Cites & Insights: Crawford at Large
ISSN 1534-0937
Libraries · Policy · Technology · Media

Selection from Cites & Insights 5, Number 4: March 2005

Ebooks, Etext and PoD

This may be the final Ebooks, Etext and PoD section—at least as a running feature. It hasn’t been “running” that often lately anyway. After six roundups in 2001 and another six in 2002, there were two in 2003 and two more last year, the latest one last July. With an eight-month gap, I don’t have much on hand.

There’s more material out there that I don’t feel compelled to note or comment on. The various library-market ebook and etext services send out press releases. The Open eBook Forum continues to trumpet record ebook sales and release best-seller lists, and once in a while a mainstream press outlet falls for the “fastest-growing sector of publishing” line, where neither growing numbers of sales nor dollars of sales matter, only percentage. (Sarah Glazer used that “fastest-growing” claim in a December 5, 2004 essay in the New York Times Sunday Book Review—and I find it interesting that the latest figures she quoted were still the Q1 2004 figures released by OeBF on June 4, 2004.) Of course, if there’s ever a year in which the quarter’s ebook sales do not set a new record, the still-tiny field is in truly serious trouble.

As of 2004, the world ebook market may have reached nearly one-tenth of one percent of the size of the U.S. print book market. It’s a start—but even the segments of print books that I believe should be ripe for ebook replacement (e.g., K12 and higher education textbooks, some reference works) represent a U.S. market at least 100 times as large.

Those sales figures also bother me because OeBF seems reluctant to define “ebook” carefully. Is a short story from Fictionwise counted as an ebook? I suspect so. Are PoD books counted as ebooks? I suspect not. My guess (and it’s only a guess) is that PoD books already represent a larger market, one that’s wholly integrated with other print book sales. The confusion shouldn’t be necessary. Ebooks may still be a tiny marketplace, but it’s a large enough marketplace for rigorous definitions. I don’t believe bookstores count magazine sales as book sales (magazines have a larger revenue stream than books, but it’s mostly ad revenue). I don’t believe OeBF should count anything as an ebook unless it’s long enough so that a print version would be called a book rather than a pamphlet—which usually means 48 pages, I believe, with certain exceptions for illustrated children’s books.

I’m thinking of dropping this section for several other reasons besides continued frustration with OeBF’s numbers and claims. The old disputes just keep going. Some folks really, truly want dedicated ebook readers—but not enough of them to make a legitimate market, at least not in the U.S. Most people interested in reading ebooks want to use etext on portable devices, a process made more difficult by the range of DRM issues. Already, despite relatively low resolution, a tablet computer makes a plausible full-page reading device for those who want to read that way—while at the other extreme, I shudder to hear some people say they’re happily reading ebooks on their cell phones. I wonder about contextual issues when you’re only seeing 100 to 200 words at a time (on a PDA) or even 300 to 400 words (on a notebook or tablet). But what about seeing 10 to 20 words at a time? Can you really absorb a book that way? Will writers and publishers start catering to people with truly short attention spans? (I’ve heard about one-minute “TV shows” to be viewed on low-rez cell phone screens; maybe we’ll have 500-word “ebooks” with 20-word paragraphs for the same market.)

I’m not making this up. That Times story includes this: “Fans of cellphone reading tell me they quickly forget about the size of the screen once they get absorbed in a good plot…” And here’s a tidbit from a manager: ebooks on a PC give you “the ability to appear like you’re working when in fact you are not,” and this guy has read “hundreds of science fiction and history books this way.” I’m sure his employer would be delighted to read that.

I printed a posting from, translated from Spanish into English on January 25, 2005. It’s titled “Why you should read e-books…and why you won’t.” The subhead is more direct: “Four reasons for electronic books to succeed, and four more reasons why they never will.” The story lists as benefits convenience, compactness, discreetness (all ebooks look the same…), and being “more complete” (that is, sometimes having added material). The drawbacks: They’re expensive, they’re almost all in English; there aren’t many of them; and they’re “too personal”—the DRM drawback. I would say that offering those as reasons ebooks “never will” succeed is almost as ridiculous as calling the death of print books “inevitable” (it’s ridiculous now, but it didn’t seem that way 12 years ago). That may be all the discussion the little piece deserves.

