What I intend to do this morning is to use my time sharing with you some thoughts about the digital world that we are in, or we are getting further into. Also doing that on the background, which is the Nordic Scene, thus the title: "The Nordic Lights".

We hope, coming from this part of the world, that we have developed something here that we could share with all of you.

I will start by telling you a little story, I like doing that, for I am from that part of the world in northern Norway where we do that. I'll come back to the story at the end of my presentation, and if you are lucky we all will know what I meant with it. But first of all, I mean, on a morning like this after a good meal, a nice evening last night with our Minister of Culture, it comes to me, some words from Ecclesiastics 31:36, where you can read as follows: "Wine drunken with moderation is the joy of the soul and the heart". Think of that. Benjamin Franklin, he also said: "Wine is a constant proof that God loves us, and loves to see us happy". Just wanted to mention that to make you feel comfortable.

My story is about this little old lady who lives not far from where I have my office, here in Oslo. Twice a week she will go to the local bakery shop and she will buy a newly baked loaf of bread and we will bring it home, and she will sit down and she will have her breakfast. The only problem is that when she gets home, she still has half a loaf of the bread she bought three days ago, and being the kind of person she is, you know, she wouldn't dream of throwing that away. She will start eating on that old loaf of bread and save the one she just bought for later use, when she's finished with the old one. And so it goes on and on and on, so you know, she never will get that newly baked bread.

Anyway, I will start by sharing with you some thoughts about what I've called "The Gutenberg Galaxy", the kind of world we live in and that we are perhaps moving away from: I'll be giving you some examples of how we do things in the Nordic Countries, and trying to map out some of the developments in the world of Libraries, in the world of Archives and museums, collections at large, that I think is important in the work that we are doing with the Memory of the World Programme. I will also dive into what I think is perhaps the most important task to deal with if you are to succede in making the Memory of the World Programme come true. And as I said, in the end I will try to figure out this about the little old lady.

You all know that we have had writing or printing languages for a good number of years. We don't know for how long, but we can trace it back at least 5000 years, and you saw some examples of that yesterday at the exhibition. If we walk through the centuries, we will find that the printing, the writing, took many shapes and forms. It would be on clay tablets, it would be on glass, on metal, on papyri as we saw, vellum and then paper, and then - some 500 yeas ago - we had this new invention. We think we invented it here in Europe - but we really didn't. The Chinese, the Koreans and the Japanese had been using moveable types for a good number of years before Johan Gutenberg started doing it here in Europe.

Nevertheless, let's for the sake of the argument say that it started here in Europe 500 years ago, and this really brought the whole perception of our world down on paper; print on paper. So for almost 500 years, this has been very much the way we have perceived the world. At the end of the last century and the beginning of this century, we can see that this single line that came from the printing press of Johan Gutenberg starts to divide, to split up. We get the possibilities of taking pictures, images, photography - and this moved on into our century, into making films; moving pictures. That's a new way of recording reality and it goes on into broadcasting and television and along this we also are given the possibilities of recording songs, and we can play them again, like gramophone records. We have broadcasting and we have the radio. So we see that in this century, where we live (and we are getting very close to the end of this century) the one dimension of Johan Gutenberg's print on paper has multiplied into a number of different ways of recording and presenting the world we live in.

I mention this because I think it is important; and you have noted from the introductions we had yesterday, that we are dealing with a very wide variety of information. It is not as simple as was just print on paper. And we are gathered here as specialists, who in many ways know far too little about each other. For example, the people dealing with sound archives, people coming from broadcasting, they - well, not to offend them - but I think that they don't know as much about restoring and conservation of manuscripts as some of the others here would do. And we know too, we need to cover all of this, this is extremely important. If you look at the guidelines for the Memory of the World Programme of UNESCO, you will find that it is broad, it is covering all these different aspects.

I personally picture myself as living actually in what I call the "Gutenberg Galaxy". And I have realised that out there is a vast universe of knowledge that would be brought to me on information carriers which are not the print on paper ones. It is the sound recordings, it is the images and, low-and-behold, at the end of this very century, at the start of a new millennium, we are given - as the Minister of culture mentioned the other day - a very powerful tool whereby we can bring together all these different types of information carriers and store them in one single medium; the electronic storage of information. And the beautiful thing about this, and the scary thing about this, is that stored in this way, made accessible in this way, we can send it around the world, fast - like that - a number of times. We can send it from Oslo to Japan, from Osaka to South Africa - wherever we have the necessary equipment to bring it. And from these digital impulses we can bring up pictures on the screen or we can have songs coming out, or we can have this printed out as text on paper.

This gives enormous opportunities, but as was mentioned yesterday, we have witnessed for the last five to ten years, a widening gap between the possibilities opened up by the new technologies and the existing legislation in the field. Legislation should handle this, by making information accessible, but should also protect copyright holders, the owners of the intellectual property, and ensure that they get a reasonable compensation. I will get back to this point, it's very important.

So we are in the "Gutenberg Galaxy" - and we have learned that there is a vast information universe out there and we would like to explore that even further, and we are given the tools. This is the background.

So - what could we offer from this part of the world ? From far North ? Well again - as we have learned from the presentations of the Memory of the World Programme, so many of these tasks are too big to be handled by one state, one nation, alone. Very often it is a question of finding allies, finding friends who would be willing and able to help you solve the problem that you are facing. We like to think of ourselves in the Nordic Countries, as having developed certain ways of cooperation that we wouldn't mind sharing with others. I know that this is something one should be very hesitant in doing, for what serves the purpose in this part of the world would not necessarily serve a purpose in another part of the world. But, betting back again to the speech by the Minister of Culture yesterday, she mentioned this book by John Kenneth Galbraith about the "Good Society" - where he says that "No single country can act effectively and alone, not in this world and in these days". And based on that I think that we have developed some good cooperation in the Nordic countries. We have an organization like an umbrella over the Nordic special academic and research libraries called NORDINFO (the Nordic Council for Scientific Information). Over the years (and this could be almost 25 years) they have developed ways to assist and to take initiatives which would make the cooperation between these libraries easier. But, most of all, to make the collections held by these institutions more accessible for the users.

I will not go into all the developments that NORDINFO has initiated over these years, but just mention one of the latest. You will hear more of this later today and tomorrow. They have set down what they call "Three Centers of Excellence" - all of them dealing with the fact that we have entered a digital world. They have created a center for Networked Information Services, another one for Electronic Publishing and a third one for Digital Handling of national library collections. And in doing so (this is something I have been watching closely for the last years) it had been evident that not only is there a cooperation between the libraries and the research libraries (also the public libraries, I should add, but that's another organization), but you'll find the same in the field of museums and in the field of archives. Every field that is given the task to take care of collections important for education and research will be affected by this.

We had a conference three years ago where we raised the question "What should be the preferred future for Libraries ?" We were discussing with politicians, decision makers, people from outside the universities, outside the libraries as well as users inside the libraries and the administrators of these institutions, to try to find the common denominators in all this.

And again, it became clear that even thought we were talking of libraries, so many people were providing input into this, pointing to the fact that this as much had to do with archives, museums and collections at large. That was a lesson to be learnt.

NORDINFO also operates into Europe, they have a special obligation given to them by the Nordic Council of Ministers to look after that, and in particular, they have been looking at ways of involving our closest neighbours up in the north, the Baltic States, and finding ways of cooperating with them. And from that work, I've also seen that it really is a good thing to join forces, particularly when it comes to handling all these different types of new information carriers.

In Norway, in many ways, we are lucky in the sense that we have quite recent legislation for not just libraries, or the National Library, the legal deposit, we have also a new legislation for the Archives and there is a good movement on for the museums. What is interesting and what I will be hammering on, all my time here, is that the kind of material that we deal with in the National Library, as they do in the National Archive (based on the legislation) is getting together because of this new storage medium. And we have much more to gain from cooperating and working together than from trying to find out in which ways we differ.

I say this now because in my years working in the field of Libraries and international organizations I have been appalled from time to time watching these organizations trying to guard their turf, and not letting the others enter, by saying that "This is, you know, far too special, this is something that only we can handle". I tell you - "Not so any longer". This new medium that has been developed, the possibility to take these collections, whether they are items in a Museum, books in a Library or papers and documents in an Archive, they could all be digitized, brought onto the form that you saw demonstrated the other day. And this is new.

Well, somebody will say - "We have different types of institutions, these are organized in different ways throughout the world, I mean we can't really pretend that history didn't happen". True, but I would think that seen from the point of view of the users, they don't care whether this would be collected in an archive or in a library, or wherever it is. They want to have quick, easy access, sort of a one-point access to all this. How do they get that ? By a corporation of people like ourselves - sitting here.

I have mentioned legal deposit; this goes for libraries. There are other forms of legislation for other types of collections. What they have in common is that normally, throughout the world, there will be some sort of legislation taking care of this. Now this is important, and also it is important that this legislation be linked with the kind of rights I mentioned earlier that has to do with copyright and intellectual property.

I know that many of my colleagues are extremely concerned about the accessibility of material, that the collections in a Library or in an archive should be made accessible, and we tend to look at the copyright holders as a nuisance. They will guard the material, they won't let it free. We have these interesting conversations with the Norwegian Broadcasting Corporation, trying to make their material more accessible to the public - and they hesitate because they know this is going to be difficult. I think that this is one of the major tasks we will have to face and find good solutions to and I do think that again, we should do this together, with other types of collections, when I look at this from the library side.

And I know very well that if we don't do this in a sensible, good way, then the generations after us - if we are not collecting material like songs, like images, videograms, broadcasts - they could rightly accuse us of not having been doing our job properly. It is well and good to talk about the virtual library or virtual reality, yet as we heard yesterday, print on paper remains dominant and still they have problems in recording musical scores in China and throughout the world. Still they have problems in keeping proper archives all over the world, and so we could go on. But the possibility is there, and we will have to find a way to organise the legislation.

Into this reality comes the programme that is named The Memory of the World. As was pointed out by Mr Abid from UNESCO, the Memory of the World Programme is not primarily a programme for making CD Roms, it is a programme that is mounted to take care of collections all over the world; important collections for the memory of the world, and this could be done in many different ways - in traditional ways, if that is the word to use. But, obviously the potential, the opportunity that opens up by using the new electronic medium is enormous and also a challenge. So in working with the Memory of the World, we have on the one hand an enormous time span, 5000 years back - and we want to bring this to the attention of people. Different types of material exist and we will find ways to handle this, which we do, each and every one specialist group. We have different types of information carriers and I see the Memory of the World Programme as one way of bringing together so much expertise in such a wide area, knowing that this is now being brought together for the benefit of the users. It shall be easier to have access to these materials and at the same time we will be saving the originals against unnecessary use.

There is another dimension to this, I've seen this in my own Library, I've seen it when people are approaching items as we saw in the Martin Schøyen collection the other day. Not just a librarian, or a national librarian, but young people whom I address in schools and who come to the library - and when you talk to them, and they realise that this very item is 5000 years old, or this book is the book that really laid the foundation for the Kingdom of Norway or the Declaration of Independence. If you gather a group of people, somehow it touches them, it's something that they feel they may not know all there is to know about, but they know instinctively that this is important. And it gives the good feeling of being close to something meaningful and important, almost sacred. And I think this is one of the forces that we really should grab and take advantage of. Make it known to people that this is not just trees or paper, and that we have in the North of Norway, in the National Library. We have a branch up there, we have a huge hole in the mountain, we have a depository library and we store all the material that we get by legal deposit. Some people will say "Why do you do this, I mean keep it there, why ?" And we will have to find ways to point out that this is important, it is our contribution from Norway to what will be the collective Memory of the World.

Now, there is an interesting question here: "Should we, or would we, be able to bring along everything that is being made public, published, either it be on print, on paper or through the radio waves or television ?" Some people, mainly historians, will say "Yes" - they would like it all. I would say no. There is no way.

Ten years ago I presented a paper at IFLA which I gave, I thought, a very good title. I called it "The end of all and forever". Meaning that it's not so any longer; we cannot collect everything and we will not be able to keep it forever. This is also an important lesson that we should take up here.

Firstly, this is the part of the Memory of the World Programme where we are identifying collections throughout the world that are in need of urgent action, so that they are not vanishing.

Secondly, we have the problems that we face today, and should collect contemporary material and store it in such a way that we are not creating a new conservation and preservation problem for colleagues in 50 or 100 years time.

You all know about acid paper. We have found out about that, and there are a lot of things to be said about the new media that is recording and bringing out information. We should also, in the context of the Memory of the World Programme have this in mind: That this is a broad concept that should make us aware, that in addition to saving important collections, we should work as we have always been doing, but perhaps more organised together. Not as different groups, but join forces so that we are not today creating new conservation problems for tomorrow. I think this is important.

And also, when we are talking about the "All and forever, and the end of it" - I read once and I thought that was a nice way of putting it - when you talk of print on paper - that "Books are trees made immortal". I liked that so much that when I talk to publishers in Norway and they are hesitant in bringing me books for the legal deposit and they say "Why do we have to do this ?" I say: "We offer yo immortality, what more could you ask ? This is something that nobody else could give you."

But then again, would we like to offer immortality to all the items being published today ? I think not. Somehow we will have to do a selection and the same thing goes for the Memory of the World Programme. You learned the other day that there have been sent out questionnaires, lists have been made up, we have tried to single out what is worth keeping and restoring and bringing with us and what is not. I think this is important, but then historians will nag me like this and say: "You are censoring this, we want it all, who are you to throw away things, important things, like these posters from this specific political party ?" And then I say: "You know, we will simply have to do this, because we cannot take it all with us." And I would use an example, saying: "What if the Egyptians had insisted on doing this in your way, would we then all be speaking Egyptian today, and walking like they did, and would we have all the material that we today have only fragments of ?" Quite clearly this is not possible. But what we should do is to give the definitions and criterias as to how we do the selections, so this is an open process, and we can have people take part in it.

We have been talking about what should be the criteria for the items that should "qualify for eternity" - both the things that we get from broadcasts, from publishing and so on today, and not least the collections that we are in particular talking about in the Memory of the World Programme. I don't think there is any simple answer to that, and I think what we will end up with, is trying to make some sort of sample that gives a reasonably good presentation of what we are doing today. We may leave out something, and maybe in 200 years time people will say "Now, why didn't they keep that ?" But then again, it should also be an interesting life to be a historian in 200 years time. If we kept it all, you could't do your speculations, you know, the facts would be there.

Which reminds me about this publisher I know; he is also an editor of a newspaper, and he said once on the telephone: "Send all the details, nevert mind the facts." We don't want to do it like that.

So, in this process, keeping in mind the changing world, the changing arena and the environment for handling information in the various collections that we are dealing with, we obviously need guidelines. We have legislation, as I've said, and we have also developed an international instrument called the Memory of the World Programme. I believe in this programme, I think it is important, but as we heard the other day, all programmes have a somewhat slow start. It takes time to make people aware of this, it takes even more time to make people start really believing in this and feeling: "Yes, we can do it - this will make a difference."

