ACCESS | Asia 's Newspaper on Electronic Information Product & Service

September 2005 No.54

The future of the book: the view from a librarian and a publisher

Although the transformation of books into electronic products has been slower than that for journals, each week sees the announcement of new titles, collections and services. The future of the book recently became a hot topic on liblicense, an online forum to debate licensing and digital products, managed by Yale University. With permission from the authors, ACCESS is reprinting below with a light edit, two contributions to the debate. The first is from Chuck Hamaker, Associate University Librarian, Collections and Technical Services, Atkins Library, University of North Carolina Charlotte, USA. Chuck was replying to Bob Bolick, Vice President, Global Business Planning, McGraw-Hill Education. The second is from Toby Green, Head of Dissemination and Marketing, OECD Publishing, in reply to Chuck Hamaker.

e-books need the e-journal treatment

Bob, I'm probably as worried as a librarian can be about the future of the monograph (no matter what format). Book publishers have been extremely slow in my opinion, to innovate. I want books especially scholarly books, to survive. What they do can't be done in journal articles i.e. the multiperspective, the careful development of complex concepts, the pulling together of an extended informed presentation on the author's topic. This form of entering into public debate and consciousness is not duplicated by the article.

At the same time, as a librarian, I see journal articles swamping whatever awareness individual students and scholars have of the means to engage in reasoned, articulate, civilized discussion. A large part of this has to do with how journal articles are identified vs. book contents. We have massive database driven indexes, we have the open URL, and we have special search tools within databases and controlled vocabularies for special topic fields: all these tools make the journal article more findable, more accessible and more usable.

How do books take their rightful place in the electronic environment as significant scholarly sources? So far the major models we have don't result in very high usage of the content, in my experience. The different silos of e-books aren't very helpful. I realize there are advocates for e-book usage results, but in my experience, they just don't get the kind of usage that justifies the kinds of investments being made in e-journals.

So, what's the solution? I would suggest we know of two basic things that can enhance the utility of books. One is purely technical and absolutely legal. The other, where I think Google is coming from, requires re-engineering the landscape of monographic literature through a massive indexing project.

Table of contents linked to chapters

On the purely technical level, I want to see tables of contents, which are often embedded in records in library online catalogs, linking directly to individual chapters and every other place a TOC appears (like Amazon for instance). We have tools that index at the chapter level right now: the library catalog and other TOC services. In many ways that is analogous to title level identification for journal articles. But we don't link at the chapter level to those electronic chapters, or instantly identify the item is print only, though the Worldcat Google indexing could be an automatic link from TOC indexing.

It's similar to where e-journals were when publishers would only support site level linking or journal title level linking. Deep linking is what has made journal articles instantly accessible. We need on the purely technical level, full scale linking to usable text in books, at least at the chapter level.

On the environmental level, we need much better comprehensive indexing. Word by word indexing, as seen in the many indexes in books, comes to mind as one approach. This is where Google Print and other Google approaches come in. They have the technical expertise to provide indexing that publishers could be supporting. It's at no cost to them [the publisher] and it highlights the content in their intellectual property.

Is what Google Print is doing, so different (especially since the text they show is so un-usable for cutting, pasting, even citing!) from the hundreds of indexes that indexing databases provide? A structure, I would remind everyone, that journal publishers didn't have to do much to cause to come into existence. I'd argue that Google Print's scanning and use of those scans, is analogous to the journal article indexing already standard practice in the serials literature.

And it can't happen soon enough if books, even e-books, are going to survive in terms of usage in the millennial generation's idea of where to go for information. I don't think we have 20 years, or even 10 to debate the future of the book. It's happening right now, and it can be lost I would guess in the next 5 years or less.

Monographs need massive indexing to survive

So, massive indexing of monographs which is what I see Google Print actually doing, is critical for the survival - the survival as usable text - of the book, to keep it from becoming nothing more than an interesting artefact of civilization.

If publishers don't want a future for the book then by all means pull out the stops and go after Google, or any other company audacious enough to provide indexing. On the other hand, look at usage data on e-books. No matter what platform, I think you will see the book is not right now in the same playing field or even the same universe as the journal article. The book's utility is being drowned out by snippets of information via articles. So I propose these two approaches:

First: better accessibility from existing tools like catalogs, Amazon and other indexes a la the Open URL structure or fixed URL structure. This is a purely technical approach linking at the chapter level everywhere a chapter is mentioned in sales and finding tools, including indexing and abstracting sources that commonly include books.

Second: massive full text indexing, with enough 'context' to let individuals know if they need to go 'get' the book wherever it is (in lieu perhaps of abstracts?)

Link from Google to full text

The constructive approach I would suggest, is to work with Google to link from the Google 'print' - actually dumb-print-version - to the full text version at publisher or vendor controlled, metered or subscribed site. Is that happening? If it is I haven't heard about it.

The whole library and vendor and publisher community has had to be engaged to create Open URL linking. Why not see if Google can support some sort of standard for linking from non-functional books to functional e-book linking? They are building the master index (that is what they do) with their scanning so why not use it to link to the 'official' copy or copies that individuals can actually 'use'?

So again, I see scanning as something other than 'scanning'. In this instance it looks to me like indexing. Scanning the whole item to provide a 'free' index for publishers and book readers and users is in the best interest of everyone concerned. Journal publishers don't opt out of journal indexes, because they know it enhances their journal sales. Why wouldn't the same thing be true of whole book indexing?

Depressed in Boston: disbelieving book publishers and new publishing models

By Toby Green

Chuck, I've just returned from the Society of Scholarly Publishing meeting in Boston exceedingly depressed and disappointed by the book publishing community represented there. I had hoped to learn some new tricks to help us improve our e-books offer - but instead found myself preaching and teaching disbelieving book publishers about what we've been doing since 2000.

