Library Out Loud

What is eResearch?
Definitions of eResearch abound and it can be pretty difficult to see any strong distinction between eResearch, eScience and digital scholarship. They are all about applying modern computer and communications technologies to the business of creating and disseminating knowledge that universities and other research institutions have been involved in all along. Looked at in this way very little has changed over the past thirty years – researchers are still reading books and journal articles, working in labs, on farms, in archives, in schools and out in the community to make new discoveries, and communicating this new knowledge at conferences or through peer-reviewed publications.

Nothing’s changed? Really? Really???
If we look at these activities from the perspective of how things are done rather than what is done then a very different picture emerges – there is a computer on every desk, journal articles are nearly all accessed electronically, and this is increasingly the case with books as well. Basic facts are simply plucked off the Internet which is also how we access the media, while observational data are recorded digitally and may even come directly from the farm or the bush to the desktop. Email and other digital communications have hugely supplemented the face-to-face contact of conferences and a range of new and exciting/alarming publishing opportunities has emerged. What is more, research has itself become the subject of digital enquiry, with research activities being stored, recorded, counted, weighed and evaluated by a variety of digital instruments.

Well, we haven’t really answered the question about eResearch yet, but what about libraries?
One of the more obvious markers of these changes is that researchers no longer need to come to the library, or at least not nearly as often as they used to – in the words of the cliché the library comes to them. Libraries have played a major role in the growth of digital scholarship from the time that online catalogues began to appear in the 1980s, followed by databases in the 90s and shortly afterwards by electronic journals, dictionaries, encyclopedias and books. Not only did the means of delivery change, the volume and type of information delivered changed as well. “Big package” deals with academic publishers hugely increased the percentage of scholarly titles available, while Open Access publishing saw a range of new publishing outlets and a shift in publishing costs away from libraries and towards authors themselves. Libraries also found themselves facing competition – where in in the past a student might have had to track down a line from Shakespeare in a dictionary of quotations, now they just googled it. Easy as. Reliance on Google happened so quickly that there was soon criticism that it was insufficiently scholarly, leading to the appearance of Google Scholar. Also easy as and, stretching the definition of the term only slightly, scholarly.

So it’s all over for libraries, right?
What has kept librarians in business so far, and for the foreseeable future, has been complexity and the growth in expectations. While the new systems work very well, they don’t work perfectly and they don’t work without a good deal of invisible effort. Interoperability is a great idea and the experience of being able to glide effortlessly from the research question to the search to the document remains the ideal, but it doesn’t always happen and even when it does this is usually because libraries (and publishers and database creators) employ teams of staff (indexers, metadata specialists, cataloguers, IT staff and electronic information librarians) to make sure that it happens. If information sometimes appears to flow like water out of a tap, this is because of a complex system of pipes and filters that much of the time we are not aware of. Even when we use Google Scholar we depend on this infrastructure. And of course there is an important step in the process which involves the translation of the research question itself into a productive search. Modern information systems aim to be intuitive but the appearance of simplicity can be seductive and can disguise the fact that the information that has been found is only a small part of what should have been found, or that much of it is dross. As the amount of information available grows so does the task of keeping up to date (while the time available remains the same at best) so understanding how knowledge systems work and helping researchers get the most out of them continues to be a major responsibility for librarians.

So that’s a lot of people finding a lot of digital stuff. What do they do with it all?
Electronic document management has become a major headache – as it becomes easier to download PDFs of articles, reports and books we tend to keep a lot of stuff just in case we need it, or save them it the best intentions of reading them at some later stage. Much of this material has non-descriptive filenames and ends up clogging up our computers and making it harder and harder to find the really useful article we saved a couple of months ago, so we end up going back to the database or the web and downloading it again – if we can find it. Tools like EndNote that began life as reference and citation managers have extended their reach to include file storage, while a range of new web-based and social facilities have come into being with innovative and challenging new ways to handle this problem. Mendeley, Academia and Papers are well-known examples.