We have, as I see it now, the foundation for all this, a basis from where we can work, a platform. What we need now is to seek partners and find good ways of really making this work. And then again, I believe that we cannot do this alone. We need the assistance of the politicians, the decision makers, the people who are looking after the nation, looking after the educational system, the health system etc. Such people never have enough money. But, we must be putting it to them in such a way that they are convinced that on a world scale, on a regional scale, this matter is important.

Finding those alliances will, I think, be the major task in the carrying forward of this programme. You may know that a month from now, or six weeks, in the middle of July, there will be another meeting here in Oslo to follow up this conference and UNESCO intends to establish a marketing Committee that will be looking more closely at the task of finding sponsorships.

How do we find the financing for this programme that we have set up ? There are ways. To give you some examples: If you look around the world today,you find that there is a lot of business going on. Interestingly also, the big 7, the G7, they are talking about the information highway. Information has become a world commodity, it's important. The part of information that we are dealing with here is also important and it's our job to make this more visible, to bring it to the attention of the decicion makers. What we could do, and I hope that this will come out of the marketing efforts of UNESCO, is to bring the lists that have been identifying the different collections worldwide to the attention of the G7 and the big international companies. If you look at an investment of 2 or 3 billion US dollars, it's quite large. You could take half a percentage, or not event that much of it, and say: "Yes, we will use some of this to save some of the world's memory which is located in this area." And if this be a procedure that would be followed up, so much really could be done.

Today, many people will say "You know, that's not possible, those people with money, they don't think like that". Well, look around you, look to the big country across the waters from here, the United States, and you will find that so much of the investment that has been done on the cultural side has been done in such a way. So I don't see it impossible, as being applied on a world wide scale. But then again, we will have to really join forces, develop strategies and find who are our allies, and who should we be approaching. This, I think, would be perhaps the most difficult follow up to all of this.

And also, another word of warning: I find, in my profession - much as I like my colleagues - I sometimes get a bit upset about the way they approach the world outside libraries. Because - for so many years, at least in this part of the world, librarianship and archival collections have been set aside from the hard world of finance, of economy and strugling to keep a business going. So many of these institutions are being financed by the Government and not financed enough, and basically, we don't like that enterprises and businesses should come in and do somehting in this field. There are many reasons for this. Some would fear that business sponsors would have a say in the running of the collection and how it should be developed and so on. Some would hesitate and say: "Don't bring this business stuff into our field." I'm not so sure that this is a good strategy any longer. When I look around the world, I see that this functions very well, And I see in Norway, and I listen to the new gospel being preached by our Government, and they are saying: "This is what you're supposed to do, if you want our blessings and continued support."

In the introduction to a "Memory of the World project - General Guidelines to safeguard documentary heritage" it is written: "The Memory of the World Programme is designed as a new approach to facilitate preservation by the most appropriate techniques of the world's documentry heritage, to enable access to it without discrimination against any users, to increase the awareness worldwide of its existence and significance and the need to preserve it and to promote the Programme and its products to the widest possible public. As soon as it was launched, the Memory of the World Programme began to arouse great interest. Requests for assistance, sometimes even appeals for help, regularly reach UNESCO. It is a daunting task and only the mobilization of all the parties concerned can translate declarations of intent into a vast world workshop to rescue, reproduce and disseminate endangered documentary treasures."

It is to be hoped that this very first international conference on "Memory of the World" here in Oslo, will give a new momentum to the work worldwide and be able to draw on experiences from national and regional efforts like for instance the Nordic ones. And bringing together into more active cooperation the different professions that are taking care of a vast variety of collections worldwide. By using the advantages of the wide range of information technologies, we have been given a wonderful opportunity for a new start and doing this in new ways. It certainly is possible to picture this vast world workshop to rescue, reproduce and disseminate endagerered documentary treasures, as is really the core of the Memory of the World Programme. But this means that the professions involved will have to join forces and also to find alies outside their own fields with whom they can share both the old problems as well as the new opportunities.

So getting back to my opening story about the old lady and her bread. Being alone and not wanting to throw away the old bread, she never seemed to get into the position to taste the newly baked one. But if she had invited some friends and they all had a meal together, and sharing the old problems as well as the new oppontunities, they all could benefit and also getting into a new and better situation to enjoy the tasty and nourishing bread from a new recipe. Like the recipe that is now being created within the framework of the Memory of the World Programme.

1 Introduction

The information technology offers a number of new options for dissemination of and access to the 'memory of the world' - i.e. to the archives, libraries and other documentary heritage collections and fonds. Computer-based information systems are built up in order to inform about the documentary heritage, to serve as finding aids for collections and fonds and to give access to machine-readable copies of documents, the original of which being stored on paper or other traditional media. There is a lot of activity in this field throughout the world. And during the past two years the use of the Internet has laid the basis for a total revolution in the dissemination and distribution of such information.

My focus, however, will be on another revolution, which is taking place at the same time, but which does not seem to engage archivists and librarians to the same extent. That is the revolution of how the documentary heritage comes into existence, or using archival terminology, how records are created. Starting with large registers and databases already two to three decades ago, an ever increasing number of records and other information is produced and stored in electronic form without any 'original' or copy on paper. Up to now, most business documents like letters, memos etc., although produced in a word-processing system, have been printed out on paper, and stored as traditional paper records. But from the mid-1990s more and more of such records will be electronic only, i.e. they will be both communicated and stored in machine-readable form and must be handled as electronic records. The increasing use of e-mail in all types of administrations is one of the strongest motivating powers in this development.

For the archival community the impact of this revolution, i.e. the management of electronic records, no doubt, represents the biggest challenge in our time, probably the biggest challenge ever. Archival material is not a collection, built up and organised according to the collector's principles and methods. Archives are fonds of records, created as part of the records creator's business process and for business purposes, reflecting and providing evidence of business activities. In order to provide evidence of activities records must be (and be considered to be) authentic, and they must be kept in an order and in a context that reflects the business process sufficiently. The task of the archivist is to preserve records which are considered to have archival value, in proper order and context, and in an authentic and accessible form over long time, in principle permanently. Considering the properties of electronic information and the complexity of modern IT systems, the preservation of electronic records represents a major challenge both to archival theory and practices.

In this paper I will give a more detailed presentation of problems and challenges related to the preservation and accessibility of electronic records over time, and discuss different approaches in meeting these requirements. My presentation will also include a summary of how the National Archives in Scandinavia approach the challenges of electronic records preservation, and as part of this, programs undertaken for this purpose. According to my profession, my scope will be from an archival point of view. But I feel sure that the issues of preservation and access related to electronic information also will be of relevance to librarians and other information managers.

2 Problems and challenges

The problems of preserving electronic information in accessible form over long time are related to some of the fundamental properties of the information technology. The problems can be summarised as follows:

• The content and the structure of electronic information are represented by symbols (the data) and recorded on media (magnetic or laser-optic devices) that cannot be directly accessed (read) by human beings. In order to retrieve electronic information it must be converted from a machine-readable to a human-readable format, which has to be done by software (a computer program) that 'knows' how the information is recorded - ideally, the same software as the one that stored the information.
• The rapid technological development implies that the hardware and software that was used for storing the information will be obsolete and unavailable after a fairly short time. To serve as the 'memory of the world', electronic information therefore must be accessed and retrieved by other software - i.e. future software, whose properties we know nothing about to-day.
• Electronic information can be altered - either by chance or deliberately - without leaving any traces of the changes that took place.
• Information stored on electronic storage devices (especially magnetic media) is not at all permanent. It will 'fade away' after a fairly short time, depending on the medium, storage temperature, humidity etc. Accordingly, in order to preserve the information over time, it must be copied to new storage devices regularly.

All these problems can be overcome by adequate technical measures and administrative routines. But you have to be aware of them and have the necessary know-how to address them in a proper way. This includes knowledge of the requirements related to preservation of archival information - i.e. archival records.

The requirements of archival records preservation combined with the properties of electronic information and the IT systems represent the following challenges for the archival community:

• Avoid that electronic records which are of archival value, are altered or deleted as part of business procedures (or for other reasons). This implies that appraisal should be carried out at an early stage, ideally at the system design stage.
• Preserve and keep accessible electronic records over a long period of time - in principle unlimited. This implies, first of all, that they must be kept through a number of technological changes and much longer than any electronic storage device can store the data. It also implies that the necessary metadata and other contextual information must be preserved and kept accessible.
• Preserve and retrieve the records in authentic form (i.e. with the integrity of their content and structure) and to establish confidence ('prove') that they are authentic. This can be achieved by administrative routines and by technical control mechanisms. Records which are not authentic, or are considered not to be so, are of little value both as evidence of activities and as information for other reasons.
• Identify, select and appraise records in modern information systems, which can have a very complex structure - for example multi-media systems (text, image, sound etc.), distributed databases and documents, GIS-systems etc. This will probably require changes both in archival theory and practices.

To face these challenges the archival community need to adopt new skills - skills related to the information technology and on a very high level. It is, however, crucial for a successful strategy that the methods, systems and routines which are developed and implemented, are based upon the established fundament of archival theory and best practices, and that the objectives of archival work are still kept in mind. Preservation and retrieval of archival electronic records cannot be carried out properly without the assistance of computer specialists, but the responsibility for the objectives, strategies and for the results rests upon the archivists themselves.

3 Approaches to the preservation of electronic records

How does the archival community approach the management of electronic records? How do the archivists address this challenge which probably is the most complex that their profession has ever faced? I am sorry to say that most archivists do not address it at all! Even for archival institutions (including both National Archives, municipal archives and private archival institutions) it seems to be true that their activities related to electronic records are very limited or non-existing. A survey carried out by the ICA Committee on Electronic Records (of which I am a member) in 1995 indicates that only a small number of National Archives throughout the world have their own program on electronic records. I am pleased to say that the National Archives in Scandinavia are among these.

Nonetheless, there is a growing consciousness among archivists about how the IT revolution in records creation will impact on the archival function, and on archival methods and activities in the future. Different archival institutions are at different stages in their approach to electronic records management - from doing nothing, via discussing and planning, up to working strategies and implemented programs. The different approaches at present and their impact on 'the memory of the world' can be summarised as follows:

• Do nothing. I am afraid that this is still the most common approach to the challenge of electronic records among archivists and even archival institutions. (And I would not be surprised if this was said to be true also for librarians and other managers of the documentary heritage). I don't know the intellectual reasoning behind this approach - probably something like 'not my problem', 'leave it to the next generation which got their PCs as a cradle gift' or simply 'après nous le déluge'.

And the approach will inevitably lead to a 'déluge'. On the one hand, there will be a major loss of archival records and other information being of value as the memory of our time. On the other hand, the information production in our time is so enormous that there probably still will be enough left - at least for some years! But these remnants will be a casual sample, out of their context and accordingly with low or no documentary value - in short, data with little or no value as information and memory of our time.

• The 'academic' approach - i.e. bring the issue up on a high and strictly theoretical level (and keep it there): This seems to be a common approach among those archivists who are trying to address the challenge of electronic records. It has the advantage that it leads to a general discussion of archival theory, an issue that probably will benefit archival methods in the future. It also has the advantage (seemingly) that archivists and other participants can feel that they address the challenge in a serious way, that they are acting to solve the problems, and that their approach is adequate and on a high level.

The disadvantage of this approach is that nothing is done to preserve the records! They are no better off than with the 'do nothing' approach. There is a Norwegian proverb saying: 'The cow dies while the grass is growing'. Related to electronic records one might say that the records die while the archivists are discussing.

Please, don't misunderstand! I do not argue, of course not, that the challenge of electronic records and other electronic information should not be treated on a high theoretical level. Archival theory and best practices are and should be, as mentioned above, the basis for electronic records management, and the theory needs to be adjusted to the new realities. But theoretic discussions in conferences and papers can never be a substitute to the practical measures and activities that are needed to preserve the records. Theoretic discussions should also be based on practical experience in electronic records management.

• The 'technical museum' approach: Some people argue that archival (and other documentary heritage) institutions in the future will have to be enormous technical museums in order to run all types of hardware and software that ever produced electronic records. This approach speaks for itself - considering what it really implies probably no-one would take it seriously. But you still meet the argument now and then, and it is therefore necessary to state that this has never been a serious approach to the preservation of electronic records.
• The archival custody approach - based on software 'independent' information (records) which is copied and migrated to new technical platforms: This approach is the most common among National Archives and other archival institutions which are engaged in electronic records preservation programs, among these the National Archives in Scandinavia. It implies that the records are transferred to the archival institution, stored in a format that is as independent as possible of current software specialities, periodically copied to new storage devices in order to avoid loss of information and sometimes migrated to new technical platforms to avoid technical obsolescence.

This approach has, by experience through several years, proved to be a success for the preservation of records. However, a software 'independent' storage format is not suitable for efficient and flexible information retrieval. Therefore the archival institutions must undertake programs in order to improve accessibility. (See also below on electronic records programs in Scandinavia).

• The distributed custody approach - 'virtual archives': In the theoretic discussion on the preservation of electronic records it has been argued that archival records in electronic form might be more efficiently and economically preserved by the creating entity (the records creator) in the context of its own computer environment. It is assumed that this should be done on the basis of standards set by archivists, and under systematic monitoring by archival institutions to ensure that standards are being met. This approach seems to gain support among archivists, especially those who have no practical experience in handling electronic records.

My opinion is that this approach may be related to specific agencies which are big enough to make it interesting, and which have some more or less permanent functions - for example, the Governmental Map Agency, the Central Bureau of Statistics etc. But as a general approach to the preservation of electronic records it is both naïve and dangerous. Naïve - because it depends upon the willingness of the agencies to give priority to preservation requirements, spend money on migrating records which no longer have any value to the agencies themselves and adjust their systems to standards set up by the archivists regarding both preservation and user services. This does not seem very realistic. Dangerous - because it offers the archivists a new excuse for not addressing the challenge of electronic records preservation.

The most interesting part of this strategy seems to be as a supplement to archival custody, especially related to complex and rather unconventional computer systems.

The preservation of electronic records requires a continual awareness of the archival function throughout the records' life-cycle. Appraisal decisions are to be taken (implicit or deliberately) already at the system design stage. And preservation measures must be considered from the creation stage on. Preserving electronic records over time will largely depend upon the willingness and consciousness of one generation to keep them accessible for the next generations. They are not like paper records which can be rediscovered after some hundred years and still be readable and informative. Consider the following example:

Imagine that in the year 2024, 30 years after the Lillehammer Olympics, there is found an 'old' file in an attic in Lillehammer. The file is entitled 'Correspondence with Mr. Samaranch' and it turns out to contain letters between the big boss of the Olympic Committee and the President of the Lillehammer Olympics, Mr. Heiberg, from 1993 - 94. The issue is the attacks on Mr. Samaranch in Norwegian newspapers, and the Olympic boss is threatening to cancel the Lillehammer Olympics if this activity is not brought to an end. But Mr. Heiberg succeeded in convincing him.

There would probably be no doubt as to the records' authenticity and reliability. But imagine that this correspondence was stored on a diskette, and that this diskette, instead of the paper file, was discovered at the same time in the same attic. Even if the title were written on the outside of the diskette, there would be a lot of questions: Is the information still intact? Can it be read by contemporary equipment? And if it could be read, is it authentic and reliable? Most likely, parts of the information would be lost, access would require special hardware and software - and who would trust the authenticity and reliability of the 'records'?

