During the last decade, e-books have grown to become a significant library resource; a 2007 international e-book survey showed that 88 per cent of respondents ‘answered that they own or subscribe to e-books’ and nearly half of the respondents (45 per cent) have had access ‘to more than 10,000 e-books’.1 Although annual US wholesale e-book sales rose by 23.6 per cent in 2007,2 take-up has remained uneven in the UK. Higher and further education continue to dominate sales, with a number of public and special libraries also committing themselves to the format. Other sectors − such as schools − are only just becoming aware of the possibilities.
In the same way that some years ago aggregators developed platforms of e-journals, now they have become an important means of access to e-books, providing large collections of titles from many publishers under a single contract and through a single interface. They constitute a major source to which UK libraries turn for the provision of e-books, as reported by Jisc: ‘Until 2005 roughly 50 per cent [of HE institutions] had been buying e-books from aggregators, but this figure has risen to 63 per cent in the last 18 months.’3
We conducted a short study focusing on major aggregators (see list, below).
Credo Reference (re-badged xreferplus) http://corp.credoreference.com/
Dawsonera www.dawsonera.com/
Ebook Library (EBL) www.eblib.com/
ebrary www.ebrary.com/corp/
Gale Virtual Reference Library (from Gale/Cengage Learning) http://gale.cengage.com/
Knovel www.knovel.com/knovel2/default.jsp
KnowUK (from ProQuest) www.knowuk.co.uk/
MyiLibrary (from Ingram Digital) www.myilibrary.com/company/home.htm
NetLibrary (from OCLC) www.netlibrary.com/
Questia www.questia.com/
Safari Books Online (available via ProQuest) http://proquestcombo.safaribooksonline.com/home
VitalSource (from Ingram Digital) www.vitalsource.com/
All offer information on their websites, and two – Knovel and Questia – allow searching from non-subscribers. It should be noted that what follows deals with the main aggregations for each company. Some companies, for example Ingram Digital which offers VitalSource for textbooks, may offer additional products. It is envisaged that the new Gale Directory Library will extend to include titles from other publishers and will thus become an aggregator product.
It was necessary to set some parameters for the study, and there are two areas which deserve a lengthier treatment than this article can provide − licensing and a comparison of software features.
Possibly one of the most difficult areas for librarians faced with collecting e-books is that of licensing and charging models. As the Jisc report stated, ‘It is clear from our discussions with library staff that there is a considerable amount of ignorance and misunderstanding about what e-book deals are on offer and what terms can be obtained from aggregators and publishers’.4
Licensing is a complex subject with probably as many approaches as there are aggregators or publishers. Some e-book aggregators offer a model in which the library owns books that have been purchased − ownership, in this case, meaning a fixed number of accesses each year. This model may incorporate a small overhead for the platform and in some cases also offer a rent-to-buy approach.
The other main model is the concurrent-use licence, with an annual subscription to a collection or part of a collection. Some models allow users to download e-books to a laptop for a limited loan period; others allow library users to view titles in the aggregator’s master collection to which the library has not yet subscribed. Finally, MyiLibrary has a pilot ‘inter-library loan’ system developed with the Canada Institute for Scientific and Technical Information (Cisti), which allows users to purchase one month’s access to an e-book by way of a credit card payment.5
In this article, we do not provide a comprehensive listing and direct comparison of platform features, as Jisc is beginning to provide this through its Academic Database Assessment Tool (see illustration), which aims to ‘help libraries to make informed decisions about future subscriptions to bibliographic and full text databases’.6 The most recently added area concerns e-book platforms, and five aggregators are compared there by means of more than 70 criteria under 10 headings, which include: Access Control, Search, Indexing, Search Results, Restrictions and Usage Statistics. Here, we have explored some of the more novel features that aggregators offer users.
What follows is an overview of e-book aggregators and their services which highlights a number of trends in e-book aggregation and explores issues facing librarians and publishers. We devised a set of questions covering five areas: content and coverage, features, pricing and licensing, working with publishers, and working with libraries. In seven cases, data was collected through face-to-face or telephone interviews while, in two others, aggregators chose to submit a written statement. The primary data was complemented by examinations of the aggregators’ websites; however, in the case of Questia and KnowUK, information was taken from their websites alone. The data was analysed using QSR NVivo7.
