Converging Frontiers

Malta

October 21-23, 2002

The summation of human experience is being expanded at a prodigious rate, and the means we use for threading through the consequent maze to the momentarily important items are almost the same as in the days of square rigged ships. We are being buried in our own product. Tons of printed material are dumped our every week. Many of them become lost; many others are repeated over and over. [Vannevar Bush, Science is not enough, New York, 1967.]

When I took up the responsibility for training librarians and archivists at University College in 1990, there were few observers of the professional scene who could have predicted where we would all find ourselves just twelve years later. My colleagues in the Department were, almost without exception, teaching librarianship and archives management much as it had been taught in the days of Esdaile and Jenkinson. My own career, then represented in the completion of the first phase of ESTC, had, however, taught me to be observant; and, not for the first time, I found myself in the uncomfortable position of having to persuade both colleagues and the College administration that, whether we liked it or not, things would have to change. Persuading administrators is, as you all know, a tricky and delicate business, since, in all places and at all times, they believe they possess a monopoly on truth! I had had to do this with three successive Directors General of the British Library, and had come to realize that when you need an administrators’ support it is essential to allow them to believe that the radical changes you propose are entirely their idea – incidentally, this has the added advantage of providing you with indemnity insurance if things go wrong!

ESTC – the Eighteenth Century Short Title Catalogue originally, but now the English Eighteenth Century Short Title Catalogue – was, at the time of its conception in 1975, and remains as such today, the largest retrospective machine-readable project anywhere in the world. It now describes over 500,000 items, in 4,000,000 copies, located in over 1,000 libraries in every continent. Many who attended the June Conference at the British Library in 1976 urged a somewhat limited project which could be achieved using cards and paper, and though there were a handful of visionaries at that conference (mostly from America) I could see that it would be an uphill struggle to finance a landmark project using the then-emerging technology. Though very crude by contemporary standards those little Singer computers (which had to have their tapes threaded with a biro!) nevertheless contained the seeds that would one day make the machines we now take for granted possible.

In order to see how far we have come in a lifetime, let me remind you that in the year 1964 the following events were noticed by few, but can now be seen as critical:

1) IBM released the Magnetic Tape Selectric Typewriter, using 16mm magnetic tape which could hold 8,400 characters on 35 feet of tape, and described this amazing machine [I owned one] as the world’s first “word processor”.

2) The Photon ZIP phototypesetter was installed at the National Library of Medicine in Bethesda.

3) Richard Clay was the first company to install a computer typesetter – the Linasec – developed by Compugraphic.

4) Rocappi at Otford in Kent started work on the typesetting of Dylan Thomas’s Collected Poems for Dent.

5) John Duncan contributed an article to the Penrose Annual on computerised typesetting entitled “Look! No hands!”

6) HMSO successfully produced at the Nautical Almanac Office the Astronomical Ephemeris from punched cards using a Monophoto filmsetter.

These developments were to have a profound effect on the way in which information is both created and stored, but there were no signs in the academic press of the decade that followed that anyone could imagine them. But they certainly had an effect on me; and when, in 1965, I started Scolar Press, one of the first innovations I instituted in the crucial area of photographic reproduction was the incorporation of a computer to control camera exposure; and somewhat later the development process for photographic negatives. Introducing yet more innovations – most of which I observed at the great Milan Printing Fair in 1968 – would have to wait, as the National Graphical Association (the printers’ union) – also established in 1964 – had set its face resolutely against change. Of course, they lost the battle; but it took several years before proprietors could modernise their printing establishments.

It is a principle accepted by professional archivists that the efficiency with which a company, organization or government functions and is able to make successful decisions is directly related to the efficiency with which its records are retained, stored, indexed and made available. Any of us who have had occasion to look at procedures in the administration of poverty-stricken sub-Saharan countries knows the truth of this. But it is a truth which those responsible for administering an organization frequently fail to respect, because keeping records never appears to represent resources well spent with evident return. Until, that is, things go wrong.

As we all know, things went disastrously wrong in the weeks prior to 9-11, when terrorism struck its nightmare blow against the United States. In the public, and not-so-public, debate since then about how such an outrage could have been committed, apparently without warning, there have been significant revelations from both the FBI and the CIA, not to mention the cobweb of committees within the Defence Department. No one agency, it seems, was listening to warnings from any other agency. The political machine of summer 2001 had other things on its mind; and it is an irony of fate that, while American security agencies had at their disposal an arsenal of sophisticated electronic dveices, the terrorists were employing an everyday technology like email and the Web with which to communicate and coordinate their plans – much of it, as we now know, buried in the pages of pornographic sites. It has taken months of careful archaeology by security agencies world-wide to begin to comprehend the pervasive nature of the Al-Quaeda network. Much has been discovered about the organization from files which were not destroyed, once America and her allies, had declared war on terrorism, but much remains to be discovered. The lesson we must learn form 9-11 is this: running a terrorist network is not much different from running any organization: there are, really, only two kinds: those that succeed in their mission, and those that fail.