4 Programs on the preservation of electronic records in Scandinavia

The three National Archives in Scandinavia (Sweden, Denmark and Norway) apply an archival custody approach to the preservation of electronic records, based upon the following principles:

• Electronic records from governmental bodies are to be transferred to the National Archives for preservation, according to the same principles as for traditional records. However, the preservation of electronic records is centralised in the central office of the National Archives.
• Because of the properties of electronic information and the information technology, electronic records (or a copy of them) are transferred to the archives at a much earlier stage of their life-cycle than traditional records.
• Records in the custody of the National Archives are stored in a format (for instance 'flat files') and on storage media that are as independent as possible of specialised software and hardware.
• The preservation is based on administrative routines for periodical copying and migration to new platforms in order to make the records accessible through technological changes.
• The necessary metadata (i.e. system description, including description of the data structure etc.) are to be transferred to the National Archives with the records. Metadata may be transferred on paper or in electronic form.

This strategy has been applied in Scandinavia for more than 20 years, mainly on electronic registers and different types of databases. It has proved to be a success as far as preservation is concerned. But the storage format ('flat files') is not very suitable for efficient and flexible retrieval, and the methods and facilities for accessing the records have so far been rather primitive. However, programs are undertaken to develop methods and software for a more efficient and user-oriented retrieval of electronic records in the National Archives. Another measure to facilitate future access to electronic archival records, is standardisation of recordkeeping systems in the creating agencies.

I shall give a short presentation of three programs on electronic records preservation and accessibility carried out by the National Archives in the Scandinavian countries:

1) The TEAM project ('Tools for Electronic Archives Management'):

This is a common Nordic project (Scandinavia, Finland and Iceland), aiming at a general development of methods and implementation of strategies on electronic records preservation and retrieval.

One main part of the project, carried out by the Norwegian National Archives, is aiming at developing methods for efficient and flexible retrieval of database records. As mentioned above, database records are stored in 'flat files', which are not suitable for efficient processing and retrieval. But the relational information of the database is kept in these files, and by means of the coherent metadata it is therefore possible to restore the database (or parts of it) with its original logical structure.

The TEAM project has developed a prototype of the necessary software for converting database records from a 'flat file' structure into an SQL database. (This process is described in Figure 1.) The software is based on Oracle and seems to be suitable for the purpose, within the limits that were set up for the prototyping. The next step in this project will be to expand the software onto a more general level and implement it as a general tool for the retrieval of database records in the National Archives. (The first TEAM project report will be publishe16this summer with a summary in English).

Figure 1: TEAM project - converting database records from "flat files" to SQL format.

2) The Noark standard for registry and recordkeeping systems:

The Scandinavian governmental and municipal administrations have a long tradition in using registries, where every document to or from the agency, and even some of the internal documents, are registered. In Norway, when computer-based registry systems were introduced in the beginning of the 1980s, the National Archives engaged in the development of a standard for registry systems in governmental agencies. This standard, called Noark, is a set of specifications on the content and the functionality of a registry and recordkeeping system. It was first published in 1984, and has been revised and extended three times since then. (All the reports are in Norwegian only). The next step now, starting later this year, will be to work out specifications for document storage (case file records etc.) and integration with e-mail and work-flow systems, aiming at making Noark a complete records management standard.

On the basis of the Noark specifications, software companies have developed their own Noark systems, and a large number of governmental bodies (more than 500) apply such systems in their recordkeeping to-day. Noark is a de facto standard only, but it plays a completely dominating role in the governmental administration to-day, and it is also entering the municipal administration. It will probably be part of our archival legislation in a year or two.

From a long-time preservation and accessibility point of view the Noark standard is extremely important. Every Noark system, used in governmental bodies, has an export function that periodically will convert the data to a standardised format for transfer to the National Archives and storage. The Noark system of the National Archives has an import function which can read the storage format and, as an automated process, convert the data back to the Noark structure of that specific system. Thus, future users of the National Archives' fonds will have the same level of accessibility to the recordkeeping system of any governmental agency as the officers of the agency had at the stage of records creation. (The procedure is described in Figure 2).

Figure 2: Noark - procedure for preservation and future retrieval

3) Standard formats for document storage:

Up to now, documents which are to serve as records of a governmental or municipal agency (letters, memos, etc.) have had to be printed out on paper in order to meet the requirements of archival legislation and regulations in Scandinavia. The reason for this is that there is no ISO or other standard for document storage which is supported by the word-processing systems that dominate the international market.

However, this attitude is now being changed. In Denmark the government have made a decision, being in force from January 1 1996, which prescribes that also electronic documents may serve as records, and that the National Archives should come up with their requirements and demands on document formats for long-time storage. This has been done, and Denmark has now a standard storage format for documents which are to serve as archival records.

In Norway the approach is somewhat different. We are aiming at coming up with a standard format for document storage, integrated in a recordkeeping system structure, before the governmental bodies are allowed to let electronic documents serve as records. A pilot project, testing document storage in integration with a recordkeeping system (Noark, see above), electronic work-flow, e-mail and digital signatures, was just finished last month and the report will be published later this year. The main conclusion is that there is still some way to go in order to integrate the different systems in a satisfactory way, but that the concept is right.

The most promising format for long-time storage and retrieval of documents seems to be a 'viewer' (called PDF) - that is the common format which the word-processing systems use for printer outputs. The problem is that PDF is not an ISO standard. But it is openly documented and available from all the main word-processing systems. The decision on a standard document storage format for Norwegian governmental bodies will probably have to be taken next year, as part of the next version of the Noark standard.

5 Summing up

The challenges related to the preservation of electronic records are at a high level, and the strategies and activities in this field vary a lot from country to country. Even though the National Archives in Scandinavia feel rather comfortable with their approach to the issue, there are still challenges which are not properly dealt with so far. One of them is to cope with the enormous volume of electronic records, especially those in local administrations. Another one is to address the complexity of modern and future IT systems.

The function of managing the documentary heritage will have to change a lot in the future. So will also the role of archivists, librarians, etc. and the role of documentary heritage institutions. New skills will be required, especially the combination of IT and archival / librarian skills, and the education and training program must reflect these demands.

In order to help archival institutions, and others dealing with archival records, to work out strategies for their electronic records management, the ICA has appointed a committee on electronic records. This committee is working on a Guideline which will be presented in its first version at the ICA congress in Beijing in September this year.

In the digital world - or maybe we should use the expression - in the world of digitisation - we can foresee changes in almost every area of education and research. My angle to the theme will be worldwide access to information in a digitised form and in particular the possibilities of education and research in Third World countries.

The background for my paper is a project initiated by IFLA, which I am in charge of. The main idea of the project is to give libraries and research centres in the Third World access to information and document centres in the industrialised world primarily through the Internet. A few weeks ago we started a trial project in Ghana, and I shall shortly comment on it later in my paper.

A conference like this is very important, because it is a possibility to create an interest in, and a world wide understanding of, digitisation as an important tool in developing programs in the Third World. Several people involved in future research programs have predicted that the information technology can revolutionise the education, research and industrial development in the developing countries. I still remember the impression of the book 'Le De fi mondial' by Jean-Jacques Servan-Schreiber, when I read it in 1980. He described the possibilities of the personal computer even if it only existed as an idea then - and he foresaw the possibilities of using it in education and research in the developing countries. It was the ideas from the discussions in: 'Le Group de Paris' he described. A group of people from Europe, Japan and the Arabic Gulf. Now we can realise those ideas.

In my paper I will start with a few remarks about the importance of establishing access to information and knowledge in Third World countries, and I will mention some barriers we meet. Then I will talk about the possibilities and consequences of the information technology and different kind of solutions. As I mentioned in the title of the paper, I also will talk about the Internet and the perspectives in the use of the Internet and how it can give access to and delivery from the digital information resources. I will give a short description of the IFLA project, I mentioned and of the main idea of the project and its perspectives. Then I will finish my paper with a few remarks about the kinds of developing programs UNESCO and other organisations might support.

Access to information and knowledge in the developing countries

The main theme the conference 'Memory of the World' is primarily dealing with the needs and interest of preserving collections with digitisation as an important tool. I also think that it is important to discuss the access and dissemination of information from the sources of value for the developing countries. This goes for new published digital information as well as old materials scanned from books or journals.

The use of new information technology means great investments regarding equipment, education and training. Is it realistic to believe that poor undeveloped countries can afford this, and become a part of a world wide electronic library? My answer is as follows: If we want to close the gap between north and south, this is the only realistic way. When we talk about the gap between north and south, I think that the most obstinate line is not the income or the wealth of the countries, but the technology, the level of skills and the access to information that separate north from south. The importance is who knows what - and not who owns what. Information is important in all phases of development, education, research and industrial innovation. In other words - access to books, journals and other kinds of information is a necessity. But if we look at the existing library services it is - with very few exceptions - extremely miserable, especially at the institutions of higher education. Their collections are small and often built of gift collections of different kinds. They are greatly lacking of current periodicals, and they cannot keep subscriptions complete, or afford to buy recently published material. To bring their library collections to an adequate standard requires unrealistic amounts of money to bur books, journals, reports etc. Compared to the former mentioned: projects with access to information in an electronic way seem much more realistic. The future libraries - at least when we are talking of the research and the university libraries - are the electronic libraries.

The electronic library accessed though the Internet

Information technology as a tool in information retrieval and information delivery has been used within the past 25 years. Access in different way to online versions of the traditional printed index and abstract volumes was the first phase, and gradually access not only to bibliographic data, but also to full-text became a possibility.

It started with a document delivery service from the database hosts, which offered printed copies from the abstracted journals etc. A great step in direction of access to full-text was taken when the CD-ROM products were introduced at the beginning of the eighties. Suddenly enormous amounts of text were available in digital form at low costs. The technological development created new communication possibilities, new interface and interaction systems, and high speed networks made it realistic to transfer not only digitised text, but also sound, pictures and videos.

Parallel to the former the price/capacity ratio has constantly declined, and the costs of the storing information have now become a less important element in cost calculation. Therefore the amount of digitised information at CD-ROM and high capacity servers accessible from networks is enormous and it is growing exponentially. In 1994 Information Marked Observatory from Luxembourg (IMO) estimated, that thirty-five-thousand electronic documents in full-text were added to the Internet every day. Probably it will be at least twice as much today.

When the next generation of CD-ROM products are introduced in a year or two, we can expect an interesting development in the amount of digitised information. The capacity of the new CD-ROMs will have a capacity that is ten to fifteen times higher than the existing CD-ROM products and they will be more suitable for full-text documents.

It is of course not information in general, but the relevant information, which is of value for libraries and their users in undeveloped countries. We need to establish access to the right information sources and to teach the librarians how to use the electronic systems. We need to discuss how we can create collections of digitised materials in areas which are very important for the developing countries.

I do believe, that the Internet will be helpful in solving the first mentioned problem.

The expectations regarding the global electronic network, the Internet, are enormous. In the library sector we discuss how it will change the whole information structure and how it will create new ways of accessing information. Actually the Internet development started 25 years ago in the US Defense Department, but today the Internet is a worldwide web of interconnected university, business, military and science networks. It is made up of huge wide area networks, of metropolitan networks and local area networks. All linked together in a network of networks.

The Internet gives access to and is used to distribute all kinds of information, letters, magazines, books, journals etc. According to the latest estimations there are now more than forty million individuals and companies connected to the Internet, and it has emerged to of the greatest educational resources ever known in human history. It combines the knowledge of university libraries and research centres in principle all over the world - at least in theory. The truth is that although we are talking about a global network, there is a strong barrier in the actual use of the Internet. Most of the worlds population cannot use computer communication in their own language, but more important barriers are technological infrastructure, costs and competence.

Let us for a moment forget the barriers and look at the possibilities. In principle the resources of the Internet are not much different from the information we find at the shelves in a large research library. The names the 'electronic library' or the 'virtual library' in reality describes the situation very precisely. The Internet gives access to all kind of information and will develop into an Internet library, where the content step by step will supplement and substitute library collections. With the explosive development in the network, practically all collections of digitised materials will become accessible in one way or another at the Internet, and the amount of information at the net is growing exponentially.

As mentioned before the access to information and knowledge is the crucial point in bridging the gap between north and south, between developed and undeveloped areas. Can the Internet be the answer? It is worthwhile trying, and that is what we want to do with our IFLA project. But before I introduce the IFLA document delivery project, I want to say just a few words about the barriers in the developing countries.

Today there is full Internet access in most parts of the world. Africa has been lagging behind, but within the last year the Internet has begun to permeate Africa too. But while the implementation of electronic networking in most part of the world has been systematic and well planned, this is not so in the developing countries. Uncoordinated experiments, inadequate facilities and lack of instruction and training in the use of the networks are the most common situations. In addition the technical solution used in connection to the Internet often is slow and bad functioning.

No doubt if you compare the possibilities of access to the resources in the Internet for university students, teachers, researchers in the industrialised part of the world with the possibilities in the developing countries there are still great differences. We have to discuss how we can support these countries, e.g. by international development programs. I believe that an implementation of the Internet at a research library will help in two ways. It can help to establish the electronic library with access to information at the Internet, and the library can act as training centres for higher education and research, regionally and maybe nationally. Even for the primitive Internet connections, it will still be possible for libraries to offer an inexpensive access to many library catalogues and to several on-line services and to use the Internet as a unique tool for having articles or information delivered electronically instead of ordering hard copies. How this can be done is what we want to find out with our IFLA project.

The IFLA-Trial projects

At the IFLA conference in Barcelona in 1993 the IFLA section on Document Delivery and Inter lending decided to start a project on Interlibrary Lending and Document Delivery in developing countries. The main idea of the project is to establish a trial project:

• To find methods for document delivery to developing countries; and

• To improve general access to and flow of information.

More specifically, the aims of the project can be described as:

1. To establish electronic network links with a regional and global approach to improve Universal Availability to Publication and Information.

2. To improve the competence of the library staff in handling interlibrary loans and document delivery systems (regional, national and global).

3. To support the negotiations with main document centres/libraries to attain favourable bulk treaties.

It was decided to start trial projects in some English speaking university libraries in Africa. Then later, if it turns out as a success, similar systems would be introduced in other parts of the world. It was also decided to build upon the Internet and to run the trial project in cooperation with other bodies as for instance IFLA 's regional Office in Africa, the IFLA Office for International Lending and CDNL (Conference of Directors of National Libraries).

The Danish International Development Assistance (DANIDA) and the Norwegian Minister of Aid to Developing Countries are both supporting different development and research projects in Africa, and both showed an interest in supporting our trial-project. For practical reasons we decided to run two parallel projects - a Danish and a Norwegian project. Both projects will be coordinated in a common follow group, with members from the IFLA Section on Document Delivery and Inter lending. The IFLA coordinating group shall:

• coordinate the two national projects;
• exchange information and experiences;
• make sure that both projects are conducted as model projects; and

• give a common evaluation of the projects to IFLA - for instance by arranging a workshop to discuss results and perspectives.