Aggregator characteristics
If aggregators can be typified, we would suggest that the provision of large collections of e-books from a range of publishers, all of which may be used on a single software interface, is the most important characteristic. Often, particularly but not only with reference collections, the full text of the entire collection can be searched in one go, and the book structure or model is visibly retained, often by an explicit and expandable table of contents in a separate pane on the screen. There will be a range of features such as the ability to search in-book, and to have password-specific book-marking, highlighting and note-taking; for the most part, books are intended to be read online, although a few aggregators may allow the e-book to be downloaded to a laptop. Because the majority of aggregators offer digitisations of printed books, there is little evidence of interactivity or multimedia elements in the collections. However, Credo Reference includes audio files (pronunciations, music and literature readings), animations and videos within selected encyclopedia. Aggregators will all offer free trials, Marc records for individual e-books, and at least Counter-compliant or Counter-equivalent user statistics (John Cox7 compares user statistics from different e-book vendors); most will allow federated search access to the collection. The majority work with consortia as well as with individual institutions.
Collection content
With respect to subject coverage, EBL claims an STM emphasis and has the strongest Australian collection. Unusually, GVRL has focused on the humanities, and on literature in particular, although Gale anticipates that science and health are destined to develop. Dawsonera, MyiLibrary, NetLibrary, Questia and ebrary are truly multi-disciplinary (although MyiLibrary initially had strengths in the medical and health sciences area). Only NetLibrary includes fiction and children’s fiction, while both it and GVRL include collections suited to the schools market. Reference collections tend to be smaller and the four aggregators included here vary from around 120 titles (KnowUK) to some 2,000 (Knovel); the general aggregators vary from around 48,000 (Dawsonera) to 170,000 (ebrary). Typically each aggregation includes titles from 60 to several hundreds of publishers. Safari, which specialises in online IT and business/management, has a smaller collection of around 6,000 titles from 26 publishers.
Four of our aggregations are clearly reference collections: Credo Reference, Knovel, KnowUK (a UK-related collection) and GVRL, with ranges of traditional reference sources including encyclopedia, dictionaries, atlases, yearbooks, and guides. Knovel focuses specifically on science and technology. The remaining aggregators do not focus on a particular form of e-book, although one – NetLibrary – has always offered a collection of free reference titles alongside its much larger main collection. The more general aggregators all include a selection of e-book types including monographs and, to a limited extent, textbooks, but their collections also invariably offer more than books.
Safari also offers ‘Rough Cuts’, which are ‘books before they are published, so [enabling] authors to make their books available online – to get feedback from their audience at an early enough stage that it could feed into the final work’ and ‘Short Cuts’ for authors who ‘have something to say, but it’s not enough to fill a book, 80 to 100 pages – somewhere between a super-article and a very short book … [it] probably won’t be printed – this is content specifically for online’.
The majority of aggregators will update collections as new editions become available. GVRL is moving towards dynamically updating the online content of certain of its titles, including The New Catholic Encyclopaedia and Grzimek’s Animal Life, moving away from the need to wait for a new edition to be published, and analogous to the process of continuous revision.
There has been a marked move over the last couple of years – possibly related to OCLC’s pronouncement8 that users are, or will become, ‘format-agnostic’ – for e-book aggregators to add non-book content to their collections, and to market them using the term ‘eContent’. As one interviewee expressed herself on adding journal material on an issue-by-issue basis, ‘I don’t think there’s any harm in doing that … if it makes sense for the patrons to find it there, or for the library to want to acquire it within our platform’.
It brings together rich content in one place in response to a search query, certainly, which users find helpful. More than 80 per cent of respondents to ebrary’s Global eBook Survey stated that integration of multiple online resources is ‘very important’.9 Questia offers journal, magazine, newspaper and encyclopedia articles; ebrary does not include newspapers, but adds sheet music, maps and reports (e.g. from Dun & Bradstreet and Datamonitor) to the mix. NetLibrary offers eAudioBooks, databases and e-journals; and Safari also has non-e-book documents and articles, while also adding some 600 hours of video instruction.
