EDINA 2010
e-journals are forever? Presentation April 2010
by Derek Law
I began my career as a Serials Librarian forty years ago next month. I published my first article two years later and for a variety of reasons was recently trying to track this down. It was about the so-called silver fork journals of the 1820’s. Interestingly (at least to me) I was able to use the original journals because I was then at St Andrews University where they formed part of the collections built when it was a copyright library. As with all copyright libraries their role was to build a collection and then keep it forever. I could no longer remember the journal title, nor is the journal in which I published (an obscure bibliographic magazine) readily available. So I know that it’s all there somewhere, but probably can’t track it down without a visit to St Andrews.
The position is not dissimilar with the digital world and I want to suggest that to a degree we are obsessing about the right things but the wrong material. And so I’d like to focus on four broad topics within the overall theme of the meeting. The first is serials themselves, the second is collections and collection building, the third is to identify the agents who have responsibility for preservation and access and finally but related to the last, trust metrics.
Serials. As a young serials librarian I can remember agonised debates about the difference between periodicals and serials. There were (and are) journals, yearbooks, annuals, periodicals, serials, newsletters, occasional publications, regular publications, series, acta, newspapers, magazines, partworks and so on. I learned that the average publisher published fewer than two titles and was typically a learned society with a journal and sometimes a newsletter. I heard a paper two weeks ago in which the well known publisher Arnoud de Kemp confirmed that this is still the case. And yet we continue to obsess about the large STM publishers, which are, in a proper sense, aberrations. I will say a little about them later, but what I want to focus on is the born digital material of average publishers. I am involved with two learned societies in the field of naval history which publish standard works with an international reputation. The societies are small – under two thousand members – and struggle hugely with issues of digitisation and making material available in what is essentially an amateur and part-time operation. The paper model for such journals is that you appoint an academic editor, base the journal in the editor’s university, then the university in effect subsidises the journal and meets the overheads in exchange for esteem and review copies for the library. No such model has developed for electronic publication. And it’s not for want of trying. I know from personal experience that approaches to university libraries and computer centres and to national libraries are met with bemused lack of understanding of the issues. So we have a big gap in developing a digital model for small learned societies.
Collections and collection building. I have argued for some time that one of the biggest failures of my generation of librarians is in failing to develop any theory and philosophy of collection building for born digital materials. There are odd exceptions of which far the best for me is the University of Texas Human Rights Collection. If we look at the history of libraries the majority of what is collected and the things which make them great are the non-commercial materials. It is the archives, the special collections, the government and NGO publications, the grey literature and ephemera which define wonderful libraries, not the long tedious stacks of largely unused scientific journals held in a thousand other collections. Our response to born digital material has then been to ignore it. Faced with the internet we have formed committees (we are very good at that) to negotiate licences from publishers and gain imperfect terms almost exactly the same as those obtained by our neighbours. And then we have gone on, not to embrace born digital material, but to digitise the bits of paper we already have. We have completely failed to embrace the wealth of material which is born even in our own institutions. I know of no university which has a comprehensive collection policy for the research outputs, research data, blogs, wikis, podcasts, teaching materials, formal records, institutional archives, PhDs, Masters Dissertations and so on of the institution. At least a proportion of this will fall under the broad heading of serials I described a moment ago. As the open access and creative commons movements inexorably spread this material will simply be lost, unless policies are put in place and action taken.
Agents. I have a very firm if personal view on this, which is that the creators institution has an irremovable responsibility for preservation and that this is normally a university. That responsibility may be sub-contracted in practice but it cannot be abdicated. But let me look at some of the agents and their roles.
a) The publishers. As I have already stated I do not believe that the majority of publishers have the knowledge or competence to undertake preservation. Remember that the large STM publishers are abnormal. One of the oldest library jokes is to ask to see a publisher’s grandmother. They don’t exist because they have been sold. It is a publisher’s responsibility to sell the last copy. Equally, why should we expect publishers expensively to maintain archives which do not make money? It is not their role nor should we expect it to be. The same applies to the recent concept of aggregators. And talking of recent the lifespan of publishing houses is shockingly short. Takeover, merger and bankruptcy make them unreliable agents. This week, Google published figures on government attempts to have material removed.[Oddly Brazil topped the list]. It is not then fanciful to imagine a government, or political party or movement buying up a publisher and suppressing material. Think of creationists buying up journals and suppressing articles, for example. Difficult in the print world, but quite possible in the digital – unless LOCKSS type mechanisms are in play. As part of the digital world we need to have mechanisms which guarantee access to orphaned or suppressed content.