If the world of the 1990’s bore a steadily diminishing resemblance to what had gone before, the world in the first decade of the new millennium will present those responsible for administering an organization with a whole new set of problems. We have witnessed, in the past four years, the omnipresent mobile phone, without which some organizations quite literally could not survive. These are increasingly, together with the PDA [Personal Digital Assistant], necessary devices in local as well as national government. The pervasive use of these devices in large companies is well-documented: you cannot administer Fedex without them. We should remember that the introduction of the telephone demanded arrangements to be made in government departments in order to record what had been agreed between two parties discussing a problem using a medium which left no trace of its existence. Of course, without accurate transcripts of a telephone conversation much of the memorandum traffic annually deposited in the Public Record Office at Kew must be taken on trust.

In theory, email traffic is a little easier to archive and scrutinize; but how many organizations currently have procedures in place to accomplish this? And how is such a vast accumulation of casual traffic to be weeded? And since email has a sender, a receiver, and any number of other recipients, will it be necessary to keep it all?

As we know, if we have ever sent an important document by one of the numerous courier services, the progress of a document from London to Bangkok can be traced on the Net. That is only possible because at every point in the document’s transfer from Britain to Thailand there has been electronic tracking input. If your document does not arrive, it is theoretically possible to determine precisely where things went wrong. For a divorce settlement between Jack and Jill, posterity may well be able to tolerate evidential loss. But what of a document forming part of a proposed treaty between two governments? As far as email is concerned, how can we be sure that the message retrieved from an individual’s file is the one sent or received? This problem is reminiscent of that posed by the photocopier which made it possible for duplicate files of considerable size to be accumulated within an organization. I know of at least seven separate files purporting to contain the critical documents relevant to the history of the ESTC: they are all different in emphasis and content. Should we keep them all? How are they to be linked?

Perhaps even more intractable than straightforward email is SMS messaging, with its elliptical vocabulary and non–existent grammar or syntax. With the mobile phone rapidly developing as a hand-held roving office with capabilities previously only found on PDAs, it is hard to see how agencies are going to be able to control and, where appropriate, archive such traffic.

This word linked is crucial; for, as we all understand, the Web is built upon the principle of linkage. Subordinate indexing is not very easy to achieve in the Web environment; but coordinate indexing – linkage – is almost trivially simple. This is the one vital area where the transition from the index card to the computer represents a transformation rather than a translation. In many of the applications I observe being adopted by the research libraries and archives it is clear that this distinction is still not grasped by those responsible for developing information systems. As we know, libraries have translated their card catalogues into electronic form, but in the process much has been lost or blurred, and new errors have been introduced. All of this lends to bibliographical enquiry an uncertainty as to what can be trusted and what not.

It has been clear for at least a decade that the ever-growing volume of electronic information generated by governments must be protected from alteration or abuse, and there have been some ingenious solutions suggested. But any security procedure can be penetrated. What is perhaps more worrying is the fact that governments, as well as large corporations, have been for some time engaged in what is termed out-sourcing: which some regard as a potentially dangerous, if financially beneficial, solution to the demands of the bean-counters. We have witnessed some calamitous consequences of out-sourcing in the prison service; what makes us so sure that in other equally sensitive environments we shall not see comparable mistakes?

All this may seem to you as obvious as the fact that tomorrow will arrive; but there are implications in the future that must be faced. Let me give you a simple example. The Metropolitan Police records of London were, from 1980 to about 1990, managed on a microcomputer that dazzled the world when it first appeared: the Sirius I, designed by Chuck Peddle. Peddle was both a genius and a visionary. He joined Motorola in 1973 to work on the 6800 microchip, but left to develop his own: the 6502, the most successful microchip of the pre-Intel period. Jack Tramiel of Commodore recognized his obvious talents, and the result was the world’s first microcomputer for everyman: the Commodore PET [Personal Electronic Transactor]. In those heady days the technology that would transform the world was being exploited by Stephen Wozniak and Steve Jobs, creators of Apple; Bill Gates, founder of Microsoft; and Chuck Peddle who turned his attention in 1977 to developing the Sirius: the most advanced computer of its time, and for some years after its first appearance in 1979. It preceded the IBM machine which was, by comparison, bulky, slow and cumbersome. The tragic fate of the Sirius, which was quickly incorporated into numerous large organizations in the United States and Europe, is a story which has yet to be told. But its collapse under Victor left a trail of devastation for those who had depended on its amazing qualities for over eight years. At one time I owned three machines: one for work, the other two for spare parts! They were finally junked in 1995.