MyiLibrary is moving in the same direction and has noted an increase in demand for certain of these formats, while Dawsonera speaks on its website of including ‘new media’, and audio was mentioned in the interview. GVRL, KnowUK, Credo Reference and Knovel – the four reference aggregators – vary. Knovel has also moved beyond the e-book to include ‘databases and other e-content’, while Credo Reference and KnowUK focus only on reference books.
While the facility for researchers to have a range of resource types available in one place has evidently been welcomed by many libraries, it is necessary to sound a note of caution. Assessing bibliographic databases in the recent past, reviewers often commented on uneven coverage or on the lack of information on coverage – we have the same situation here. Other than in GVRL, which links to InfoTrac for journals, it is not clear how or why particular journal articles or issues have been selected
for inclusion among the e-books. One has to hope that library end-users will realise that they are being offered only a very small, and selective, window into a journal world.
Market trends
One of the principle objectives of our study was to explore recent trends in the markets aggregators are addressing for e-books. Historically, e-book publishers and aggregators focused on the academic library sector, initially within higher education and latterly, through the initiative of the Jisc e-Book Working Group, on post-16 education. Public libraries and special libraries too have, in recent years, benefited from a broadening of the range of e-book titles offered by aggregators. These sectors still constitute the principal market, as typified by ebrary which sees its users as ‘based in universities, further education colleges, corporations and public libraries’.
However, our study revealed that, in several instances, aggregators are now exploring the potential of the children’s and school library sectors. For several years, NetLibrary has expanded considerably its collection for North American school libraries so that it now constitutes a critical market. While this remains a focus in the US, there is a suggestion that ultimately the collection will develop to encompass the needs of schools internationally. Similarly, a significant proportion of GVRL’s collection is used by US schools, some of which will be internationally relevant.
Another aggregator dipping its toe into this pool is Credo Reference: ‘If a public library in the UK subscribes to Credo then we actually create a second site for them. They’ll have access to two versions of Credo – our standard version and a customised student one. It’s a subset of books that can be targeted at a particular subject area.’ Credo Reference is currently conducting book trials in a couple of libraries, and indicates that it is likely to focus on the secondary school student but, assuming that the market is viable, ‘there’s nothing to prevent us from targeting the more junior people’.
Over the past five years we have been conducting a series of workshops on e-resource provision for the school and children’s library sectors, and know that the time is ripe for a critical mass of e-book material to be made available. There remains, however, the problem which aggregators have identified of the fragmentary nature of this market, with their often small, devolved budgets. This is gradually being acknowledged by school libraries and school library services, and also more recently by Jisc and Becta.
Geographical coverage
Several aggregators, such as MyiLibrary, are UK oriented, and are now looking to extend their international content. However, one of the early criticisms levelled by UK libraries at some of the general aggregators which were established in North America concerned the predominantly US bias of their collections. Our study suggests that there is a discernible move towards offering a significant quantity of e-book titles from non-US publishers.
NetLibrary, for example, indicates that more than 60,000 titles from the collection of more than 160,000 emanate from non-US publishers; ebrary now offers more than 70,700 titles from more than 74 non-US publishers and aggregators – including 35 major UK publishers – from its growing repository of more than 170,000 titles. Dawsonsera and Safari indicated that they were now becoming more Eurocentric, and the latter was drawing in content from Indian publishers; EBL has ‘a good mix of publishers in all regions – Australasia, UK and Europe’, and suggests that there is no discernible bias: ‘I think we have [none], because we’ve tried to be a little bit stronger with UK-based publishers to date.’
The move towards internationalism is also reflected in the development of multi-lingual interfaces and multi-language content. NetLibrary, Credo Reference and MyiLibrary are among those offering or developing multilingual interfaces. The range of interface languages is broad, extending beyond European to some of the Asiatic languages. The ebrary platform and interface is currently available in English, Spanish, Portuguese, French and Swedish. This year, Credo Reference will be developing a new enhanced version of the system that will enable it to take multi-language works, and following market research into the needs of UK libraries it has identified Polish, Urdu and Chinese as priorities. The new ebrary Reader, currently in beta, supports any language type, which means that any e-book, in any language, can be viewed and used.