At present we are in the planning phase of both projects, but we hope to start in autumn 1996. The Danish project will be established in Ghana and the Norwegian project will take place in Kenya. Both projects will probably run for two or three years.

Financial support to buy the necessary equipment and to establish Internet connection will be a part of the project. Beside, there will be arranged training workshops for library staff members and researchers, who are going to use the Internet during the project period. As a part of the project we try to work out different types of agreements:

1. Bulk treaties with Document Supply Centres.

2. Agreements with database hosts.

3. Agreements with other libraries, e.g. twin-library agreement.

4. Document supply support by vouchers.

(A voucher is a reusable plastic card, which represents a standard payment for a lending transaction).

Our expectations to the IFLA project are great. In a way the Danish part of the project has already been started. In April this year a planning seminar took place at Legon University in Accra. At the seminar we discussed how the document delivery project could be organised and structures and technical solutions that would be appropriate. It was very promising to see the enthusiasm and expectations from management and staff in both university and library, and to note that government bodies except their responsibility for the project too, and see the interest from the commercial communication companies in supporting it.

Of course we hope, that the IFLA project will validate the assumption that the special efforts from the project will improve strongly the availability of publications in the area. An evaluation of the project is of course an important part. We have to analyse the problems we meet, and we have to discuss which solutions we can use to overcome these problems. We want to investigate different kinds of structure and technical solutions. We need to find out how we can educate and train people both in working with computers and in using document delivery systems.

We will carefully analyse the economical consequences of the project and try to estimate the costs of an Internet-based document delivery service compared to more traditional library solutions. We are not only looking at the financial part but also evaluating the effect of the document delivery service.

For both the Danish and the Norwegian project the dialogue to the libraries in Africa will be an important part of the project. We want to establish contacts not only between the libraries involved, but also personal contacts between librarians, subject specialists and researchers in north and south. The Internet will be an important communication channel in this dialogue.

Memory of mankind and founding the futures by preserving the past are keywords for this conference. They are also keywords, when we look at the perspectives of our IFLA project. I do not doubt that research and education at the institutions in the Third World can benefit immediately from the information accessible in the Internet. However, there still is a lot of important material of value for research, for instance in humanities, which are not accessible in digitised form. A national library with a satisfactory collection of the national literature or literature about the nation is more an exception than a rule in most Third World countries. In Ghana for instance, the most comprehensive collection of Ghanain literature and literature about Ghana exists at the Library of Congress. I hope that the combination of the Internet access and the efforts of digitising this kind of national literature for the libraries and countries taking part in the project could be a part of our IFLA project or a spin-off of the project.

Perhaps this could be done in a memory of the world project - a project running parallel to our IFLA project. I can promise that our project will strongly support and gladly participate in such a project.

Through the support of NORDINFO, three Centres of Excellence for the Nordic library community are established in the Nordic countries. The Centres of Excellence shall make knowledge on the use of information technology in the library environment available to libraries. This is done as ordinary consultancy work, participating in projects, seminars and papers, and via the World Wide Web. One of the three centres, the Nordic Digital Library Centre, has its focus on digital handling and access to national library collections of audio-visual information.

1 NORDINFO

NORDINFO is a Nordic institution within the Nordic Council of Ministers. The main purpose of NORDINFO, the Nordic Council for Scientific Information, is to promote Nordic co-operation within the field of scientific information and documentation, principally in connection with the research library system. NORDINFO shall also promote Nordic interests in a wider international context.

NORDINFO's task is to be instrumental in the development aiming at better and more efficient ways of disseminating information to research workers and other users of scientific and technical information resources in the Nordic countries.

On this background, NORDINFO has taken the initiative to establish three Centres of Excellence in different areas of interest on information technology. And to help realise these centres, NORDINFO is supporting them financially as well. The three centres are:

• Nordic Net Centre (NNC), located in Lyngby, Denmark and Lund, Sweden;
• Nordic Centre of Excellence for Electronic Publishing (NordEP) in Esbo, Finland;
• Nordic Digital Library Centre (NDLC) in Mo i Rana, Norway.

Although the centres work on different areas of interest, they co-operate to a certain extent as well through co-operative projects. They also share a common information channel, called NordELIB.

The centres are presented briefly below, and NDLC is given a more in-depth presentation.

Additional information may be found on World Wide Web at:

http://www.hut.fi/NORDINFO.

2 NNC - Nordic Net Centre

NNC is a Nordic centre of knowledge about networked information services. NNC aims to support and inspire the application of information and network technology in libraries and other knowledge centres. NNC will also contribute to increased co-operation between libraries and networking sectors and support Nordic development activities in the area. The centre is a 3-year project, carried out by LUB - Lund University Library in Sweden and DTV - Technical Knowledge Center & Library of Denmark.

Activities

The Internet and its client/server technologies, i.e. World Wide Web, WAIS, electronic conferences etc., is the focus area for NNC.

The activities comprise an education/demo/test/support centre with, consultancy service, courses, seminars, and workshops. The Centre communicates software and hardware experience and is on the Internet with extensive electronic information service.

Additional information may be found on World Wide Web at http://www.nnc.dk/.

2 NordEP - Nordic Centre of Excellence for Electronic Publishing

Electronic publishing includes creating, storing, presenting and distributing documents in electronic form. It offers an alternative for distributing existing information but it can also serve as the only form of new publications, which may be distributed on magnetic or optical media or via networks. NordEP is a project carried out by VTT Information Service in Esbo, Finland.

The Nordic Centre of Excellence for Electronic Publishing aims to:

• follow the development of electronic publishing from Nordic libraries', information services' and publishers' point of view;
• increase and promote the knowledge on electronic publishing and implement-ations among Nordic libraries, information services and publishers.

Additional information may be found at http://www.vtt.fi/inf/nordep/nordep.html.

3 NDLC - Nordic Digital Library Centre

3.1 Main goals

The main goals of NDLC are initially as follows:

• achieve a better understanding for the potential of digital representation of audio-visual information;
• investigate ways in which high speed wide area networks will give the Nordic libraries new possibilities;
• increase the amount of knowledge in the Nordic libraries with regard to various aspects concerning this area among Nordic libraries, archives and museums;
• represent an attractive resource of competence for Nordic library communities.

These goals can be reached only through a close co-operation with other communities of interest in the Nordic and other countries through projects, seminars and more or less informal contact.

3. 2 The National Library of Norway

NDLC is organised as an integrated part of the National Library of Norway, located in Mo i Rana at the Arctic Circle. The National Library of Norway, established as late as 1989, is an institution where one will find all kinds of information, information carriers and professions at one place. Although relatively small, it is well developed with extensive laboratories, heavy use of information technology, and a well educated and trained staff. The information repositories at the library includes, at least in a Norwegian scale, large amounts of photographs, sound recordings, films, radio and TV broadcast programs, and of course written material. This environment represents an ideal context for NDLC.

3.3 Some examples

At the moment, NDLC is running or participating in several projects on digital handling of and network based access to audio-visual information. Some of these are described below.

3.3.1 Networked poetry

Networked poetry, where the Nordic library community is encouraged to make presentations of local writers including text, images and sound recordings of the writers themselves. Through this collaborate work, it is possible to make a broad and different view of the writers in the Nordic countries. Internet and the World Wide Web is used for access to the information, and the information itself is distributed among participants in the project.

3 3.2 Still images

The National Library of Norway has been working on database systems for photographs for several years. Galleri NOR is the major result of this work, and today it is the production database for photographs at the National Library of Norway. Based on of-the-shelf Internet technology, this database can make large numbers of photographs available through the Internet. The database contains both digital copies of the photographs and a text based description of each of them. It is build on an Oracle DBMS, using Oracle Forms for registering metadata. The main search user interface is based on the use of World Wide Web.

Until now, only low quality digital copies of the photographs are put into the database. But NDLC wants to look into the area of high quality digital copies in the database, with the implications on data amounts and access control this will have.

NDLC is also working on a project on old maps together with the National Library of Iceland. In this project, one is looking on both preservation and availability of old maps, mainly the map collection at the National Library of Iceland. A demonstration of the access interface based on WWW is already available on the Internet.

3.3.3 Network access to digital video

The Archive for Sound and Image in Stockholm, Sweden, the State Library in Århus, Denmark, and NDLC are running a project which aims is to make a set of commercial films from the 1950s available as digital video via computer networks. Again, standard Internet technology is the base for the work. Early in 1997, about 800 short films from Denmark, Norway and Sweden will be available on the Internet through a common database. The users of this database will mainly be researchers in the Nordic countries, but it will also be available for education. Some of the films will be available for everybody.

4 NordELIB

To promote co-operation between the three centres, and to achieve a better dissemination of information from the centres' activities, NordELIB is established. It works as a common meeting place for the centres, and uses projects, meetings, WWW and open email distribution lists as tools.

This and Dietrich Schüller's paper make a pair: we are coming at related subjects from different angles. Together we are going to talk about the AV (audiovisual) media: the spectrum that includes:

• film (various gauges, for cinema, other professional or amateur use)
• sound recordings (whether on disc, cylinder or tape)
• radio
• television (broadcast, cable or satellite), and
• video (tape, disc, rental or sell-through).

And what else? What about new media like video games, CD-ROM multimedia, and the Internet? Or old media like piano rolls, mutoscopes and optical toys? An extended discussion might include these, but for the sake of simplicity let's keep to the five core media.

I am assuming, by the way, that in this forum I don't have to justify the crucial cultural importance of preserving AV materials. I can think of other audiences and other places where - even today - I could make no such assumption.

Let's begin with film. It comes in a variety of standard types or formats. For example:

• The long, narrative feature film has been the major commercial format for more than 80 years - the economic foundation of most industries, and most obvious expression of cinema as both art and entertainment.
• The short comedy or drama, once a hugely popular format in pre-television days, is now rarer and more specialised.
• The documentary is real life, structured and distilled: a mainstay of most industries.
• Before television, newsreels were a weekly record of major events; now they are a revealing window into yesterday.
• Animated cartoons - the fantasy world which the movies created.
• Advertising films - short and long - have been there from the beginning.
• and amateur and home movie films, created in huge quantities, are both unique and a social mirror.

There are some standard television formats too. Here are several:

• Newscasts and current affairs programs are today's newsreel: longer, more frequent, more discursive.
• Sport floods the airwaves and reaches vast audiences. How many watch the World Cup? How many will see the Atlanta Olympics?
• Commercials - the aspect we love to hate - are prolific and important: the true mirror of values.
• Variety and chat shows have huge audiences and enormous influence on public opinions and perception.
• Series and serials are the commercial foundation of television - whether they be drama, comedy, or other kinds.

Television feeds on itself: these days there are even shows about television and about commercials. In addition, everything ever made on film, in any format, can potentially end up on television.

In fact, the moving image and the sound recording have utterly revolutionised communication and art in the space of a single century. Nowhere in the world is 'safe' from broadcast, satellite or cable TV or radio. The grammar of the moving image is now as instinctively familiar to us as the grammar of written and spoken language. Could the initiators of this new age - people like Thomas Edison, Alexander Graham Bell, Guglielmo Marconi, the Lumiere Brothers and the rest - have imagined where it would all lead? Or that these images and sounds would have anything other than transitory value - that at the end of this amazing century we would be mourning what had been lost, and be preoccupied with how to preserve what can still be saved?

I want to show three examples of moving images - documentary, feature film and television news - to pose the following question (which I won't answer): How would you rate these as part of the Memory of the World? And how would you compare their importance to written records of the same events?

• The Inauguration of the Commonwealth of Australia, January 1, 1901 - Australia was almost certainly the first country to be born in front of a movie camera. This is footage of the formal signing ceremony in Sydney which brought the nation into being. It was photographed on 35mm film and has survived, as so many films do, by accident.
• The Story of the Kelly Gang (1906) - Australia's, and possibly the world's, first feature length dramatic film was based on actual events and personalities that are now a central part of Australian folklore. The actor playing the central figure, Ned Kelly, wore Kelly's actual armour for the camera - art recreating life. Only fragments of the film now survive, and the visual effect of chemical decomposition of the nitrate film is obvious.
• 'People Power' Revolution in the Philippines as reported on SBS Television News, Sydney, on 22 February 1986. The news cameras documented a seminal event in modern Philippine history, seen around the world as it happened - the deposing of the Marcos regime by the peaceful demonstration of popular will.

The hard questions

Let's turn first to the hard question - in preserving moving images, how much do you keep? There are three alternatives:

• Keep everything. Many would see that as desirable, albeit ideal. No one, to my knowledge, has been able to do it yet.
• Keep a deliberate selection. The next best choice. But it's difficult - it requires reference points, like a selection policy and known priorities. It also requires knowledgeable people, trained to make skilled and rational judgements. Even so it remains to some degree an arbitrary process.
• Keep whatever survives. In reality, this is what we do in practice, all too often: when events have overtaken us and we have no other choice.

There is, of course, a fourth alternative - keep nothing. It's certainly the easiest one. Consider the consequences of keeping no coverage at all of your national audio-visual heritage. Why, indeed, does the world need a memory?

Having chosen what to keep, who are the custodians? The usual answer is - audio-visual (AV) archives. This is actually a collective noun, using the term 'archive' non-traditionally, for a wide variety of institutions or departments of institutions, large and small, public and private. They go by many names. Some have a wide brief: others are characterised by particular formats, subject matter or motivation. Their purpose is to collect, manage, preserve and provide access to the AV heritage, and most identify with one or other of the international federations which globally represent the field of AV archiving.

Supplementing the culturally-motivated efforts of publicly-funded AV archives are individual production companies, television networks and others who have in-house departments to manage what are now seen to be corporate assets. Here, of course, the motivation is commercial prudence - protecting an investment to allow future exploitation - but the practical result is the survival of the material.

Helping both is the private collector: individuals whose passion, sentiment and ingenuity have saved so much, but whose own resources cannot ensure permanent preservation.

Realities of survival

What is our record so far? What has survived? What is preserved?

All countries have suffered loss. When we look at the silent film heritage - that is, films made before about 1930 - we find the count varies widely, but probably none has done better than about a 60% loss rate. In my own country, the loss rate is 95% and in some countries it is virtually 100%. Think what our reaction would be if we lost that percentage of the world's literature for the first three decades of this century!

For sound films, post-1930, the figures are better but again vary greatly. Some countries have lost almost everything made before 1950, 1960, even 1970 - and much of what is left is in serious trouble.

When we move to television, we remember that much live-to-air programming was not recorded - so we can never recover it. And as with film, much more has been discarded or has decayed. Videotape can have a short life, and unlike film it can be recycled - and often is: the new obliterates the old. The further back you go, the smaller the surviving portion - and, usually, the worse it looks. If you can still find equipment to play it, that is: the shelf life of video technology is often even shorter than that of the tape itself.

The fact of survival is of little use unless you can find what you are looking for. So, how well are the world's audio-visual collections documented and catalogued? Again, the available information varies greatly - but, on the whole, the answer is: not nearly as well as the world's literary heritage. In many cases, the catalogue information is scanty or non-existent: and what there is, is not necessarily available in a standardised or readily accessible form.