b) The national libraries. Again one might argue that the nation state – a construct of the mid-nineteenth century – and by extension national libraries are recent and therefore untested resources compared with the universities which have existed for nigh on a thousand years. However they have a clear responsibility for commercial material which one would expect to be collected under legal deposit and will continue to build their own special digital collections I have no doubt. One might argue that historically their track record on automation is weak compared with the major research universities, but that is very different from their record on acquisition. The BL has also played an important role within the community in helping to develop the UK Research Reserve, which also offers an interesting potential precedent in the digital environment. I have to say that while relationships between the BL and the HE community have been excellent and productive under Lynne Brindley, this has not always been the case. Personalities, changed policies and priorities and the inevitable funding crises do not make me, at least, wholly comfortable that this relationship can be taken as a given, rather than a relationship which has to be permanently worked on and refreshed and which could change.
c) JISC. I have a very soft spot for JISC having been part of it since it was invented and having had a hand in its first act – the first ever national site licensing scheme anywhere in the world, which brought us BIDS. JISC is, of course, not the agency to act as a digital archive, but it has a track record of being the agent for change, the agent for developing standards, the agent for building communities of practice and the funder of relevant research. It has supported services from BIDS to SUNCAT, from MIMAS to EDINA which provide and can in future provide the backbone of a coherent national strategy.
d) The Universities. I have myself no doubt that it is the universities who must bear the responsibility for the preservation of much and possibly most of the born digital material. Universities exist for the creation and transmission of knowledge. It seems to me they have an absolute responsibility for managing the preservation of what they create. Many universities will do this themselves but need advice and support on what is needed. Others may choose to sub-contract it to national libraries or to agencies such as EDINA, but I have myself no doubt that it should form part of their core values. I would go further and suggest that universities should have mandates in place to ensure this. Interestingly, I believe that there has never been a better time to address this issue. I was recently in Washington for the launch of the Blue Ribbon Task Force report on digital preservation. It is an excellent and thorough report with excellent background work but rather timid recommendations – such as researchers should be asked to consider whether they might deign to deposit research data. Now I say this in the spirit of “there but for the grace of God go I”, but the recent fiasco at East Anglia seems to have occurred through an abdication of managerial responsibility and childlike reliance on scientists knowledge and understanding of information management, freedom of information legislation, and computer audit – when in practice the researcher is the single point of failure in the system. It should provide a wake-up call for every university to treat its born digital assets with care and to develop and implement mandated institutional policies.
e) Finally trust metrics. If I give printed matter to the Bodleian Library I know what will happen. It will keep it forever – or endeavour to do so – and it has a cracking record so far. If I give my files to a university computer centre, I know what they will do. They’ll lose them within three months. I still feel that we lack a proper set of tests to acknowledge what are trusted repositories. Interestingly the Maoris developed five tests for their oral archives a thousands years ago which seem to me to offer the basis for trusted digital repositories:
i. Receive the information with accuracy
ii. Store the information with integrity beyond doubt
iii. Retrieve the information without amendment
iv. Apply appropriate judgement in the use of the information
v. Pass the information on appropriately
While I am personally clear where the responsibility must lie, this will be meaningless if it is a whole series of isolated activities. This requires a collective infrastructure of the sort we are already experimenting with. Broadly this is needed to ensure aggregation, to ensure discovery, to ensure access and to ensure preservation. We will need a preservation registry such as PEPRS (bearing in mind my very broad definition of serials), we will need common description standards, we will need access agreements and we will need LOCKSS type arrangements. We will also in my view need to support and fund a number of regional centres which can absorb the material of smaller institutions or those without trusted repositories. And no doubt agencies such as JISC and the BL will then have to operate and negotiate at the international level since this is much more than a national problem.
I have been above all a serials librarian throughout my career. Before I retire I would love to be able to identify the silver fork journal on which my first article was based then go to a digital repository and access both a copy of that and of my own first article. It is hugely encouraging that all of the right players are here in this room. What they now need to do is to commit to creating, funding and delivering the sort of digital infrastructure which libraries delivered in the age of print.