Substantial Articles

Litzer, Don, and Andy Barnett, “Local history in e-books and on the web: one library’s experience as example and model,” Reference & User Services Quarterly 43:3 (Spring 2004): 248-257.

I sort of miss the old RQ with its claimed meaningless title, but it’s been Reference & User Services Quarterly for seven years now; whatever the name, it’s the refereed scholarly journal of the Reference and User Services Association (RUSA) of ALA. This article is refereed—it was originally submitted March 3, 2003; accepted for publication October 30, 2003; and actually published a few months later. It also has the characteristics of a refereed scholarly article, for good and for bad, including a justification for what could be called a “how we did it good” study.

In a way, it’s too bad that this form of publication almost requires the 2.5 pages of justification, methodology, and assumptions: the “scholarly apparatus.” When you get to “The evolution of one library’s ‘local history on-line’” on page 250, you get to the heart of the matter: A description of an innovative low-budget project to make a library’s special local history holdings usable, which in this case means digitizing them—and a carefully-done study of actual use of that local history collection (at the McMillan Memorial Library in Wisconsin Rapids, Wisconsin). It’s a dynamite story, really about etext and the library as digital publisher more than it is about ebooks. Highly recommended both as a careful study and as an example and model for what many more libraries can, should, and will be doing now and in the future. Here are the first paragraph and last sentence of the closing half-page “Implications and conclusion” section:

If the use statistics provided in this study are close to representative, they indicate strongly that, as a public service, the use of digitized local history made available by a small to medium-sized public library on the Web is significant and worthy of the investment made in it. Whether a library digitizes in-house or out-sources; whether it digitizes its own materials or accepts donations of digitized documents, the demand exists, waiting to be satisfied, for digitized local history.

…Digitization of local materials is not merely a high-tech information transfer, but another way in which libraries can unify their communities by reminding them of the history and legacy they share.

Cox, John, “E-books: challenges and opportunities,” D-Lib Magazine 10:10 (October 2004), 17 pp.

Cox is deputy librarian at the National University of Ireland, Galway. His report describes “[t]he experience of a group of Irish university libraries” offering a Safari Tech Books Online collection. The experience “shows that, with the right combination of product and subjects, e-books can thrive among students and faculty, while librarians can create more dynamic, relevant and flexible collections than for print.”

I should note that, while I had already printed off Cox’s article for future comment, he’s also one of the alert Cites & Insights readers who noted that I used the wrong URL for the last issue in the Topica announcement—and, along with pointing that out, mentioned the article and attached a portion of the overall project report. That portion includes this paragraph:

By way of a benchmark from outside Ireland it was interesting to read a discussion of Safari use at York University in Canada in the August 2004 issue of Walt Crawford’s Cites & Insights newsletter [4:10]… York’s subscription comprised 150 titles, attracting 3157 sessions and 29,511 hits in the period from 1 September 2003 to 31 May 2004. The NUI Galway subscription of 64 titles (54 to mid-January) realised 1578 sessions and 11,307 hits as shown in the same period. This level of use was closely aligned to that at York, allowing for the lower number of titles in the Galway subscription. Crawford, who has tended to be somewhat sceptical regarding e-books, is impressed with York’s experience and with Safari, noting that 3000 hits from 150 books is “…success by any measure I’d care to use….If you’re looking for the ‘but,’ it’s not coming. Safari’s model makes sense for that kind of book and these massive, rapidly changing manuals make sense for most readers in e-book form.”

I would mildly object that I’ve never been skeptical of the potential for etext and ebooks in areas where they simply work better than print books—and needing to look at three to 15 pages of a massive software book or manual is precisely one of (several) such areas.

But now I’ve given away the conclusion and my opinion, haven’t I? This case study covers a group of Irish academic libraries that looked at the whole ebook marketplace, saw the range of possibilities and problems, and recommended a one-year trial “to focus on business and computing, two closely linked areas with strong teaching programmes at all seven universities.” Safari Tech Books Online was the unanimous choice for the trial. The article carefully describes the process, the Safari operation, and how it all went.