The point of this little digression is to remind you that when making arrangements for the retention of electronic records it is necessary to consider carefully how it is proposed that both hardware and software can be migrated upwards without loss of data. This is no trivial matter. For one thing, there does not exist in either the United States or Britain a software library - which will be essential in the future if we have to read documents (which may well come from highly respectable sources) that have been produced on obsolete hardware using obsolete software. In the period between acquiring my first Sirius in 1980 and migrating all my files to the IBM standard in 1995, I used a variety of software packages for the preparation of reports and research initiatives within the British Library. None of the software used for these purposes can be found today; and certainly not the software I developed myself!

It is an accepted fact that all records are subject to two distinct phases: in the first phase they are records, and as such dealt with according to the prevailing principles of records management. In the second phase, having been appraised and weeded, they become part of an archive. Where paper is concerned these two phases have never been in conflict; but with records in electronic form the situation is dramatically different. The use of a Sirius computer to handle records prior to examination for retention or disposal presents no great difficulties. But the availability of a working Sirius fifty years from now to verify and archive files stored in the unique disk format invented by Peddle is quite another matter.

You will probably think that I am exaggerating the nature and scale of the problem. But reflect on the fact that for some years now we have in the developed countries been exporting our now inadequate hardware and software to developing countries, where a simple telephone conversation is still a miracle! Just as we dumped our letterpress machines on Africa and India when printing by offset lithography became the standard in 1965, so now there is a brisk trade in obsolete equipment and obsolete software throughout the developing world. And the beneficiaries of this trade are not exclusively the deprived countries of sub-Saharan Africa!

There have been many responses to what some see as a crisis in the records management and archives community: how to balance ever-dwindling resources with objectives which have both popular and governmental support. The word most commonly encountered in discussions of how to achieve the impossible is co-operation. As used by those who have the least resources in countries like South East Asia and China the word means asking a richer country to assist in resolving its difficulties; as used by those with the most resources it is a way of diverting attention from the need to do more for less. As I was preparing this paper the Canadian Government announced [October 2] that a new agency was being formed by amalgamating the National Library of Canada and the National Archives: the new agency is now called Library and Archives Canada. The official prose describing the benefits expected to accrue from this astonishing decision deserves quoting. As Heritage Secretary Sheila Copps puts it:

“The Government of Canada is committed to making history, culture and Canadian voices accessible to all Canadians and to encouraging research, discovery and the sharing of knowledge. The creation of this modern, dynamic world-class organization addresses an increased public appetite for knowledge about Canada. The new agency announced today will strengthen the visibility, relevance and accessibility of the collections and services of both the National Library and the National Archives. … We will provide leadership and support to Canada’s archival and library networks while continuing to represent them on the international level. We will seek to develop more partnerships with other communities and knowledge management professionals to create networks and synergies. As a result, the Library and Archives of Canada will become better positioned as a leading knowledge and information management organization – unique in the world.”

I spent many years of my professional life advising Chief Executives and Directors General of the British Library on matters to do with automation and staff training, so I recognize flannel when I read it! There is a principle at work here: which is to inflate the rhetoric in defending a cure to match the reality of the crisis for which the remedy is proposed. I don’t know what the experience in Malta is, but I can assure you that every initiative to introduce the concept of partnership in the British Library in the past ten years has been little short of disastrous. And the exercises which have been undertaken to pave the way for such developments have cost vast sums of money which could have been put to better use in improving services to the public. Time and again I read in the library press how this or that initiative is going to improve services, provide better value for money, &c. &c. Such promises belong to the world of Erewhon where more is less and better is worse.

One of the treasures of the collections in the Malta library is a copy of Ptolemy’s Cosmographia printed at Rome in 1490. It shows Malta at the centre of the Mediterranean – itself the centre of the known world. And of course we know the extraordinary role this little island has played in the history of that body of water before you reach the Pillars of Hercules. It is said that the inscription above the pillars read: Ne plus ultra. With the development of post-medieval Europe that inscription no longer applied, for there was everything beyond! And here we are exploring some of the ways in which the records of civilisation everywhere on this planet can be managed so that we never again find ourselves in a Dark Age, such as Europe endured for a thousand years. It is a chilling thought, but if we can make sense of it all – and I include the flannel as well as the contradictory issues which seem to plague all institutions these days, then we will have done well. But to achieve this we must look at the problems facing the management of records and archives with surgical precision. There are not going to be any easy solutions; and though we must respect political niceties, the politicians will be long dead and we shall still have the fundamental problems to solve.

That we have been engaged in a love-affair with technology for fifty years can hardly be doubted by any rational observer of events since 1964. After a lifetime’s involvement with technology I am disinclined to believe that you can solve the problems created by technology simply by throwing more technology at them. There are deeper issues at stake here than simply making Bill Gates richer than he already is. That is, I believe, why we are here …

Robin Alston