Finally, there's the question of physical management: how appropriately are these environmentally-sensitive collections stored, organised and maintained? Very few, in fact, are in a situation that could be described as ideal; the spectrum runs mostly from 'acceptable' to 'dangerous'.

The reasons for this state of affairs will sound familiar: lack of environmentally-controlled storage, lack of equipment, lack of funds and staff. Underlying these may be less visible, but equally important, circumstances, such as the absence of suitable skills and training, and of an appropriate organisational structure that can support both preservation and access.

A satellite's eye view

Now, a snapshot of film and television archiving activity around the world.

Europe and North America have, relatively speaking (and I emphasise the term relatively) the oldest, most mature, and generally best developed situation. It's where AV archiving originated and where it has possibly the greatest cultural acceptance. It also has, in terms of film and tape longevity, the most benign natural climate.

In Latin America there is an active array of film and television archives, with committed staffs, and a large and diverse multinational film and television history. It is less visible to the world, resources are tighter and the climate is less benign.

Across Africa the network of AV archives is generally young and thin, and moving image preservation has to compete for limited resources in political and economic circumstances where, all too often, it is unlikely to have a high priority. Africa's profile in international AV archiving forums is low.

In the Arab states the profile also tends to be low and the degree of wider networking limited. (As with Africa and Latin America, this is an outsider's perspective - I've not yet been fortunate enough to visit AV archives in any of these regions in order to view the work first-hand.)

Finally, there's the Asia-Pacific region, the largest geographic area identified in this conference, and home to half the world's population, as well as the world's largest and smallest countries. It contains some vast film and television industries, together with economic extremes and political diversity. From country to country there is great variation in both the survival of the AV heritage and the capacity to deal with it. I'll return to this area later.

Nature of the moving image media

Let's travel back from the satellite to consider the special attributes of the material we are dealing with: its content and its physical nature.

In content terms it contains, obviously, pictures and sounds. But depending on the maker's intent and the user's point of view, the pictures and sounds may be (for example) an artistic work, a historical record, information, propaganda, entertainment, or simply a commercial product. Sometimes it can be several or all of these things simultaneously. All can be possible justifications for preservation.

Consider now its physical character. In theory, its content is infinitely 'copiable' - though in practice there's usually a quality loss involved - although a reel of film or tape, like other media, can have its own artefact value if it has particular attributes which can't be copied. It is fugitive and environmentally vulnerable and fragile - you can't just put it in a shelf and expect it to look after itself. It damages and defaces easily and requires skill and care in handling, especially over the long term.

It survives best within precise ranges of temperature and humidity, matched to its technical nature (nitrate and acetate film, colour or black and white, videotape; each has specific requirements). Degradation accelerates in proportion to the degree of variation from these conditions. Most of the world's film and tape stocks are not held in anything like ideal circumstances: and in tropical countries, heat and humidity have a terrifyingly rapid effect. Chemical breakdown is the end result: to the well-known instability of the now obsolete nitrate film stock can be added the newer phenomenon of vinegar syndrome - the breakdown of modern acetate film, the basic material of film making today.

It is a distinguishing feature of AV materials generally that they can't be perceived directly by the eye or ear - they are intelligible only via a technical device. This seems to me to be a fundamental paradigm shift in human communication - think about it! - and it means, in turn, that the custodian or archivist has immense potential to manipulate reality and change history without anyone ever knowing. The ethical implications are profound.

Mechanics of survival

Many of the world's cultural artefacts have survived because they've been long forgotten and neglected. Unfortunately, this rule doesn't apply to film and tape. They have a self-destruct mechanism built in, which means that neglect equals loss. But that's not all: there are other factors which affect survival.

The more copies made of a given film, the greater the chance of survival. This augurs well for the Hollywood blockbuster, but not so well for films with smaller markets. Some films - such as home movies - typically exist in only a single copy.

Nor is it just a matter of statistics. There are numerous documented examples of the deliberate destruction of AV materials. Sometimes these are for ideological reasons, to remove the memory of a particular time, even of culture. Sometimes, more mundanely, they are destroyed for contractual or legal reasons, to prevent uncontrolled exploitation - the result of films or TV programming being 'product', as the industry terms it. Sometimes it just happens through coincidence or ignorance - like cleaning out old rubbish. And in the television industry, where videotape is constantly recycled for both practical and economic reasons, it takes just seconds for a magnetic field to obliterate all evidence of yesterday's mega-production.

Legal arrangements, such as statutory deposit of AV materials in recognised institutions, have become increasingly important as an encouragement to survival. But there is another side to the coin: deposited materials have to be cared for or they'll be lost sooner or later. Guaranteed acquisition doesn't mean guaranteed preservation. If a collection is not documented and managed it's already in the process of slow self-destruction - and without alarm bells ringing! Management, in turn, presumes organisational structures - people, procedures, resources, equipment, activity. Unfortunately, so many of our film and tape stocks have so little related infrastructure that they are effectively graveyards.

Realities of preservation

Perhaps this rather gloomy outlook can at least be lightened by the march of technology. Surely digitisation is going to offer a way of getting past these limitations?

Unfortunately, no. At least not yet. It's true that digital video is already with us, and progressing rapidly. But its effect is often to simply add another layer of technology to the expanding range of current and obsolete equipment and skills which AV archives have to maintain. Even when a reliable standard is settled, the cost and logistics of converting an analogue collection to preservation copies in the new standard may be out of the question.

When we turn to the preservation of film images, with their much higher 35mm quality, it's clear that there is no cost-effective digital alternative in view in the foreseeable future. Until there is one, the celluloid strip will remain the preservation medium.

This means that any kind of high quality preservation of moving images is expensive. There are the costs of film and tape stock, equipment, staff time - as well as of storing and managing the resulting copies and originals. And even if the money is there, the infrastructure of skills and equipment may not be. The job may be too large to outsource to another country or even another city. It's only going to happen if the infrastructure is built where it's needed. That in turn will mean the training of the people who will do the job - often in skills that are unusual or obsolescent and that aren't supported by the mainstream media industries - as well as the assembly and maintenance of equipment to play non-current tape and film formats.

All of this makes net demands on governments. The political will needed is considerable - and not always available.

And now to Access....

I want to contrast these rather depressing realities with the possibilities now being opened up for access to the audiovisual heritage. After all, the whole purpose of preservation is future access (why else would you keep anything?). Maybe - just maybe - it will be the engine that can drive solutions to some of the needs I've outlined.

Traditionally, moving images are randomly accessible through cinemas, broadcast TV, festivals and other events - where you choose from what the programmers have put on offer. From other sources, like video stores or AV archives, you can check catalogues and select what you want to see, when you want to see it. To that pattern, we are adding new possibilities. These include cable TV - which is serving new, minority 'niche' markets - laser discs and video CD's, and 'juke box' sources from which you can retrieve images mechanically or electronically. And now there's the Internet... soon it will be fairly easy to send good TV quality moving images from anywhere to anywhere, over the Web.

Imagine this scenario. I'm sitting at my home computer. I log in my credit card number, key in what I'm looking for, and I start to surf the moving image databanks. Whether they're located in Canberra or Oslo makes no difference: I won't even be aware that the software is checking many catalogues simultaneously around the world against my search criteria. It comes up with answers and identifies the sources - let's say several AV archives in different countries - who can provide what I want. It gives me film titles, accession numbers, references to written and graphical material - and the email addresses of archivists at these institutions, with whom I can dialogue for further information. When I am ready, I can view my selections on-line, or download them. I can print out hard copy documentation - say, a film poster, critique or photograph - that's been turned up by my search. I can place orders for products matching my search criteria: a book or video? a CD-ROM? a video game?

All this time, I haven't left my seat. By their nature, the AV media come to me, I don't have to go to them. Now technology can remove traditional access limitations, multiplying availability and visibility.

How does this help the problems of preservation? Well, I have just paid for my search, and some of that money goes back to the supplying archives to help their preservation program. Popularisation and ease of access raise awareness of issues, grow expectations and profile, change perceptions and build political will.

Let's be realistic. There are no panaceas or magic wands. This is just one way in which access can help preservation. But in the end solutions rely on committed individuals doing hard work and making hard choices. While such a computer search is possible now, in theory - the technology exists already - practicality limits availability. For example, there's copyright (the rights holder wants a fair return on investment). Then, we don't all speak the same language, which limits the potential to search databases across languages and cultures. Finally, many collections are unable to tap their potential because of the cost of converting material to digital form and therefore making it accessible.

On-line access will be available where the infrastructure is and the market will influence the demands and the agenda. As is so often the case, the developments and benefits will be unevenly distributed. But I wouldn't assume they'll exactly mirror the present status quo. Maybe we can, country by country, change it.

A case study - South East Asia

To draw the threads together, let's explore a case study from my neighbourhood - South East Asia.

It's a dozen or so countries, of very diverse size, languages, cultures, economic levels and political systems. It's also a dynamic region, with rapidly growing economies and a large population, large media industries, a mooted regional TV channel, expanding broadcast, cable and Internet networks. The Philippines, for example, makes 150 feature films a year. Linkages like ASEAN (Association of South East Asian Nations) are very meaningful structures for trade and cultural development. The countries share a geographic neighbourhood as well as, increasingly, a single language of international contact (English).

The countries also share a relatively short tradition of AV archiving, as well as a climatic problem - tropical temperatures and humidity eat rapidly through film and tape. There have been huge losses of film and TV material across the region. In many cases, the archiving structures are young and embryonic, and the political will to deal with the issues has often been lacking. So, of course, have resources, infrastructure and skills. For the most part, these archives haven't been part of the international professional network: their needs were not even well known amongst themselves, let alone the world at large, and the work had low status and low profile.

What to do? We began at the bottom. We just helped ourselves. Concerned individual archivists began reaching out, visiting and talking to each other. In 1993 some of them gathered in Manila for a conference-workshop. This led to help from above - from ASEAN, the Australian government, UNESCO, Kodak and others - for an ambitious training program, which in turn led to a regional professional organisation, SEAPAVAA (South East Asia-Pacific Audio-Visual Archive Association) established this year. Based on the 'united we stand, divided we fall' maxim, it gives visible shape to a professional community, a profile to the work and a means of sharing resources. That, again, is leading to other things, such as an annual conference and the development of a permanent university course, based on a UNESCO curriculum, for training the region's AV archivists through the Internet, and so on.

This energy has been a catalyst in encouraging organisational structures in individual countries to be revisited. The principles and concepts of regional data standardisation have been adopted with other aspects of AV archive management to follow. (This will aid efficiency - and be crucial for access). We used the structures and interests of international relations as a resource. We have tapped some political will. The new ASEAN TV channel will make demands but also be a stimulus; it's an opportunity. We will use events, like conferences and training courses, as occasions which can also attract attention and work as a catalyst.

Summation

The challenges facing us in South East Asia are basically no different from those facing AV archivists everywhere:

• There is a preservation crisis. We won't solve it just by throwing money at it (even if we had the money) but by developing infrastructure, seizing opportunities and changing perceptions.
• We have to effectively communicate the realities, consequences and needs as we see them - and the solutions we propose. Perceptions and political will won't change unless the information gap is filled.
• There is no short cut to developing the infrastructure of facilities, equipment and people - but a regional strategy will make the best use of available resources.
• Support won't come unless we pursue it. We have to build our own support base. We have to generate the political will.
• Structures for AV archiving will vary from country to country for good reason. Whether they conform to this or that traditional model is not important; what matters is that they be appropriate and effective in their particular context.
• Training is a major challenge. AV archivists have learned 'on the job' for a hundred years. It no longer suffices. Formal university courses are at last emerging, and are desperately needed.
• A codified philosophy for the profession is just as urgent and, with UNESCO's support, is now in progress.
• An orientation to access as well as preservation will be essential to encourage growth and political will.

I believe coordinated action - national, regional and global - is a strategic necessity if we are going to save the audiovisual memory of the world. We haven't always been good at this. Now we must be. Much has been irretrievably lost; much more is on the brink; trained and committed people are the foundation of development; and we need a strategic and collective approach. Nitrate - and acetate - won't wait.

Is this the end? Or the beginning of the end?

No. I think it's just the end of the beginning. The best is yet to come.

Audio and video recordings are documents of ever increasing importance and significance. They are indispensable sources for many scholarly disciplines, the only true representation of orally transmitted cultures, and, in a time of ever increasing electronic communication, they constitute one of the major sources of our contemporary civilization. No evaluation of politics, history, everyday life and, of course, all kinds of music and performing arts, would be possible without the use of audio-visual documents.

The safeguarding of all these documents is widely associated with the keeping of books and other written materials. This is partly, perhaps, because textual libraries have existed for more than 4000 years, while audio-visual archives have been in existence for only less than 100 years. There is, however, a fundamental difference between audio-visual documents and printed materials which is commonly overlooked. The difference lies in the nature of the documents and in the different degree of redundancy of information.

Printed matter represents human thoughts by the use of a stock of symbols. A certain amount of redundancy is intrinsic in speech and writing. Letters, sometimes even words, may be omitted without any real detriment to communication. A good example are the scripts of Semitic languages which generally do not represent all vowels which are spoken. But still, even complex texts like philosophical tracts can be communicated that way. In contrast, the audio-visual document is an analogue representation of a physical status or event: every part of such a document is information. While a speck of mould in a book does not normally hamper the understanding of the text, comparable damage on a photograph would cover up information, and, on a magnetic tape, it could even render the tape unreadable. Seen, therefore, from the perspective of redundancy, audio-visual documents call for a higher degree of protection and security than written materials.

A second reason for increased efforts to safeguard audio-visual data carriers is the vulnerability of the carriers and their components. While paper is far from being without problems - as can be seen from contributions elsewhere in this publication - there are specific problems related to the stability of audio-visual documents, as will be explained below.

A short survey of the most widespread audio and video formats and their specific stability issues will help to understand the problems.

Phonograph Cylinders

Cylinders, originally developed for use as dictating devices, have been used since around 1889 for original recordings in the academic world and later also as mass produced recordings for the entertainment industry in competition with early gramophone (shellac) discs. While industrial production ceased in the late twenties, they continued to be used for field recording until the fifties(!). Most cylinders are made of wax, some of the mass replicated cylinders are made from celluloid [7]. There are about 300,000 cylinders (2) in the custody of recorded sound collections world-wide. They are extremely brittle and fragile and if they have been stored under conditions which are too humid, they suffer from mould. Fortunately, most of these holdings have already been transferred onto modern media and thus their contents, which are frequently of unique historical value, are already safeguarded (3).

Shellac Discs

Coarse groove gramophone discs, commonly called shellacs or 78s, were the main mass produced audio format of the first half of our century. It is estimated that the world-wide stocks of this format amount to 10 million discs. They were produced from 1898 until the mid-fifties. The discs consist of various mineral substances bound together by organic substances like shellac or similar binding materials. Although breakable when dropped, these gramophone records are fairly stable and there are no reports of a systemic instability problem [17].