Few other aggregators offer non-English-language material, although some are addressing this dearth, and recognise the need to extend the volume of multi-language content. NetLibrary has been conscious of this issue for some years, and now offers approximately 7,500 non-English-language e-book titles in more than 20 other languages. Critically, there is a move towards serving the markets in China, Japan and India. Ebrary’s selection also includes more than 29,800 non-English titles. Credo Reference offers collections of bilingual dictionaries and, given the predicted multilingual interface, anticipates developing collections of monolingual non-English reference titles. While the higher education market predominates for GVRL, the emphasis remains firmly on English-language titles. However, it is conscious of the market potential, and the
collection already includes titles in Spanish.
Developments in platforms
Aggregator platforms offer some very sophisticated means of access to e-book content. Before discussing these it is interesting to note the comments of some students in the SuperBook survey of e-book usage in University College London libraries,10, 11 who consistently used the browser or PDF reader ‘find’ function rather than the much more elegant search feature provided within the software. When asked why, they replied that they were unaware of, or had not noticed, the search box, let alone the advanced search feature. Allen McKiel noted in his analysis that the ebrary survey ‘provides indications that e-book collections and the research tools they provide are not well understood by a sig-nificant percentage of faculty and students’.12
However, aggregators believe that they have a clear view of how their systems are used. Safari noted: ‘We find it is used much more for practical learning where someone has a very definite problem to solve or a research topic – they spend on average about 30 minutes, so it’s not for immersive reading’, while MyiLibrary accepted that ‘it’s all about discoverability – if you’ve got stuff… we can improve discoverability’.
In addition to the conventional searching facilities, some aggregators offer more innovative approaches. Safari – because of its IT specialism – has developed the search facility to specify computer and programming code, and has cleared copyright on this content as well. This answers the recent debate on LIS-E-Books on how back-of-book CD-Roms are dealt with by some e-book publishers. Knovel has a range of features including searching for data by range, or data that is buried in a table or database: ‘Knovel will return just the relevant rows… Knovel is smart about units. If you search for a value in lb/ft³, Knovel will find the relevant results even if they are published in g/cm³.’
Credo Reference has looked at different learning styles – visual learners or individuals who are ‘exploring’ topics – and have developed tools specifically to help them. The Concept Map explores visually the terms associated with an initial search term and provides a mind-map of subject nodes, which link to text in the various reference books: ‘I’ve seen people who are visual learners – their eyes just light up when they see that… you can choose a topic like global warming or perception or whatever, and just – it helps you get ideas about how to broaden that… you are looking at the inter-relationship and exploring a topic.’ Further developments are under way to significantly enhance the Concept Map to give users more control of output and searching.
As we noted above, for the most part aggregator collections comprise digitisations of printed titles, but our study revealed a small degree of interactivity developed by some aggregators. While all aggregators are constantly improving their software, the reference aggregators – possibly because searching across the whole collection is so important – seem to have advanced further. Credo Reference has added user-definable timelines, interactive maps and dynamic statistics tables. Similarly, ‘Knovel’s interactive tables can show or hide rows and columns, [and with] Knovel’s equation plotter, you can simply select the values of interest on the curve and export them to Excel for further manipulation, again with the source citation.’
Variations on book-marking, highlighting and note-taking are common to most platforms: ebrary notes: ‘The current ebrary Reader™ allows end-users to take notes alongside individual pages and store them on their personal bookshelves for later retrieval. [The] new Java-based Reader, currently in beta, offers improved annotating features, with any combination of:
- multiple highlights and notes per page
- resizable and movable notes
- highlights with or without notes attached
- color coding of notes and highlights.
Some aggregators offer referencing help by providing citations, sometimes in as many as half a dozen styles, and Knovel automatically exports a citation with any text or table. Ebrary also offers automatic citations when text is copied and pasted and printed. Ebrary’s citations, available in several different formats, also include a URL hyperlink back to the source. The interim results from SuperBook offer an important caveat, in that, even where there is a choice of styles, it does not necessarily match the needs of students and academics. MyiLibrary is fully integrated and Credo Reference and ebrary are working on integration with citation software such as RefWorks and EndNotes. Those that do not offer this kind of help recognise it as desirable and are likely to make it available soon.
For the most part, e-book aggregations are marketed as independent collections. However, a different approach is offered by Credo Reference, the implication being that the collection is not necessarily totally self-contained but can and should be linked to the wider resources of a library, be they print or electronic: ‘We are going to make it possible to continue searching through other resources.