And do librarians play a role in all of this?
By default we tend to be good at dealing with published documents. That’s where libraries came from after all and library staff tend to see a very big picture – we buy and organise books and journals and help students and researchers use these collections to find information. Then increasingly we find ourselves helping with the process of knowledge creation and dissemination itself, making sure that the literature review is properly carried out, assisting with referencing, advising on choice of journal or publisher, advising on titles and keywords to optimise discovery of articles and even tracking the impact of work after publication. Library staff at all New Zealand universities played a significant role in PBRF over the past couple of years and that sort of work is not going to end any time soon. As long as universities and other research institutions are competing for money and prestige then they will be seeking to maximise the impact of their activities; published outputs are a major part of what a university produces so we can expect that the need to measure and evaluate them will continue. Thoughtful and nuanced promotion of research publications is essential if they are not to sink without trace and there is no one single factor that will ensure this. Simply choosing a so-called ‘A’ or high impact factor journal will probably not be enough on its own.

What other sorts of things does the library do in the eResearch area?
The Library makes a huge contribution to the dissemination of Massey’s research through our digital repository, Massey Research Online, – since 2007 all Massey theses have gone into MRO as a condition of enrolment and a retrospective scanning programme has seen the inclusion of all Massey doctoral theses in Massey Research Online. This makes our doctoral research available to the entire Internet through Google Scholar and our graduates have a permanent link to their work. The Open Access principle applies to published works as well, and by depositing their articles in Massey Research Online researchers can ensure that they are available to their maximum potential audience. Open Access scholarly publishing is a major international trend and is mandatory for publicly-supported research in many countries but, as in other domains, Internet openness in scholarship is not without risks and costs. The Library is increasingly a source of advice in this area, particularly with the flood of low quality or fraudulent journal titles soliciting articles and authors’ fees.

And what about the future?
A couple of areas we are likely to move on in the next few years are stable researcher identifiers and open data access. Attempts to resolve confusion over author names has been a preoccupation of librarians for a long time, and with researchers regularly changing their names – or at least the form of their name under which they publish – and moving between institutions, it can be pretty difficult to work out who published what. At the same time, the research assessment environment makes it important to do this accurately. There are various researcher-identity schemes out there and at present ORCID looks like the best bet.

Attitudes towards research data are also changing rapidly. In the past research was carried out and data collected but only results and conclusions were published, but the Human Genome Project and similar international collaborations have shown the benefits of making the entire research data available. While these are early days, we can expect more and more published articles to come with links to the original data itself. The aim is to raise standards of data description – not only what the numbers are, but how they were collected and why – and to improve the standard of the research itself by subjecting it to greater public and peer scrutiny. Increased efficiency is another desired outcome from this development, which holds a particular appeal for funders – rather than the same or similar data being gathered over and over, carefully managed and described datasets will be available for reuse, or at least for comparison. For many researchers this will be a step too far – access to or ownership of data is of huge academic advantage in a market in which they are competing for research funds, positions and promotion – but over time greater openness about data promotes a clearer separation between observational or experimental research and subsequent analysis, and a possible decline in the incidence of data massaging and academic fraud.

But what has that got to do with libraries?
Librarians have a long tradition of storing and describing information and then locating it, and tend to understand the relationship between these things. For data to be useful it has to be securely and permanently located and then properly described through metadata that makes it accessible to Internet searches. If the open data movement is to bear fruit it must have a sound infrastructure and follow well-accepted standards. Just as libraries have retained a central role in the transition from print journals to a largely electronic environment, they are equally well-placed to curate future digital data collections.

But we still haven’t defined eResearch in a simple sentence!
That’s true. Probably eResearch relates to anything that fosters and promotes digital scholarship and we’ve largely concentrated on the library’s part in this. It might be useful to mention a few things that libraries don’t do. A lot of what is known as eResearch relates to high-performance computing and data communications and this is properly the domain of ICT specialists and researchers themselves. Just as librarians don’t, as a rule, interfere in the process of creating books, so we don’t generally involve ourselves directly in the research process itself, beyond our role as locators and organisers of information; by and large the task of capturing and analysing data happens in the academic domain supported by ICT. However digital technologies have tended to create greater complexity and to break down these sorts of clear boundaries, so we can probably expect to see a greater intermingling of scholars, ICT staff and librarians as the trend towards data curation takes hold. And yes, we still don’t have a simple, single-sentence definition of eResearch.

Bruce White
eResearch Librarian
eResearch on Library Out Loud