indexes are literally the keys to successful access tto digital material but attempts to produce a detailed index to digital text images can require a very high input from qualified staff; an expensive solution. A cheaper option may be to use optical character recognition (OCR) software to convert the textual images to raw text and then apply such automated text retrieval techniques such as inverted index files to this material. Unfortunately the relative inefficiency of OCR can still mean a substantial input of staff time to edit the text into an accurate form for automatic indexing. PixTex/EFS, a software product from Excalibur Technologies Inc., provides a way around this dilemma by its use of Adaptive Pattern Recognition technology. Neural networks and fuzzy matching provide links to the search terms within the un-edited text. The text is linked to the original scanned image and it is the latter which can be displayed to the user to read as a 'clean' document. This project has concentrated on scanning the sort of printed material to which researchers would like the more dynamic access offered by computer indexes, but for which the Library is unlikely to have the resources required to convert this material to a completely clean machine-readable form in the immediate future. This includes items such as the Printed Catalogue of Seals and the Printed Catalogue of Music Manuscripts. A range of printed documents for use by historians, such as calendars and ancient university matriculations and alumni lists have also been scanned. They are currently available to historians at the University of Sheffield and the efficacy of the application is also the subject of research by the University's Information Studies Department. Further details of the 'Initiatives for Access' program are available from the British Library's Web server, Portico or in our newsletter.

The British Library's Chairman, Sir Anthony Kenny, has established a working party of the five British legal deposit libraries and also the British Film Institute to work towards submitting a proposal to the British Parliament for an extension of legal deposit to non-book materials. They are attempting to work closely with publishers and other organisations, which are both collectors and providers of such information. All the sorts of concerns and issues which I've heard mentioned in the past two days on the question of legal deposit remain 'on the table' for this working party. These are early days yet, so I'm afraid I don't come to you with any conclusions or even suggestions. I would merely like to draw your attention to the fact that in Britain we are attempting to begin to address the problem.

With 'eLIB' and 'Initiatives for Access' the primary aim has been to use digital technology to enable far wider access to library collections, both in traditional reading room environments and remotely by networks or optical media-in other words, the production of digital surrogates. This has inevitably raised the question of the quality of such digital surrogates. Certainly within the BL the general view has been taken that image quality should be both 'fit for purpose' and should be as far as possible 'future proof'. If such aims are realisable then digitisation does take on a strong preservative function also, if only in strengthening the case of librarians and curators for significant reductions in user access to the original materials.

But the question remains: what do 'fit for purpose' and 'future proof' really mean? Are there as yet any agreed quality standards by which the output of digitisation projects may be judged? There are a number of technical standards available, or perhaps it might be more accurate to say that there are a number of documented technical options available. But even public adherence to these options will by no means guarantee quality. To date within the BL we have had to rely on that old standby 'professional judgement', which given the range of expertise we can call upon in our institution is probably as reliable as anything else currently, given the absence of any agreed international quality standards, and-I like to think-better than some.

International Activities

I would also like to draw your attention to another project in which the British library is heavily involved but which is not part of its IFA programme-using digital technology to combine access with preservation. The International Dunhuang Project links scholars, librarians and conservators in Beijing, London, Paris and St Petersburg in the task of cataloguing, conserving, digitising and making accessible one of the finest collections of written and printed Buddhist texts in the world. Discovered in caves in Chinese Central Asia at the beginning of this century and then dispersed to the four cities previously mentioned, they contain a treasury not only for scholars of Buddhism and historians of China but also for those studying the history of paper making and printing. The collection contains 'The Diamond Sutra', in The British Library, which is the world's oldest surviving printed book from the Ninth Century AD. Not only will digitisation be able to bring the entire collection together in a virtual way but it will enable the computer technology to assist in the reassembling of many thousands of fragments. Once this process has been completed the source material can be archived under the strictest of controlled environmental conditions and access rights. However as with the IFA projects the question of quality has been left to the subjective judgements of the scholars and curators participating in the project.