Our search results and entry pages will provide a means for users to continue their search at an Opac or additional library resources. These options will be library-defined and offer a gateway to further research. We really see reference works, particularly subject encyclopedias, as a foundation to your research, a gateway to the library.’ Similarly, ebrary allows users to link out to additional electronic resources such as bibliographical databases, dictionaries, translations and other online resources in the library and on the web through ebrary InfoTools™ software. The InfoTools menu can be customised by the library to link to the online resources of their choice.
Other aggregators mention that their platforms do not prevent the possibility of facilitating hyperlinking to external sources if this has been undertaken by the publisher, and that they are willing to explore this further: ‘And I do have some ideas – you know I have these visions of ways that we might be able to inject the citation information with some ways to the OpenURL or whatever.’ (EBL)
Bibliographic issues
A perennial problem that has been articulated for the past decade in research into e-books 13-15 is the fraught issue of bibliographic control of the format. Given the lack of a single national bibliographic source for the foreseeable future, the identification of new titles can be a difficult and time-consuming activity. One important development has been the integration of e-book titles within several international bibliographic sources, such as Global Books in Print and WorldCat.
Throughout the study some aggregators recognised the significance of this issue, and acknowledge the need to be more proactive in expl-oring the implications of bibliographic control. Several aggregators do put their titles in these bibliographies. For example, NetLibrary and Knovel have their titles within OCLC WorldCat, and GVRL within Global Books in Print. Although they do not work through the major bibliographies, Safari notes that, for its business library, all the titles are indexed in ABI-Inform, which offers another form of discoverability. Ingram Digital has worked with Microsoft Live Search Books to make its e-books content discoverable through Live Search Books with a tie back to MyiLibrary.
Several aggregators commented that, since they are integrated within a library supply system, and since libraries have traditionally used suppliers as the major source of selection and acquisition, they had felt no there was need to make e-book titles available in the international bibliographies. However, as a consequence of this study, one aggregator was interested in exploring the implications of having e-book titles listed in the international bibliographies.
Working with libraries
We were keen to explore the perceptions that aggregators held about working with libraries. There is an evident desire to offer an attractive deal for libraries, while maintaining a balance in negotiations with publishers. This is epitomised by the following approach: ‘You used to buy 500 of our 1000 titles for this much money – why don’t you buy 800 for the same money? With the power of “e-” we can deliver that at no more cost to the publisher, and they can pay the authors what they always paid them. So everyone gets the same amount of money they used to get, but the institution gets more content.’ (MyiLibrary)
One issue of particular concern related to the promotion and marketing of e-books within libraries. Over several years we have been running workshops about the management and promotion of e-book collections, and anecdotally it is evident that librarians from all sectors face the challenge of creating an awareness of new e-collections and developing skills strategies to exploit them.
In the limited research that has been conducted into this sphere 16, 17 there is evidence to suggest that students and academic staff in university libraries appear ignorant of newly established e-book collections, and that it is incumbent on librarians to do more to facilitate usage. Some aggregators share this view: ‘There is an issue with publicising the stuff and having a collection that is attractive ... I’m sometimes frustrated – a library will buy 12 e-books and then after a year they’ll go “Well nobody used them”, and I’m thinking, well no, because they weren’t made available. I think libraries have to go through an educational process.’ (MyiLibrary)
While libraries are seeking new ways of promoting e-book collections – with varying degrees of success – several aggregators thought libraries might make greater use of their promotional products. NetLibrary, for example, was concerned that there had been low take-up of its marketing materials.18 Credo Reference has ‘Promote It to Your Library’ promotional materials and case studies of ways of using the reference collection, and GVRL has posters. Meanwhile, ebrary offers promotional materials as part of its ‘getting started kit’ for new customers, as well as videos explaining product features, quick-start guides, and complete user guides; and recently launched a live web-based training programme.
Impact of new technologies
All the aggregators to whom we spoke are very conscious of new developments, particularly the increase in interest over e-book readers and mobile technology. EBL noted that titles were ‘already available on handheld readers – the books can be transferred to a PDA, and we are working with some device companies’. Credo Reference said that ‘part of our development for our next version will enable mobile technology, but actually Credo today looks pretty good on the Blackberry – you know it does work’. Safari acknowledged this would be delivered soon: ‘We are looking at two things – one is making Safari available on hand-held browsers – could be phones, could be a variety of things’. Likewise, NetLibrary says it is exploring the possibilities, and Dawsonera offers some PDA compatibility (‘Yes. Some will work with the software, some won’t’) but was less convinced by e-book readers, which were not seen as ‘very important yet’.