Instantaneous Discs

Prior to the introduction of magnetic tape, which occurred in the late forties and early fifties, so called instantaneous discs were the only medium for audio recordings that could be played back immediately. The total number in existence amounts to three million. Practically all of these discs are irreplaceable originals, many of them of great cultural, historical and scholarly importance. Unfortunately, the largest group of these instantaneous discs, the 'acetate discs', are at the greatest risk. These discs are laminates of aluminium, sometimes glass cores with a lacquer coating of nitrate or acetate cellulose which is soft enough to be cut by a recording machine, but hard enough to withstand several replays. With age, the coating becomes brittle by a hydrolytic process: the lacquer then breaks apart, and flakes off. Thus a considerable portion of the holdings world-wide have already been lost. Even if transfer programs were to be hastily established, further losses of irreplaceable information cannot be prevented. Every day, hitherto intact records are being affected by this phenomenon.

Microgroove Discs

From the late forties onward microgroove discs (vinyl or LP records) replaced shellac discs and only relatively recently has this format been superseded and replaced by the compact disc (CD). The total number of microgroove discs in sound archives world-wide is estimated to be more than 30 million. They are made mainly of polyvinyl chloride [18]. No systematic stability problems have arisen so far on a greater scale, but their stability in the long term, thinking in centuries, is unknown.

Magnetic Tapes

Magnetic media in the form of tapes on open reel or housed in cassettes are the most widespread carriers for audio and video data. Early audio tapes used acetate cellulose as the base film material, which is also used for safety film. Acetate cellulose has a tendency to become brittle through hydrolysis caused by the moisture contained in the atmosphere. This brittleness generally constitutes a serious problem in the play-back of old audio tapes. Severe cases of hydrolysis end up in the so-called 'Vinegar Syndrome', whereby acetic acid is set free in ever increasing quantities, which has a catalytic, and therewith accelerating affect on the process of decay. This has been experienced in film archives, especially in hot and humid climatic areas [6,25]. Affected films become soft and limp, ending up as powder or slime. In principle, this may also happen to acetate audio tapes. Fortunately, however, no disastrous losses similar to those in the film world have been reported. Still, acetate tapes, which were produced until the mid-sixties, are at risk, and transfer onto other carriers must be envisaged.

Another group of historical audio tapes used polyvinyl chloride as the base film material. As with vinyl discs, these tapes have not exhibited any systematic instability; the long term prospects are, however, unknown .

Polyester is the base film material which is used for all modern audio and all video tapes. It has the greatest resistance of all base materials to mechanical stress and the influence of humidity. No systematic stability problems have occurred so far but, again, its stability over very long periods (centuries) is unknown [3,9,33].

Of the various magnetic materials used to store the information only metal powder, as used in more recent high density audio and video formats, has given cause for serious concern: early tapes of this kind suffered from corrosion but this problem now seems to be under control [15,35]. There is, again, no precise answer to the question of how long metal particle tapes will keep their information undistorted and readable. At this point it must be emphasised that, contrary to laymen's expectations, the magnetic information on properly stored and handled tape is not at risk.

The greatest problem related to magnetic tape is the stability of the pigment binder. A considerable number of audio and video tapes, especially amongst those produced during the seventies and eighties, are suffering from pigment binder hydrolysis. The atmospheric moisture is absorbed by the pigment binder causing the polymer to hydrolyse and lose its binding property [4,5,13]. Tapes of this kind deposit a smear of magnetic particles onto the replay heads. This clogs the heads and swiftly makes the tape unreadable. Processes to render such tapes playable again are available, but the restoration process is cumbersome and time consuming. This problem has been found especially in hot and humid areas where many tapes do not last longer than a few years (4).

Compact Discs

The compact (CD) disc and its forerunner, the laser vision disc (video disc) have both suffered from delamination, reflective layer corrosion, and crazing. All these problems render such discs unplayable. They have occurred mainly during the introductory phase of these products, and it seems that the problems have now been solved. The long term stability, however, especially of the varnish on the upper side which protects the reflective layer, is under systematic investigation [2,11,22]. More research is also required into to the stability of the recordable CD (CD-R).

In summarising, it can be stated that the stability of all polymeric materials over long term periods is limited. This has a major bearing on the stability of audio-visual data carriers: the vast majority of these consist of polymers, in many cases of a sandwich of polymers, where possible interactions between layers have also to be taken into account. It can, therefore, be unequivocally stated that, with rare exceptions (5), there is no eternal audio-visual data carrier.

To achieve the aim of 'eternal' preservation for the information requires, therefore, that the information sooner or later has to be copied. In the analogue domain, however, each copy differs, if only slightly, from the original. With multiple copying, therefore, the information tends to zero. Even if we assume a fifty years lifetime of an average audio/video carrier, we would need twenty generations of subsequent copying to cover a millennium. Everybody who is aware of the decay of quality in the analogue domain, eg of a video after three or four generations, can imagine how little of the original quality will be maintained. Clearly the way to overcome this problem is to transfer all information into the digital domain where it subsequently may be copied to 'eternity' without any alteration and loss. Over the last fifteen years since the introduction of digital techniques great hopes have been expressed in the audio-visual archive world that this technology would offer simple solutions to overcome the pitfalls of the analogue world.

However, the digital revolution has differed from what had been expected. Instead of the replacement of the hitherto few (if not single) professional analogue formats by one or two generally accepted digital formats, the competition between the producers of audio-visual equipment has led to the development of several competing formats which - due to an incredible progress in technology - were often outdated after a comparatively short time and superseded by new developments. This has sometimes occurred even before the market had accepted the previous format. It has been especially true in the development of digital video formats where, to date, some eight formats have been developed, none of which has yet reached a dominant market position.

This situation inevitably leads to the threat of the obsolescence of hardware. There are several audio and video formats of which carriers exist in good condition but they can be only be transferred with difficulty and at great expense because of the lack of hardware and spare parts (6). Advancing technology is a highly aggravating factor in this process. While it is possible, though not inexpensive, to construct a new cylinder replay machine to play these early recordings with better fidelity than any machine of Edison's time [21], it is impossible in practice to build a CD-player or a digital video recorder once mass production has ceased and the last machine or its spare parts have been used up.

Seen from this perspective, the long term preservation of audio and video data - if we speak in centuries - is hopeless: the carriers are unstable, the commercial lifetimes of the formats seem to become shorter and shorter and the amount of data to be stored is too big to allow manually operated subsequent copying from one commercially available format to the next, even if only 10 % of the amount of data available today is stored.

The solution lies in automatically accessible, self controlling and self re-generating archival systems [26,27]. The features of such systems are:

• The management of audio-visual data as computer files in mass storage systems, eg jukeboxes of magnetic tape cartridges.
• An open file architecture to accommodate all audio-visual data together with catalogue/content information and written text.
• The access time of such systems is not of major importance.
• Data integrity is controlled automatically, and copying of the information onto new carriers (refreshing(7)) is done automatically before errors cannot be longer fully corrected.

Once new storage media and systems are available due to technical development, automated migration (transformatting) will be carried out.

First thoughts in this direction have been expressed during the Second UNESCO Consultation of Users and Manufacturers of Audiovisual Equipment in Vienna, Spring 1989. The first in-depth discussion took place during the Joint Technical Symposium in Ottawa 1990 after a presentation of a more elaborate paper on that topic by the author [26]. For audio-visual custodians which have devoted so far all their professional endeavours to protect their precious documents 'to eternity' it was provocative to suggest giving up the idea of the 'eternal document' in favour of the idea of the 'eternal computer file'. In fact, mass produced documents such as cylinders, shellacs and LPs have become objects in their own right, with a certain intrinsic value, sometimes even venerable relics. Consequently it was and still is not easy to accept that in the long term the preservation of the original artefacts will not be possible, at least not for the vast majority of the existing documents.

In the following years, however, this change of paradigm has become more and more generally accepted. In 1992 it was discussed at the Tonmeistertagung in Karlsruhe [27], and in 1994 an Archival Group of the German Phonographic Industry was formed to draw up specifications for digital mass storage systems for the safeguarding of the master recordings of record companies, along with all the associated information: sleeve notes, labels, booklets, photographs, contracts etc. First installation are expected to take place 1996/97. In connection with this development, similar endeavours are being undertaken by several radio stations within the ARD, the community of German broadcasters. These pilot projects are under development and expected to become operative also by 1996/97.

While at the beginning of this debate there was a certain danger that safeguarding audio-visual materials would be carried out by using the newly developed data reduction (compression) algorithms, it has, at least in the audio domain, become generally accepted that data reduction (compression) is considered to be unethical because of its prejudicial technical effects on the further use of the material (8).

If audio-visual preservation in the long term can only be successfully carried out in the digital domain then it would be interesting to have information on the order of magnitudes incurred. A study carried out by the Library of Congress [12] gives an estimate on the worldwide holdings of audio-visual materials which can be used as a basis for calculating the digital storage capacity required for their safeguarding in the digital domain. The result is impressive: the worldwide audio holdings are estimated to amount to 45 million hours, corresponding to 30 Petabyte (ie 30,000 Terabytes)of digital storage space; the worldwide video holdings are around 9 million hours, amounting, in digital uncompressed format, to around 1 Exabyte (ie 1 million Terabytes). The annual growth rate is reported to be 5-10 % (9). Even in the unlikely event that by radical selection only 10 % of the world-wide holdings would be declared worthy of being kept for 'eternity' the remaining 100 Petabytes would still be an enormous challenge for the computer industry. It must be remembered that digital film preservation is not included in this calculation as it requires enormous storage capacities: 3.6 Terabyte per hour for the full representation of 35 mm colour film seems as yet not economically feasible. Also, photographic stills and, most important, textual information - heavily debated in the context of 'Digital Libraries' - have not been taken into account.

In view of the rapid development of digital systems is does not seem utopian to think that these future storage requirements will be successfully and affordably met. While in the beginning of the debate during the early nineties the biggest available mass storage system was capable of capacities up to 30 Terabytes, current development allows for the storage of 2.5 Petabytes. In principle, there is no limit to the further expansion of the capacity of such systems. Present systems are based on magnetic tape cartridges which are derived from digital video formats, but other media, eg optical tape may be available in the future.

The problem of digitisation seems not so much to be linked to hard- and software technology, the real problem is the transfer from the analogue to the digital domain. This is a very labour-intensive process which requires, depending on the difficulty caused by the condition of the original document, a time factor of 1.5 up to 10 of the duration of the document. On an average, a factor of 3 must be calculated. Therefore priorities have to be set in the transfer to the digital domain: in the first instance, only carriers who are endangered should be transferred, and, additionally, those documents which are in frequent demand. The question of 'what is endangered' is not so easily answerable. While it is obvious that all instantaneous discs and all historical tapes made from acetate cellulose are at risk and must be transferred, the greatest problem today is the prediction of the life expectancy of magnetic tape; while many tapes have survived successfully for several decades, tapes of more recent production particularly have caused replay problems. Of greatest importance, therefore, is the intensification of systematic research into the life expectancy of audio-visual data carriers, especially of magnetic tape, and also of replicated and recordable CDs (10). It is imperative to know what kind of holdings are of immediate risk in our collections and which can wait. Without proper research tools we would waste time and money transferring the stable parts of our holdings, while other documents rot away unnoticed.

Two other important questions relating to the analogue-to-digital transfer are: do we yet have the appropriate transfer technology and do we have the appropriate digital resolutions? Taking an example from the audio world it must be mentioned that the conventional transfer by mechanical pickup systems of carriers like cylinders and historic discs may be superseded in the future by optical, non-contact transfer methods. Also the current 16 bit resolution of the CD and other commercial digital audio formats is about to be replaced by 20 and soon by 24 bit resolutions which allow for a better, more accurate, capturing of the original document. Therefore, digitisation should not be hastily taken on board. Most important, furthermore, it must be understood that every transfer is more or less preliminary. Consequently, the original analogue carriers must be further preserved for future, better transfers, unless they disintegrate.

As a result of this complex situation, the strategy for their safeguarding of audio and video materials in the long term has to be twofold.

In view of the enormous amount of analogue materials in association with the labour-intensive task of transfer, taking into account the possible further improvement of transfer technology and digital resolutions, all effort should be undertaken to prolong the life of existing carriers to the maximum possible extent. As the complete preservation of all present holdings is not desirable, the necessary selection process will also be easier from a further distance. Emphasis, therefore, must be given to systematic research into the prediction of life expectancy of audio-visual data carriers and into measures to retard their decay. Hopefully, it will also be possible to develop, on a greater scale, measures to rejuvenate already deteriorated carriers.

For the long term, however, it has become clear that self-controlling and self-regenerating digital mass storage systems are the answer for the safeguarding of audio-visual documents. This kind of concept will also provide a solution to overcome the scepticism vis-à-vis the safeguarding of electronic documents as expressed by Jeff Rothenberg [24] and also voiced elsewhere in this publication. Such mass storage systems are, at the same time, an indispensable pre-requisite for the functioning of all kind of services in the forthcoming information age, to mention only 'digital libraries' and 'video on demand'. Their prices will come within the reach of average budgets. Contrary to sometimes expressed fears this concept does not call exclusively for huge, centralised stores: it will also allow individually tailored solutions for smaller applications. Thus, such systems could also be a solution for the preservation of documents in southern countries: While in hot and humid environments conventional audio-visual preservation is generally insufficient due to the notorious lack of funds for the proper air-conditioning of storage areas, mass storage systems, requiring relatively small floor space, could be effectively air-conditioned at low cost.

(1) This article is an updated and extended version of a paper originally read at the Symposium 'The Eclipse of Memory' convened by Fondatione IBM Italia and the Academia Nationale dei Lincei in Rome, December 1996 [30].

(2) The estimates of worldwide holdings of audio and video materials are taken from a recent study [12]. The author is indebted to Mr. Gerald D.Gibson, Head Curatorial Section, Library of Congress, for providing access to data.

(3) In the United States most of the important collections of cylinders have been transferred onto modern carriers by the Federal Cylinder Project, co-ordinated by the Library of Congress. Similar attempts have been made by individual collections in Europe. There are, however, still many collections in Eastern European countries awaiting transfer onto modern media. The author, with the help of the Austrian Federal Ministry for Science and Research, has recently undertaken a survey in Eastern European countries to locate collections of endangered audio-visual materials.

(4) This phenomenon is aggravated by the fact that many archives, especially under hot and humid conditions, use air conditioning systems that cool the air only. Without simultaneous dehumidification, the relative humidity rises in such systems [32].

(5) These exceptions are the metal parts, the matrices or stampers of grooved and compact discs, which consist mainly of nickel. These metal masters will - under favourable storage conditions - probably last for a very long period.

(6) A sad example is the two-inch video format, which was in use from 1956 to around 1980. Replay machines have been hastily abandoned and the machines still available are too few in number to transfer the stocks of two-inch tapes, of which most are still in good condition.

(7) The author follows the terminology recently suggested in [8]. 'Refreshing' means the copying of an (endangered) carrier onto a new one of the same type (within the same system, ie hardware/software configuration). Migration means the transfer of data from one carrier of one system to another carrier of another (new) system (hardware/software configuration).

(8) Data reduction - based on perceptive coding - as employed in audio consumer formats and in digital broadcasts may be acceptable for listening purposes. It is a limitation, however, and prejudicial to the future use of the material. Data reduction should, therefore, not be used for archival purposes [27,29]. In the video domain, because of the enormous storage requirements, the use of data reduction may be even more tempting. Recent thinking, however, discourages the employment of data reduction algorithms for video archiving as well [16].