In the international arena one organisation which could have a role in the task of promoting quality standards for digitising library archive material is UNESCO. Indeed in 1992 it launched a new programme, 'Memory of the World' to protect and promote world documentary heritage. This programme is also designed to democratise access and ensure a wider dissemination concerns itself with the most endangered hand-written, printed and audio-visual holdings. An International Advisory Committee (IAC) was established with senior representatives from throughout the international library and archive community and has held two meetings, in 1993 and 1995 receptively. A number of projects have been accepted for inclusion in this programme, from Eastern Europe, Spain, Yemen and Turkey, where extremely precious material, of great cultural significance not only for the countries in which the material is currently lodged but also for the wider world, is in grave danger of disappearing completely because of the surrounding social and economic circumstance of its location. A fuller description of these projects can be found in the August issue of the IFLA Journal [4]. At its last meeting in Paris the IAC promulgated its 'General Guidelines to Safeguard Documentary Heritage', to be used as the controlling document for the programme. It covers the scope and management of projects; the establishment of registers of collections; criteria and methods for selection; issues of preservation, access, distribution and education. The document is almost entirely the work of Jan Lyall, assisted by colleagues from the National Library of Australia and Australian Heritage Projects and I think she and they deserve warm congratulations on producing such an impressive and useful document.

Within the appendices to this document will be found a report from the Technical Sub-committee of the IAC. Chaired by Dietrich Schüller, Director of the Phonogrammarchiv of Vienna, it contains representatives from IFLA, ICA and the International Federations of Film, Sound and Television Archives. Its task has been to recommended the standards to be adopted for ensuring the long-term preservation of and access to 'Memory of the World' material. Taking into account the wide-ranging definition of document as occurring across all types of media, and the apparently conflicting demands of maximising access to the endangered material while promoting the greatest possible protection for the documents, it has been proposed that a programme of digitisation of these documents is the best solution. (And here I must declare an interest-I am the IFLA representative on this sub-committee.) However, in the first draft of our recommendations we did not feel able to do more than provide minimum standards of digital capture which would enable sufficient access to the material using existing commonly available forms of IT. The 'fit for purpose' and 'future proofing' quality standards which we have struggled with in the BL remain an issue for further discussion when the sub-committee meets again, hopefully in March, next year. Fortunately, since the last meeting almost a year ago, the amount of information about what judgements can be made in this area has increased and it may be possible to make recommendations on the basis of best practice from within existing digitisation projects around the world.

One country in the World which has never been backward in applying new technology to solve problems is, of course, the United States. We have already been fortunate at this conference to hear from two of the key American protagonists in confronting these challenges and I would counsel everyone to pay close attention to their experience and advice. The US also has some key national organisations in the library and archive professions which are already playing a prominent, I would say crucial role, in defining and integrating standards for building and maintaining digital collections. I would cite the work of The Commission on Preservation and Access (CPA), the Research Libraries Group (RLG) and the Association of Research Libraries (ARL) as perhaps the most crucial in this area.

The number of different digital projects concerning libraries and archives currently on stream or planned to begin shortly appears legion so I will just mention two or three initiatives from the many worth considering.

Perhaps the most well-known digital preservation project in the US has been the CLASS project at Cornell University, co-sponsored with Rank Xerox and the Commission on Preservation and Access. This project has been tackling the problem of 19th century brittle books by providing a quality page image for the reader and a quality reprint on acid-free paper. Anne Kenney, Associate Director of Preservation at Cornell has written extensively about the experiences gained from this project [5]. The project has found that digitisation can be competitive in cost with both preservation photocopying and microfilming; and it will produce images as good, if not better. Kenney has attempted to draw out from these some quality standards which could be applied to all digitisation projects; standards which would produce images to enable access without elaborate computer equipment but which would be images of preservation quality. I think that these proposals are worthy of serious consideration by all who are concerned both for access quality and archive quality.

Another project of interest is HELIOS (Heinz Electronic Library Interactive System) which commenced just over a year ago at Carnegie-Mellon University (CMU) [6]. It is converting over one million pages of the papers of a late political scion of the Heinz family. The process of capturing the digital images of the pages is not extraordinary in any way. But in order to index this data effectively the project is applying OCR technology to the scanned images. In this it resembles the BL's own investigations of the EFS software but it takes the whole automatic process a stage further. Using a software system called CLARIT, produced by the Claritech Corp., itself a CMU spin-off company, the project is using a tool to perform intelligent post-OCR correction on the text data. The need for human editorial intervention on the text by archive staff is not completely dispensed with but is reduced considerably by this process. Furthermore the retrieval engine uses natural language processing (NLP) techniques taken from work on artificial intelligence and linguistics. Indexing is based around attempts to extract concepts which characterise the document content rather than rely simply upon the occurrence of word strings in the text. The argument runs that whereas in STM subjects there is often a rigorous use of terminology which permits word-based systems to operate effectively on texts, in this area no such discipline applies to documents within most historical and archival material and a more concept-based approach needs to be applied. That view is borne out anecdotally for me from the work we have been doing in the EFS project on historical materials where the breadth of word use within the document database can lead to unacceptable imprecision in the retrieval. I have not yet had an opportunity to read the papers which the project cites as evidence for its success but I intend to do so urgently. If these do justify CMU's claim to have built 'a better mousetrap', then this would be a significant breakthrough for indexing digitised text archives.