Conclusions
Overall, aggregators represent a major, perhaps the major, source for e-books in libraries, bringing together vast collections of general academic titles and quite large reference collections for easy access. While English collections dominate, there is a discernable move towards developing multilingual access and content. Markets are focused in academia and public libraries, but schools and special libraries are increasingly represented, and all aggregators acknowledge their needs.
Aggregators provide access to, and the ability to interrogate, the digitised texts in innovative and helpful ways using an array of software features. While charging models are complex, aggregators believe that they represent a fair deal to both the publishers and the libraries, both of whom have to be accommodated in negotiations.
During the study we were listening to the voice of the aggregator, to their perceptions of how they might enhance the library culture of e-book collection and use. While many offered summaries of strengths that were unique to their collections and interfaces, several offered insights into the aggregators’ role as trusted intermediaries.
These perceptions are reflected in the closing extract:
‘A lot of our work is about bringing together the publisher and the library… [with] thousands of publishers and millions of consumers (libraries, or whatever it might be). Someone needs to pull this together… and that requires a bunch of stuff – technology and transfer mechanisms and delivery mechanisms and all that stuff. But it also requires relationships with publishers and trusted relationships… [with] libraries and library consortia.’ (MyiLibrary)
References
1 Ebrary’s Global eBook Survey. 2007 (www.ebrary.com/corp/collateral/en/Survey/
ebrary_eBook_survey_2007.pdf ).
2 David Rothman. ‘U.S. 2007 wholesale e-book sales: $31.7M, or 23.6 percent over 2006 – but should they have been still higher?’ TeleRead, 18 February 2008 (www.teleread.org/blog/2008/02/18/
us-2007-wholesale-e-book-sales-317m-or-236-percent-from-2006-but-should-they-have-been-still-higher/).
3 A Feasibility Study on the Acquisition of e-Books by HE Libraries and the Role of Jisc. The Higher Education Consultancy Group, 2006 (www.jisccollections.ac.uk/workinggroups/ebooks/
studies_reports.aspx).
4 Ibid.
5 http://cat.cisti-icist.nrc-cnrc.gc.ca/screens/
myilibrary_e.html
6 www.jisc-adat.com/adat/home.pl
7 John Cox. ‘Making sense of e-book usage data.’ The Acquisitions Librarian, 19 (3/4), 2008 (www.library.nuigalway.ie/coxebooks.pdf) (Archived by WebCite(r) at www.webcitation.org/5WRHRWnKO).
8 2004 Information Format Trends: content, not containers. OCLC, 2004 (www.oclc.org/reports/2004format.htm).
9 See 1.
10 Chris Armstrong, Ray Lonsdale, and Dave Nicholas. ‘SuperBook: planning for the e-book revolution.’ Library & Information Update, 5(11), 2006, pp. 28-30.
11 C. J. Armstrong and R. E. Lonsdale. Initial analysis of SuperBook qualitative survey [unpublished], 2007-08.
12 See 1.
13 C. J. Armstrong and R. E. Lonsdale. The Publishing of Electronic Scholarly Monographs and Textbooks. Ukoln, May 1998; and Report G5, Library Information Technology Centre, April 1998 (www.ukoln.ac.uk/models/studies/).
14 Chris Armstrong, Louise Edwards and Ray Lonsdale.
‘Virtually There: e-books in UK academic libraries. Program 36(4), 2002, pp. 216-227.
15 See 3.
16 Promoting the Uptake of E-Books in Higher and Further Education. Gold Leaf, 2003
(www.jisc-collections.ac.uk/workinggroups/ebooks/
studies_reports.aspx ).
17 See 10.
18 e.g. its marketing kit is available at: http://library.netlibrary.com/marketingkits.aspx
Ray Lonsdale (rel@aber.ac.uk) is Reader in the Department of Information Studies at Aberystwyth University and Chris Armstrong (lisqual@cix.co.uk) is Managing Director of Information Automation Ltd.