Global discovery trends and the library’s changing role
The continuing rise of search engines as a powerful discovery gateway have left many academic libraries concerned about losing grip on ‘discovery’. Timon Oefelein maps shows why quite the opposite is true.
NOTE: Paragraphs three and four, referring to the Ithaka survey, have been updated since this article was published.
‘Discovery’ remains a hot topic. All over Europe librarians are eagerly discussing what implications the latest discovery trends might have on their library. And quite rightly so because it’s one of their core historic functions, that is, to help their patrons find and access the right content. But some argue that the Googles of this world have taken over.
They argue that students and researchers use the library less and less as a starting point for information gathering. And perhaps there is some evidence to back it up.
The prestigious and long-running Ithaka S+R US Faculty Survey 2015, issued April 2016, surveyed more than 9,000 faculty members from US academic universities about their views and attitudes towards the library’s core functions – gateway, archive, buyer, teaching support, research support, and undergraduate support.
One key takeaway was that only 58 per cent of faculty members from the sciences and medicine viewed the library’s ‘gateway’ function as ‘highly important’ – the overall number for all subject areas including humanities and social sciences was higher at 70 per cent, but still much lower than several other core functions. For example, 78 per cent of academics viewed the ‘undergraduate support function’ as highly important. Further, the library’s role as ‘buyer’ scored even higher at around 83 per cent. Is it true then, are the Googles of this world starting to take over the ‘gateway’ function from libraries?
At first glance the global external referral statistics from the last few years that we see at Springer Link back up the report’s basic findings. Around half of the visits to the SpringerLink platform came from search engines, in the order of Google, GoogleScholar, Bing, Yahoo. Another one fifth was from abstracting services, repositories, as well as library sources, including OPACS and discovery services. (Source: Springer Nature Strategy and Market Intelligence)
Further, one fifth of traffic was not directly identifiable, also known as ‘dark traffic’. We suspect that most of this group comprises visits from: web links in email alerting services, bookmarks and standard emails, as well as web search traffic from secure https mode and direct visits to our website.
Interestingly, just under one per cent of all visits come from social media, with Facebook and Twitter leading the pack, followed by Wikipedia, Reddit, LinkedIn, and ResearchGate. The traffic from social media has remained stable for the last four years with no significant change in volume.
The basic external referral picture remains similar at the institutional level. I have looked at the referral trends at over 50 medium-large academic universities throughout North Western Europe in the last four years and seen no major diversions from the global picture. The variations are insignificant.
It comes as no surprise then that some universities are making adjustments to their discovery set up. Utrecht University took the lead back in 2013 by taking offline one of the library’s main search boxes with little complaint from users.
Simone Kortekaas, from Utrecht University Library, presented her findings at UKSG in 2014 with her presentation ‘Thinking the unthinkable – doing away with the library catalogue’.
Her last slide nicely summed up her position: ‘Without your own discovery tool you might feel stark naked. However, we have to admit that others can do a better job on discovery. Focus on delivery and rethink the way you can provide value for your users.’
And yet, the delivery of content has arguably a significant effect on the true discovery of content. After all, nearly a quarter of all visits are from ‘dark traffic’ which is linked to delivery format and technology.
The Netherlands continues to lead the way with the UKB consortia coordinating the country’s transition to a unified information infrastructure, namely the OCLC World Share platform. Part of this project involves moving the holdings information of most Dutch Universities to the cloud-based World Share platform which will in turn optimise back-office processes.
The move will also allow the participating librarians to benefit from OCLC’s global holdings network. The question is, do these trends mark the beginning of the end of an era of home-grown library discovery technology and local OPAC infrastructure?
Certainly, the situation regarding the librarian’s role in the discovery process is far more complex than an initial reading will yield. Librarians still have substantial influence over the way researchers and students discover content. This influence is, however, less direct and less visible than before. But it deserves further consideration.
For me, true discovery goes beyond simply finding content: it and includes accessing it, obtaining it in the right format, and consuming a suitable and optimized platform. For a researcher to truly discover content he/she must travel seamlessly through all these stages. A bottleneck at any stage will hinder discovery, leaving the researcher frustrated. Like most of us on the net, we all want content quickly.
According to a Microsoft study which surveyed 2000 people, the attention span of an average internet user in 2013 was a mere eight seconds, down from 12 seconds in the year 2000. The reported noted that a goldfish had an attention span of 9 seconds! The report concludes that the participants ‘struggle to focus in environments where prolonged attention is needed.’(Source: Attention Spans, Consumer Insights, Microsoft Canada, 2015.