In discussing both issues of technical and quality standards and methods of image retrieval I would also like to mention the work of Howard Besser, Visiting Associate Professor at the University of Michigan's School of Information and Library Studies and Jennifer Trant, Manager of the Getty Art History Information Program's Imaging Initiative [7]. Clearly standards for capture, storage and transmission of images are of vital importance for the long-term viability of image archives but so too are accurate and valid descriptions of the image files themselves; descriptions of the circumstances and the technologies which have been used in the creation of that image. Without this information it may be difficult for future users of, particularly pictorial images, to make worthwhile comparisons between images or understand exactly what that image represents from the original source. This area is of great importance particularly to those who are contemplating the creation and use of pictorial digital image stores. I would urge you to familiarise yourself with Besser and Trant's work and to participate if you can in the discussions on this matter. The work is being carried forward under the auspices of Getty AHIP and the Coalition for Networked Information (CNI).

Conclusion

The set of questions surrounding access quality and archive quality, technical standards and quality standards seems to me to continue to go round and round amongst those of us involved in these issues, within our projects and programmes and at seminars and conferences, without any clear resolution. It may well be of course, because this is still 'a period of transition from established analogue technologies to . . . rapidly changing digital technologies' [8] as the CLASS project identified. Those involved in tackling these issues are still relying on gut feeling, politely called 'professional instinct'. As I said earlier, if the professionals involved are of a high calibre this may not result in anything bad, but of course it may not coincide with the 'professional instinct' of those in a neighbouring institution.

In conclusion, the key word here for wide access is inter-operability. In IT terms this means conforming to the same sets of standards, and in the British Library we have tried to travel along that route. Earlier in our 'Initiatives for Access' programme, we produced a substantial technical standards guide to indicate the range of standards that we were prepared to support. We are in the process of updating the guide now. The 'eLib' programme is also drawing up a similar guide and we have agreed to cooperate on the production of a joint technical standards guide, which is a sensible move when UK university libraries and the British Library are very likely to be sharing digital data before very long. Computing areas of the British Library and the Library of Congress have also agreed to work together on standards definitions, though we have not yet defined the manner in which this partnership will be conducted.

Nancy Elkington talked in some detail about the establishment of the National Digital Library Federation. I think that this could possibly be the most important development in digitising activity within libraries that has happened recently. The Federation is going to establish a collaborative management structure and a co-ordinated funding strategy; it will also adopt common standards and best practices to ensure full information capture, availability and enduring access.

This is a bold vision containing some tough challenges, but I think it does begin to provide a forum in which these challenges can be addressed and a forum, moreover, composed of institutions which have a good recent record of commitment to facing and overcoming these challenges. As the eventual aim of this proposal is global Internet access and as the Federation's mission is to obtain the material for its collections from across the nation and beyond, then I hope that this forum can become an international one where librarians, archivists, conservators and information technologists can meet and define the technical and the quality standards which will enable us to digitise, to preserve and provide access that is truly global, durable and beneficial for our users.

Joint Funding Council's Libraries Funding Review Group:Report. Bristol:HEFCE,1993. Available at http://www.ukoln.ac.uk/services/papers/follett/report/.

  1. For fuller details of the programme see: http://www.ukoln.ac.uk/
  2. For scholarship, research and innovation: The British Library's Strategic Objectives for the year 2000. The British Library:1993. http://minos.bl.uk/diglib/access/strategic-objectives.html
  3. A. Abid Memory of the World - Preserving the Documentary Heritage. IFLA JOURNAL 21(3) August 1995 pp 169-174
  4. e.g. Kenney & Personius Joint Study in Digital Preservation CPA,1992 and Kenney & Conway in Digital Imaging Technology for Preservation RLG,1995
  5. Galloway, E.A. & Michalek, G.V. The Heinz Electronic Library Interactive Online System (HELIOS):Building a digital archive using OCR and Natural Language Processing Technologies. The Public-Access Computer Systems Review. 6(4), 1995.
  6. http://www.ahip.getty.edu/intro_imaging/home.html
  7. Kenney & Personius ibid


Return to 1995 NPO Conference

Last updated 28 July 1998