(9) The calculation of the required digital storage space is based on the full, unreduced (uncompressed) signals [29], the following bit rates, and assumptions:

For the 45 million hours of world wide audio holdings, calculation is based on the forthcoming format of 48 kHz/20 bit. It was assumed, that half of the material is mono, half of it is stereo.

For the digital video signal (standard definition, i.e. 625 lines/50 Hz or 525 lines/60 Hz, not HDTV) the bit rates/storage space depends on the method of coding: composite coding, which combines the luminance (Y) and colour difference signals (U/V), requires around 70 GB of storage space for the hour. Component coding, which for better quality handles and stores the luminance and the colour difference signals separately, requires 122 GB/hour (CCIR 601).

Taking the component format CCIR 601 for calculation, 8.6 million hours represent 1050 PB. Assuming, that historical material could be coded - without loss of quality - in the composite format, and further assuming, that this would be applicable to 50% of the worldwide holdings, the sum of all video materials would still require impressive 825 PB.

(NB: Because of the binary nature of Bytes is has become a tradition within the computer world to use the ISO prefixes kilo-, Mega-, Giga-, Tera-, Peta- and Exa-, as factors of 1024 (2¹⁰) instead of 1000 (10³). This practice is not generally followed by newcomers from the audio-visual world, partly because of unawareness, partly because calculation is annoying while the difference is generally negligible, and the conventional, standardised understanding of prefixes is better imaginable. For the purpose of this publication the colloquially acquainted factor of 1000 has been taken).

(10) There is an ever-increasing activity of research in the field audio-visual data carrier stability, carried out by archives themselves or commissioned by them from the polymer research institutes. Key players, amongst others, are the Library of Congress, the National Archive of Canada, the Australian Film and Sound Archive, the British Film and Television Archive and the Bibliotheque Nationale de France. Most recently these activities have been joined by the EUREKA project EU 892, EUROCARE AVIDA which is coordinated by the Phonogrammarchiv of the Austrian Academy of Sciences. Project partners are the Österreichisches Kunststoffinstitut, Vienna, the Deutsches Bundesarchiv, Koblenz, and the Centre for Polymeric Archival Materials of Manchester Metropolitan University.

[ 1] AES (Audio Engineering Society), AES22-xxxx, Draft AES Recommended Practice for Audio Preservation and Restoration - Storage of Polyester-base Magnetic Tape, 1996

[ 2] AES (Audio Engineering Society), AES28-xxxx, Draft AES Standard for Audio Preservation and Restoration - Method for Estimating Life Expectancy of Compact Discs (CD-ROM), Based on Effects of Temperature and Relative Humidity.

[ 3] N. ALLEN et.al., Factors Influencing the Degradation of Polyester Based Cinematographic Film and Audio-Visual Tapes. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[ 4] N. BERTRAM and A. ESHEL, Recording Media Archival Attributes (Magnetic), New York 1980

[ 5] K. BRADLEY, Restoration of Tapes with a Polyester Urethane Binder. In: Phonographic Bulletin 61/1992

[ 6] K. BREMS, The Archival Quality of Film Bases. In: Orbanz, E. (Ed), Archiving the Audio-visual Heritage. Proceedings of the (Second) Joint Technical Symposium, Berlin 1987. Berlin 1988

[ 7] L.S.BURT, Chemical Technology in the Edison Recording Industry. In: Journal of the Audio Engineering Society 10-11/1977

[ 8] COMMISSION ON PRESERVATION AND ACCESS, Ed., Preserving Digital Information. Report of the Task Force an Archiving of Digital Information. Washington DC, 1996

[ 9] M. EDGE, The Deterioration of Polymers in Audio-visual Materials. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[10] J.-M. FONTAINE, Conservation des Enregistrements Sonores sur Bandes Magnétiques: Étude Bibliographique. In: Analyse et Conservation des Ducuments Graphiques et Sonores: Traveaux du Centre des Recherches sur la Conservation des Documents Graphiques 1982-1983. Paris 1984.

[11] J.-M. FONTAINE, The Preservation of Compact Discs - Principles of Analysis. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[12] G.D. GIBSON, Audio, Film and Video Survey. A report of an International Survey of 500 Audio, Motion Picture Film, and Video Archives. Library of Congress, Washington DC. 1994. Partly published in: IASA Journal 4/1994

[13] I.GILMOUR and V.FUMIC, Recent Developments in Decomposition and Preservation of Magnetic Tape, in: Phonographic Bulletin 61/1992

[14] A. HÄFNER, The Introduction of Digital Mass Storage Systems in Radio Broadcasting: A Report on the Progress within the ARD. In: IASA Journal 3/1994

[15] F. HAYAMA et al., Study of Corrosion Stability on DAT Metal Tape. Paper read at the 92nd AES Convention, Vienna, March 1992. AES Preprint 3237

[16] J.HEITMANN, Zukünftige Archivierungssysteme. In: Fernseh- und Kinotechnik 50,7/1996

[17] W.R.ISOM, Evolution of the Disc Talking Machine. In: Journal of the Audio Engineering Society 10-11/1977

[18] S.K. KHANNA, Vinyl Compound for the Phonographic Industry. In: Journal of the Audio Engineering Society 10-11/1977

[19] G.A. KNIGHT, Factors Relating to Long Term Storage of Magnetic Tape. In: Phonographic Bulletin 18/1977

[20] F. KRONES, Guidelines for the Conservation of Magnetic Tape Recordings. In: FIAF, Preservation and Restoration of Moving Images and Sound, Brussels 1986

[21] F. LECHLEITNER, A Newly Constructed Cylinder Replay Machine. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[22] W.R. NUGENT, Issues in Optical Disc Longevity. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[23] A.G. PICKETT and M.M. LEMCOE, Preservation and Storage of Sound Recordings, Washington 1959. Reprint by ARSC, 1991

[24] J. ROTHENBERG, Ensuring the Longevity of Digital Documents. In: Scientific American 272(January)/1995

[25] D. SCHÜLLER, Sound Tapes and the 'Vinegar Syndrome'. In: Phonographic Bulletin 54/1989

[26] D. SCHÜLLER, Towards the Automated 'Eternal' Sound Archive. In: Boston, G. (Ed.), Archiving the Audio-visual Heritage. Proceedings of the Third Joint Technical Symposium, Ottawa 1990. 1992

[27] D. SCHÜLLER, Auf dem Weg zum 'ewigen', vollautomatischen Schallarchiv. In: 17.Tonmeistertagung Karlsruhe 1992, Bericht. München, etc. 1993

[28] D.SCHÜLLER, Behandlung, Lagerung und Konservierung von Audio- und Videoträgern. In: Das audiovisuelle Archiv 31/32, 1992 (1993), 21-62

[29] D.SCHÜLLER, Introduction to Bit Rate Reduction and its Consequences for Sound Archiving. In: IASA Bulletin 2/1993, 32-35

[30] D.SCHÜLLER, Informazioni audio e video. Dalla preservazione dei supporti fisci alla preservazione delle informazione. In: Tullio Gregory e Marcello Morelli (Ed.) L' eclisse delle memorie. Roma 1994

[31] D.SCHÜLLER, Jenseits von Petabyte - Zum weltweiten Speicherbedarf für Audio- und Videoträger. In: 18.Tonmeistertagung Karlsruhe 1994, Bericht. München, etc. 1995

[32] D.SCHÜLLER, Preservation of Audio and Video Materials in Tropical Countries. In: IASA Journal 7/1996

[33] L.E. SMITH, Factors Governing the Long Term Stability of Polyester-Based Recording Media. National Institut of Standards and Technology (NIST), Washington 1989

[34] S. SMOLIAN, Preservation, Deterioration and Restoration of Recording Tape. In: ARSC Journal, 2-3/1989.

[35] SONY (Ed), Archived Stability of Metal Video Tapes, MPG Technical Report, vol. 6

[36] J. VAN BOGART, Magnetic Tape Storage and Handling. A Guide for Libraries and Archives. Commission on Preservation and Access, Washington DC, 1995

[37] G. WELZ, On the Problem of Storing Videotapes. In: Orbanz, E. (Ed), Archiving the Audio-visual Heritage. Proceedings of the (Second) Joint Technical Symposium, Berlin 1987. Berlin 1988.

Chair: After six brilliant speeches, we have now come to the plenary discussion where you all can take part. I would ask the speakers of this day to come up here and to facilitate this discussion. I will ask everybody who is speaking to start by saying his name and institution for the sake of the records of this conference.

When I saw this invitation program, I was reminded of my childhood. I grew up in Sweden in a province called Gavleborg I stayed with my grandparents for several years, just under a magic mountain at the edge of a big plain. In this mountain there dwells a mythic figure, the mother Oma. And not far from this place, a few kilometres, there was a huge stone with runic inscriptions, about this high and this broad and with inscriptions on both sides and on the edges. And the picture of this stone is on the inside of the back cover of the invitation program. I am mentioning this, not primarily from the preservation aspect, but from the access point of view. The way for me to access this Memory of the World was to take a bike and ride these two kilometres and look at it. And so it was for other people, some maybe in other centuries used a horse, or a car to go to this place.

Although we have had an information revolution since this stone was made, a thousand years ago, the print revolution which occurred in Asia in the 9th and 10th centuries and changed history there, and in Europe in the 15th and 16th centuries, the way to find the information in libraries and archives didn't change all that much. But now we are in a digital information revolution and that will have a new impact on how we get information. This is the possibility - who will distribute the information? We don't have to go to the stone collections or the libraries and archives any more. They will come to you. Much of this conference deals with how it should be with libraries and archives and museums, to see how to change our way to accommodate this new experience. The discussion should relate to the questions of today - the Nordic approach for the Memory of the World and the conditions of preservation. Also you can make reflections and questions relating to the pilot project explained yesterday.

Fathi Saleh, from Egypt. Thank you for these different lectures that we have had today, they were all excellent. I have some few questions for the last lecture of Mr Schüller, which is of a very special interest for me. The first question we were talking about that when we store audio-visual archives, we should be as true as possible, otherwise it's unethical.

PROFESSOR SALEH AND DR SCHÜLLER THEN HAD A TECHNICAL DISCUSSION RELATING TO AUDIO ARCHIVING AND MAINTAINING THE INTEGRITY OF AN AUDIO DOCUMENT.

Erik Norberg, National Archives, Stockholm. When I see these gentlemen sitting in front of me, representatives from eminent libraries, archives and audio-visual archives, I can't help making a few special comments. We should consider a special kind of a Nordic corporation in this field, if there is one, it is obviously based upon a similar cultural tradition. It's based upon a similar legal framework, a similar constitutional basis, which makes things easier to cope with. I've seen in the latest years cooperation grow between archives and libraries and museums in my country, cooperation which has led to much more knowledge in paper conservation, in very concrete standards for permanent paper, and we are now working on finding aids, information exchange which we only can cope with in cooperation with our colleagues from libraries and museums. If I should recall the rune stone of Johan Mannerheim, that is, it would actually be a third aspect of this one, not only the preservation and the access ones. It's so interesting that this rune stone is a part, not only of a national cultural heritage, it's part of a common Nordic cultural heritage. That is why it is in this leaflet, I suppose. And to me, some of the advantages of the Memory of the World programme are not only that it gives us the opportunity to highlight single special documents in our archives, it could generate the cooperation between regions, an international corporation, between, quite clearly between different kinds of institutions and not the least, it will show us that we have to see the documents or the archives or the libraries in a common international context, it's all part of international cultural heritage.

Miroslav Musil, Slovak Commission for UNESCO. Ladies and gentlemen, allow me to come maybe from this highly sophisticated discussion to some more basic departure points. I would say that this conference, besides being well conceived and organised, is on a symbolic place in a Nordic country, and we see many examples of a Nordic joint approach, and this is precisely the message I would like to raise also to Mr Abid for the regional workshops. Because we have seen impressive and encouraging results of the pilot projects, but if I understand well, I'm not a technological expert, these results were obtained by individual, more or less unrelated approaches and this may not be the ideal way for the future for countries with limited resources like my country, but also many other countries represented here. We would appreciate that UNESCO and those countries that are the most advanced in the area of digitising, would provide us, a sort of optimum model, method to use. I was working with the WHO - World Health Organisation - so just as WHO provides the most recommended therapy for disease, we should be, maybe, recommended the most efficient therapy to cure our archives, our documents. Just let me give an example of how we tried to get a joint approach. At the 28th session of the UNESCO General Conference last year, we presented a draft resolution aiming at providing digitising equipment under the expert sponsorship of UNESCO for four countries - Croatia, Hungary, and the Czech and Slovak Republics. And the idea was, after having provided the equipment that the equipment would circulate among these countries to be to the common use for all of them, at least for the documents registered on the World Memory Register. Unfortunately the project received only a fragment of the requested budget, so for the moment it's stagnated, so if you happen to know about sponsors, please let us know.

So my idea is that we would appreciate to be able to avoid the children's diseases of those who start with trying to save and distribute their precious, most precious documents, and not to reinvent the wheel each time that a new country starts. And now, maybe I will express a naive reflection, or maybe too demanding, but I think that, if already thanks to the kind assistance of UNESCO and its regular program, these kind of digitising centres were built in some countries of the region, that it would be ideal if the same equipment could be lent or rented in advantageous conditions to neighbouring countries. I think I'm coming back to the Nordic example, I think that the Nordic union - or joint approach - is not only given by the geographical proximity but also to the historical interconnection of these countries. And we have several other examples of this sort of historical links, let's say the countries of the former Austro-Hungarian Empire, or the former Soviet Union, we have here Baltic countries and so on. So a lot of documents we are actually in need of are at the present in Vienna or in Budapest, so this is another reason why these countries should really be acting together. So, if I may conclude my contribution to discussion, it would be a suggestion of a conclusion or a recommendation of this conference that we should start a new type of cooperation that has not existed or functioned until now. I think this could be one of the really good outputs of the conference.

Jean-Pierre Wallot, National Archives of Canada. I entirely agree with the last phrase of my predecessor who just spoke. I have two series of comments, one on the sessions today, I must say that again like my predecessors who spoke, I was struck by the importance of the cooperation, and I would like to give you perhaps another, not another point of view, but a different experience, the Canadian experience in audio-visuals, since this has been the last session mostly about audio-visuals. In Canada, although Archives have been cooperating pretty much in the past ten years, perhaps a little bit more, it is not a natural, let's say, inclination and cooperation within libraries and archives has not been particularly great and with museums and other heritage institutions, even less. Except in the very specific circumstances where, for instance, an archive is in a library, or a library is in an archive or things like that.