Much of what we are doing is along the lines of what you advocate, in particular a move to chapter-based publishing. At OECD we publish around 150 books annually. In January 2001 we launched an e-library containing all our books published since 1998. To begin with they were only available as a single PDF file.

In 2003 we started to break some of the books into chapters, as well as offer the complete file. Our aim was to enable the various discovery tools find an individual chapter - just as they find individual articles in journals. In fact, we are using an e-journal platform to publish our e-books because we couldn't find an affordable and flexible e-book platform to do the job. And why did we take this step? Because we didn't want our books to be drowned out by e-journal articles!

The results are impressive. In 2000 we had around a core set of 250 libraries worldwide buying >60% of our books. Today we have 550 libraries worldwide with online access to 100% our books - an impressive growth in reach and accessibility. Around 200 of these libraries still choose to buy a print collection.

Slow start for downloads

Downloads of our books started slowly, I think because users didn't expect to find books online so weren't looking for them. We also didn't have a persistent linking system in place making it hard for librarians to link from their OPACs. In 2003 this changed when we launched our(persistent) EasyLinks and downloads took off.

Last year for the first time, we delivered more e-books (or e-chapters) than we did printed books. Usage levels are on a par with our e-journals, so you are quite correct: if book content is delivered in an e-journal-like way usage does increase.

This positive message is one I'm trying to get across to specialist book publishers - but they seem to find it difficult to accept that an e-journal model works very well indeed for scholarly monographs. Our business model is just like e-journals: i.e. multiple-simultaneous access; no DRM (we deliver standard PDF files); IP access; remote access; walk-in usage; licence to include excerpts in course packs; e-ILLm - in short, a licence to use our content as much and as frequently as possible!

(For any book publisher reading this, you may be interested to know that our print sales between 2001 - 2004 declined at a similar rate to the previous three-year period but, following a change to the way we promote our books (and having had some better titles!) they have increased a little since Nov 2004! Clearly, delivering standard, unprotected PDF files is not destroying demand for print.)

We've learned that e-books can't be handled exactly like e-journals, but the discovery principles are the same. We are trying to improve our systems so that the various discovery systems (Google, Google Scholar, Scopus, etc) will find our content easily, and the introduction of an OpenURL system due in 2006 will be another step to making access easier still.

Books or journals: readers don't care

I agree with you that efforts need to be made to improve indexing systems - but importantly I think they need to be compatible with e-journal indexing systems because in our experience readers couldn't care if the content they need comes from a book or a journal.

Books are not easy to break into chapters. We've tried to find ways to automate our production process so we can break them into chapters easily. The problem is rooted in a book's structure - lots have an hierarchical organisation (Part/Section/Chapter) so at what level(s) should they be broken? So far we've failed to find an automatic process and continue to break them manually at the end of the production process.

We've had to persuade our authors to provide abstracts for each chapter (or part or section), something we thought would be difficult, but which actually came readily. We've had to work on the chapter titles so they make sense when seen out of context in search engine results. Some short books don't lend themselves to be broken up, so we leave these complete.

Despite doing all this, and providing our EasyLinks, we're discovering that many librarians are not making the links to their OPACs. We're planning to introduce downloadable MARC 21 records in 2006 to make this task easier for librarians, but it doesn't seem to be a standard reflex for some cataloguers to make the link - I'm sure this will change over time.

Librarians want to pick and choose titles

Another oddity we've learned is this. Ask a librarian what they want and they'll say they want the ability to pick and choose book-by-book. This presents a huge challenge for publishers since it really means that some intermediary will have to offer this service (just as in the print world). This takes control of the way the books are presented and marketed out of the hands of publishers (and because we couldn't find an intermediary who could provide the service we wanted, we did our own thing).

Yet, as you will have understood from our sales figures, a great many librarians have acted differently and subscribed to all our books. The bulk subscription model sounds a lot like a books 'big deal', but the benefits seem attractive to a lot of librarians. We are now moving to offer our books via the e-books vendors to meet the needs of librarians who don't want all our books - but I think there's room for both the big books deal as well.

Perhaps librarians need to start thinking more carefully about this model if they really want to work with publishers - without it, you'll have to work with intermediaries.

The intellectual property question is one that puzzles me. I understand that problems may exist when Google scans books from libraries since the copyright owner is not involved. But, if it leads to greater chances of discovery I think publishers will learn to appreciate the benefits. The trouble is that it will take time for the benefits to become apparent and in the meantime everything may get bogged down with lawyers.

Does greater discovery lead to greater usage?

As for Google Print, where publishers and Google are working together, here we might be able to see if greater discovery leads to greater usage (and sales). My feeling is that for most higher-priced monographs greater sales isn't really what will happen. Why? Because most Google users are in a hurry and want access now - if they can't get it they'll move on.

Also, and more significantly, most readers of scholarly monographs have little or no purchasing power - so connecting from Google Print to online bookshops is not the ideal situation. The link, as you suggest, should also be to library systems via Open URLs (just like Google Scholar links to e-journals). But that's of little use unless librarians are making the connections to e-books and if e-books remain poorly presented by the aggregators.

Our next steps will be to enable reference linking from our books (via CrossRef et al - which will also mean other e-publishers will be able to link into our e-books when they are cited!); to further extend our StatLink service (whereby we offer links to Excel files that underlie the charts and graphs in our books; to improve our indexing and search service; to continue to push our metadata out to other channels to improve discoverability; to launch an RSS service.

I just wish some more book publishers would join us in doing all this, it's fun and it works!

Toby Green is the Head of Dissemination and Marketing, OECD Publishing, Paris France. Visit here & here for OECD's award-winning e-library & here for the new title alerting service.