When you combine these facts with Tenopir’s findings that on average a researcher in 2011 looked at just over 300 scholarly articles per year (up from 150 in 1977) and spent some 31 minutes per article in 2014 (down from 48 minutes in 2005) then it becomes clear that researchers are under huge information pressure, with little time. In this light, seamless access to content is critical. (Source: Carol Tenopir, ‘Measuring (and increasing) the Value of Library Services’, Center for Information and Communication Studies, The University of Tennessee, 2014).
As we all know from experience, pay-walls and intermediate access steps will slow down, if not halt access. But the librarian is in a good position to help steer the path of access by choosing the best delivery format for their library infrastructure and user needs. While there are indeed many pressures operating on the librarian, from faculty requirements, budgetary limits and technological interoperability limits to university policy, to name but a few, the market does provide choices. Each of these choices often establishes a certain access path, a product format, a consumer platform and a certain useability experience.
Seen in this way, then librarian can influence the process of true discovery. On findability, a librarian might ask ‘Is the content indexed by Google and professional discovery services?’ or ‘Is the content marked up in semantic micro-data so that Google can machine read it?’ After all, true discovery is more likely if the content is easily findable on Google and seamlessly accessible via IP recognition.
Tuula Hamalainen, head of library at VTT Technical Research Centre of Finland, observes: ‘Discovery is indeed a complex phenomenon. On the one hand you have search engines which allow very large amounts of information to be searched, but they do not always yield highly accurate results.’
‘On the other side of the spectrum you have specialist bibliographic databases which give highly accurate results but their searching scope is limited. In this discovery landscape we believe that it is absolutely imperative that publishers do everything they can to make their content highly discoverable on all the available tools, this is of paramount importance to us.’
Further, there is the question of format. In many cases the researcher is perfectly happy with consuming the content in HTML format. But there are other cases where a researcher will want to interact with the content: click on an inline citation, annotate text and save it in a personal library. In the case of a larger ebook a researcher might like to browse through a low-resolution preview version before downloading the entire book.
One good example that demonstrates the importance of delivery format is the mobile app that Springer built for the British Library. I visited the library to brainstorm with them new innovative ways to help increase the access points to their licensed e-content. Prior to the app, access was restricted to a limited number of PC terminals in their reading rooms.
My thinking was, how can we get patrons to access the content, ‘How can we bring the content to them rather than patrons having to use the PC terminals in the reading room?’ I suggested to the BL the potential for a mobile app. They welcome this idea and a partnership project was born. The app was recently released on GooglePlay and ITunes and patrons now enjoy the benefits of access via their mobile devices whilst on the BL premises.
Commenting on the new app, Nigel Spencer, research and business development manager at the British Library, said: ‘The new mobile app allows us to really leverage the impact of our licensed e-content by delivering it directly to the mobile devices of our patrons. We expect this to make a positive impact on usage.’
All these formats exist and every time a librarian acquires an ebook – or other content – they do so in a specific delivery format and technology. If the ebook is a textbook, the student might even prefer a low-cost PPD print version to the e-version, for the summer break or home use. The DRM attached to the content is therefore of critical importance for true discovery.
If the user is freely allowed to share copies of articles and chapters with their fellow patrons at the same subscribing institutions then this will ultimately facilitate discovery. If there are no limits to the number of simultaneous users that can access the content at any given movement in time, again this will aid discovery.
How the user interacts with the content and follows links to further reading is also important for true discovery. ‘Does the publisher platform contain an article recommender?’ Similarly, ‘Can the user look up other articles by the same author via an ORCID ID?’ All of the above are examples of how the librarian can influence the way that content is found, accessed, and consumed.
Asked about discovery, Dr. Jord Hanus, Head of Research Affairs at the University of Antwerp, replied: ‘The quality of the content alone is not enough in ensuring ‘discovery’ by the scientific community… accessible and ‘findable’ metadata is in all likelihood much more important… being properly findable in Google truly is paramount in reaching a broad enough audience (as an author) or finding the most relevant papers (as a reader).’
The role of the library has always been an evolving one, and here we have seen that even when search engines are the main discovery tool, the library can affect the likelihood of their users finding the content that they need. Whilst the library’s influence on discovery is less direct than prior to the prevalence of Google search, their influence certainly exists.
With a focus on the opportunities outlined above, librarians can significantly influence the user’s experience by being aware of the needs of their users and matching those needs in their collection.
Timon Oefelein is Springer Nature’s account development manager for north-western Europe