When we came to the audio-visual problems, these were a major problem in Canada. Don't forget Canada is only 30 million people, but it is spread over thousands of kilometres. There are so many and we're beside the United States, with the number of production units of TV channels, recording and so on is so great that it was very difficult to come to a resolution, and we're losing a great amount of audio-visual heritage in Canada. So we started in 1993, a joint venture study, first of all, by all the stakeholders. By stakeholders I mean the producers, the distributors, the actors, the broadcasters, the archives, libraries, all those who have audio-visual material, made the study of all the holdings across the country, at least the ones that could be identified, and we came up with - quite fairly I think - good ballpark figures of what we had, what was in danger and so on. And the proposal that came out of this report - which is available if anybody wants, you just have to write to me - is essentially that to create a consortium of all these stakeholders to monitor and maintain a program of preservation of the past, but partially work on the present and future creation of audio-visual records, so that they are preserved in a sense that the budget for their safekeeping is included in the production and things like that. This report recommends in particular, the creation of this consortium, but also the centralisation of regional facilities, because it is impossible to build 500 repositories to keep safekeeping audio-visual material. It takes very rigorous standards of temperature. Mr Schüller has expressed that, so if we had one per region, that would already be a great achievement.

Secondly, to have one or two at best, excellent centres for the reproduction, in particular for the reconstitution of movies, whether black or white, or colour. These are very labour intensive, very technical and costly equipment and so on. And there's a whole list of proposals. But what is very striking, is that out of the about 3 million hours of either audio or audio-visual material that was identified as being actually preserved, part of it not being in danger was only identified as the core value that had to be saved at all cost - 250,000 hours. And this would cost $45 million over the next ten years, spread over ten years, to save the past. What we've got left of the past, and not much is left. Imagine if you were in a big country like the United States or France or Germany where, I presume, the old things are extraordinary, and so it's a costly business, and I don't think you can ever attack this audio-visual preservation unless you have a kind of a collective effort done by a series, and you specialise and you interchange your specialisations otherwise it's a loss, I mean you may as well write it off.

This brings me to my second sets of comments. I'm changing my hat, I'm speaking a bit like the Chair of the International Advisory Committee on Memory of the World. And it's not particularly about what has been shown here at this conference, it's just as a comment in general. If you read the document that has been developed by Jan Lyall and that you have received at the beginning of this conference, there are a series of criteria for Memory of the World. Now in the two committees that I've worked on this issue, there were two tendencies. One was, let each country decide what they believe is at Memory of the World kind of value and, or else, let's select. And in the end the select tendency won the day, so please don't submit all your holdings to Memory of the World, only those that you really feel, take as an example I would say the world heritage type of approach to these matters. You can still identify nationally significant, regionally significant records, but only a part of these would be Memory of the World. And that would help the Committee and I'm sure they're going to thank me for, they're going to think I'm very charitable for them, to not get all the holdings of all the archives of this world, proposed for Memory of the World label.

Ian Wilson, Archives of Ontario, Canada. Many of the presentations in the last day or two have focused on the technological issues and the technological problems. Listening to the presentations on the pilot projects, I gather simply putting multi-media holdings onto a CD ROM and marketing is straightforward now in a technological sense. And the presentations today, yes there are challenges, there are problems, but there are some solutions that are coming down the line. The problem with all of them is that they are expensive, and the speakers that we've had today and only one or two, particular, Mr Rugaas this morning, touched on, hinted at the funding issues, the funding problems around it. And I was wondering if there's anything we can do here in this session to try and explore how do we get at the funding issue. Because I think we have to realise we have an extraordinary asset, the concept Memory of the World is a very powerful one, and the speakers who led off yesterday morning, and especially the Minister, explained it far more eloquently that I could in terms of just how powerful a concept it is. More powerful I think than the Olympics. I think in terms of getting corporate sponsorship, our problem may be more, how do we manage it so that it's appropriate and sensitive sponsorship and that it doesn't overwhelm or take over the Memory of the World.

I think it's timely, because in Canada as we look at planning for the next few years, the archival system is starting to say, 'Well, we have an extraordinary opportunity coming up around the millennium'. The media, the popular media will be hungry for content, for retrospectives on the last thousand years, on the last hundred years. We have an opportunity as archivists and librarians to make it clear that the heritage of the last millennium, the heritage of the last century particularly is a documentary heritage, that it's multi-media, it's all inclusive, it's diverse in terms of its representation but it's extraordinarily fragile, and as we've heard here, saving it, preserving even a portion of it is going to be expensive and require major social and political commitment to do it. It's also valuable, and I think what we have to recognise is that the current media devour content, and as librarians and archivists we have marvellous content. The question is, how are we going to ensure that we use that value to help preserve and maintain it? Microsoft is buying up rights to museum and art gallery digitised holdings. The Vatican Library and IBM have gone into a partnership to develop digitised holdings from the Vatican Library in a way that will, we think in time, be not only cost recoverable but also turn a profit, and do it while making that material more cheaply available than it has been. For anyone coming from overseas to use the Vatican Library, it's been a very expensive process. For anybody who can avoid that cost by staying at home, they will pay money to do it, and certainly some of those pilot projects we saw yesterday would have tremendous appeal, not just in their country of origin, but to the many places that scholars and in fact, immigrants have come from and settled in the New World and other parts of the world, it is still an extraordinarily valuable heritage.

So I've been curious, I guess, whether any of the projects that have been undertaken so far have piloted new ways of funding, of self-funding, of different types of partnerships with the private sector to ensure that we can do more of this, and some of the really marvellous stuff will help sustain and support the things that might not otherwise get funded. How can we develop and use this potential we have of the idea of the concept of the value of our holdings to help ensure the preservation of those holdings?

Abdelaziz Abid, Programme Officer, Memory of the World. Thank you Mr Chairman. I am very grateful to these very useful comments from our colleague from the National Archives of Canada. And I would like to react also to what Mr Musil from Slovakia was saying. I think that I'll try to react to both comments at the same time. It's true that our pilot projects have been developed in a very individual, unrelated, uncoordinated manner. They were projects to test different sorts of ideas, and mainly to test the digitisation as a potential method of enhancing access and drawing attention to documentary heritage. Now that we have developed guidelines and the technical committee which is chaired by Dietrich Schüller has recommended a set of standards which you'll find in one of the appendices, one of the annexes of the guidelines, that was the agreement of the technical committee after three meetings and very technical discussion. This is Appendix D. I think that from now on, all projects have to be carried out according to recommended standards and therefore, there will be unity, there will be a coherence, and there will be even more, I would say, we should find some sort of editorial line for all products that have to, that will be developed under the program. I think that we should seek a kind of very strong recognisable Memory of the World signature, so that as soon as you see something you will recognise it that you see, oh, this is interesting because it brings the signature of Memory of the World. I think that the Internet, from what we have seen this morning from our colleague from the National Library of Norway, I was extremely impressed by the appeal of the presentation of this morning, and I would be very much interested in developing, for Memory of the World, a web, very strong site with as much images and sound and everything from Memory of the World as possible, to enhance wide access, world-wide access to this material developed under Memory of the World.

Very short comments about the regional project mentioned by Mr Musil and then the funding: I think that Mr Musil is completely right in pleading for a strong regional approach for the Memory of the World project. And even if we have not been successful with the General Conference of UNESCO to get the necessary funds for our regional projects with Slovakia, Hungary, and the Czech Republic and Croatia, I think we still should try to get that project done in partnership, we have to work together. And the small amount of money which is available to us, probably the best use of that money would be to have a group of people trained and familiar with the recommended standards for digitisation and Memory of the World. And probably now the most suitable place for that, and the area I would say, is the National Library in Prague. And I was talking to Mr Knoll and asking him to host, for some time, a group of colleagues from these countries so that you will have a core group of people that would be able to carry out a regional project. I still think that we will be able to find some money for such a good idea, and there is very spectacular material from the region which we could use for this project. And I think that Mr Wallot was saying we have not to list and to nominate everything for the World Register - I fully agree with him of course - but when we think of some of the beautiful material we have in central Europe, I think that we should start with a registration of that material and then embark on a strong regional project, once we have trained a group of people.

My last comment would be about funding. That's a huge problem, and I think that we have reached the level where we really need to launch a strong fundraising campaign and we have started exploring ways for bringing funds, but serious amounts of money for the projects which are on that list waiting for funding. I think there is ground for us to be optimistic, there are quite a number of partners who seem to be interested, and who want to cooperate and I think that all these aspects are inter-related. That we will be able to attract funds if we are able to develop very strong, powerful and appealing projects. Thank you, Mr Chairman.

I would just like the chance to take the opportunity to make a short additional comment. The work of the subcommittee on technology does not end - did not end - with the recommendation which you find printed in the general guidelines. This is only the recommendation on digitisation and this will be constantly reviewed and possibly, in the near future, be reformulated. We have been asked by the International Advisory Committee to take on board a New World look. One is the collaboration of standards to preserve the originals, whatever they may be. And this group is just working, and we will have first results by the end of June for internal discussion. I would like to inform you that another working group under the Chairman, Chief Adolf Knoll from the National Library in Prague, is working on guidelines on the harmonisation of access to electronic documents within the Memory of the World domain, in order to ensure that the access is in more or less the same way and not every electronic document has to be accessed with different programs.

Virginia Betancourt, National Library of Venezuela. I was also happy to receive the information in regard to technical criteria and for reproduction of audio-visual materials, and of course have been reminded of the high costs and risks because at the National Library of Venezuela by law we are responsible also for the audio-visual archives. But of course, I'm also satisfied to see the result of this Nordic cooperation and wanted to insist in regard to Mr Abid's comments, that is the sponsoring regional project and the newspaper projects - 19th century projects in Latin America - is a pilot project of UNESCO, which would imply a lot of training of personnel not familiar with bibliographic control of serials and automation of serials, which is a complex operation. So I would recommend then that emphasis should be given to up-to-date information to interested parties in regard to these technical criteria, but mostly to promote cooperation among library managers. You can't, from UNESCO, force regional cooperation unless the national librarians are convinced that it is required. It is easier to work by yourself, things are done quicker and you don't, if you are the nucleus of a cooperative association of national libraries, you have to spend time and effort and resources in this endeavour. And therefore it is better to work by yourself as the initial idea, it's cheaper, it's quicker.

But I would say that the only way for projects to have impact is if they are not only interested in the content, but also with the wider range of images or information, whatever the content is. Therefore I would greatly insist on the fact that not only the Nordic experience, but also the Latin American experience - which is Iberoamerican - be taken into account, and basically the result of that experience is you have to have governments involved in our countries - we don't have so many industries around - and you have to have managers really recognising that it is indispensable to work on a cooperative basis. And thirdly, I think that deciding which country would in a region or in a sub-region be responsible has a lot to do with the degree of development of technical and managerial development. Not so in how complex, how interesting the collection is, but how able is the manager to handle the collection, to sell the collection to others. So I've been satisfied with what I've heard, but I'm missing this strategic effort to convince others within the region at various levels.

I come from Barbados in the Caribbean, and like previous speakers I must say I am overwhelmed by the level of presentations. I have two questions, one I'd ask now, I represent a radio and television station, which may be considered a fairly small outfit in that it's on, our island is 166 square miles. We have about 34 years of archival material which must now be stored. The question is, and recognising that there may not be totally agreed standards as yet, is there an approximate cost of storing one hour of archival material using today's technology?

Isaac Mabhikwa, Zimbabwe. Within the demonstration this afternoon on the degeneration and loss of quality and the eventual total loss of the property in audio tapes and then our records. Can we have it then on record that the digital signal has been exposed and tested for time, wear, and loss of quality that it is now the final answer? And also there was mention of film yet to be, or rather some work still going on on the storage of film and, bulky as it is, what about the laser disc, is it not good enough as a storage format for film? And if not, what are its limitations? And maybe from UNESCO, do we have experts working - this is on the film issue - do we have experts working on it to try and find out the best possible format for the storage?

Ray Edmondson. There's no final answer, I wish there were. It seems to be that the experience of people working at audio-archiving every year, every six months, there's a great final answer, somebody discovers new technology or a sudden solution to everything. There never is, and we learn to make very conservative judgements about the preservation choices that we make. So we know, for example, at this stage with film, the most cost-effective way to preserve film is on film digital alternatives, if you're going to maintain the same level of picture and sound quality, it's just not a possibility, it's just far too expensive and less reliable. But we always face choices about quality as opposed to other things, and these are always value judgements and the judgements are going to be different in different circumstances, and there never is. So, on the one hand you can be very purist about, saying we can only do this and nothing else, on the other hand you can make some judgements.

Let me give you an example from Australia. Twenty years ago, when there was very little money being spent on preservation in Australia, I think it was a practice to copy 35mm nitrate film onto 16mm. This gave you a much reduced picture quality, and today people who look at that practice are appalled that this was ever done. One answer to that is, at the time it was better to have that than nothing at all. At least we now have those films surviving in some level of quality; if that had not been done, they wouldn't survive at all. This is the value judgement that was made at the time, in circumstances that applied at that time and do not apply now. And they were always difficult value judgements, and in the light of subsequent history we'll never get it right, we'll always be blamed for doing the wrong thing. So it always involves the kind of judgement in any situation. Sometimes you must, you might make a judgement that you have a certain amount of good storage available for a very large collection. OK, you choose, you make a choice about what goes into good storage and about what doesn't go in. But at least you make those judgements consciously and defend them, and they're the basis on which you build towards other things.

Dietrich Schüller. Let me join this question more from a technical side. It is impossible to store the amount of digital data which is necessary to represent full film quality on any optical disc which is available now. So the laser disc you are talking about may be a good tool for dissemination of films in an acceptable video quality, so that besides the critical scenes you would not see any difference.

Rolf Dahlø, Norway. I'm speaking here as the Chairman of the International Organisation for Standardisation - ISO - TC46SC10 Physic Information and Documentation: Physical keeping of documents. It's a very long name but it explains what it is all about and of course, some of the questions here bring up the questions of standards for preserving documents, and I think that is one of the fundamental issues that has not been brought forward in this conference yet.

We have managed to make a standard for permanent paper, that has been referred to, and I'm must confess that I know too much about all the problems surrounding paper as an archival medium that I'm not very happy about this standard, but at least it was an improvement and of course, this is a standard that is under attack from people with commercial interest, and it is a standard that will need to be safeguarded by the preservation community. I don't think that everyone of you is aware of that. And we have been given a scope by ISO that says that we are responsible for all physical media except photography and optical memories and we have given some thought about digital media and also audio-visual media. Our answer so far, is that there is not sufficient knowledge - scientific knowledge - that would allow us to draw up any kind of standard for some kind of permanent media in these fields. The good thing about digital information is that you can have an internal backup without quality loss, but that is all you could really say for it. Some committee in ISO tried to draw up a standard for permanence for ROM and that standard was saying that for the period that the manufacturer could expect that a ROM should be useable, you should devote one third of the ROM for the facts and you should use one third of the ROM for the information and you should use one third of the ROM for moving information from bad sectors to good sectors. And I think that is a very good example of what problems we are facing with digital media.

Sir, I'm just warning everyone that is expecting that future will solve the problems, well, future isn't here yet, we are faced with some real problems and some of those problems ought to be answered by standardisation and by going into it - national standardisation - and there is a lot to be done and there is a lot to be achieved by national standards, because some of you will obviously have some help in your own countries when you are faced with problems and you can refer to that. This is the way to handle it, because this is the advice of a national standard. Thank you.