Safari is very much a pseudobook service, oriented to those wanting just a few pages. Each access only delivers the equivalent of three print pages (after searching, which can be within a title or across the collection). You can’t buy the titles, only a subscription for a term of access. At NUI Galway, there was a three-user limit—and that meant three users for the collection, not for a given book. On average, users spent less than five minutes in a session—and all of this makes perfect sense for the “I just need a few paragraphs” mode of etext/ebook use. I don’t take issue with Cox’s comment that, for software-related textbooks, “content may be viewed as disposable after a certain period.” Many “computer books” are certainly written as disposable products; it’s hard to argue for their lasting significance (sez I, who also doesn’t argue for C&I’s lasting significance).

Highly recommended. It’s a readable article and up to D-Lib’s high standards. Cox goes through usage and related survey results in some detail. Yes, many users commented on eyestrain, which was “the most predictable finding in the survey.” Yes, users came to see the collection as “a collective reference resource” more than a bunch of books; that’s as it should be in this case. Most of those surveyed—83.5%—showed “emphatic support” for the need to use printed books in addition to Safari.

From the conclusion: “Where content matches need, e-books can support the academic mission effectively, saving time and adding value as a collective online reference resource rather than a set of individual titles.” That should be true; this is another case study that demonstrates that it is true.

Garrod, Penny, and Jane Weller, “Ebooks in UK public libraries: where we are now and the way ahead,” UKOLN Issue Paper 2, July 2004.

This heavily footnoted 14-page paper offers a good overview of the ebook scene in the UK, with a balanced discussion of the advantages and disadvantages of ebooks. It’s interesting to see here, as in most recent articles, the shift from ‘90s discussions of when and how ebooks will drive out print book; this time, the authors cite Rod Bristow of Pearson Education UK arguing “that this is unlikely to happen, quoting the history of media in general as evidence of integration rather than outright replacement.” Garrod and Weller note the ongoing confusion between ebook as carrier and ebook as content, noting that in a 2002 survey on UK public library use of ebooks, “of the 13 respondents who claimed to provide ebooks, 12 were referring to CD-ROM. Only one…actually lent dedicated ebook devices…preloaded with a range of ebook titles.” The authors claim, I believe correctly, that the term “ebook” increasingly tends to mean content rather than platform.

There’s one stumbling block here, toward the end of the “What is an ebook” section: “Implementing ebooks is a complicated business, and publishers are anxious to protect their profits and have taken steps to ensure that ebooks do not suffer the same fate as the music industry during the ‘Napster’ episode.” It’s unfortunate that the authors simply accept that the “music industry” suffered some horrid fate because of Napster and its ilk; the evidence is just not that clearcut. The paragraph goes on to discuss the problems caused by DRM; apparently the UK has a law very like DMCA.

I also found it a little odd that the authors regard library-owned PDAs, preloaded with ebook content, as “less of a financial risk to libraries than dedicated ebook readers.” iPAQs and Pocket PCs cost much more than dedicated ebook readers—but they’re a lot more likely to stay in production. They go on to say, “Smartphones may be the technology of choice for future generations…” which could be true, but is also a little disturbing in its implications for coherent text longer than 50 or 60 words.

Those are quibbles. Particularly for UK readers, this paper offers an excellent survey of what’s available (the various ebook services), current problems and advantages, and some of the grant-funded projects actually in place. Recommended as a good survey discussion.

Littman, Justin, and Lynn Silipigni Connaway, “A circulation analysis of print books and e-books in an academic research library,” Library Resources & Technical Services 48:4 (October 2004). Read as a downloaded preprint; pagination not available.

At the time this article was written, both authors worked in OCLC’s netLibrary division—and the analysis concerns a netLibrary collection at Duke University. A yellow flag for possible bias may automatically go up; in this case, I don’t believe that’s a problem. Indeed, part of the scholarly apparatus that takes up several pages of this 22-page article (probably closer to 11 pages in print: the preprint is doublespaced) is a thorough analysis of previous research comparing ebook and print book usage.

It’s a tough comparison to make. As the authors note, most studies (including this one) fail to consider in-library use of print books, thus inherently undercounting print usage (it’s impossible to use an ebook without that use being counted)—and in some academic libraries, studies have shown more in-library reshelvings than actual circulation, suggesting that the undercount may be 50% or more. Additionally, an ebook “access” is typically only a few pages, particularly with services such as netLibrary; the equivalent of a single print book circulation may involve several ebook accesses.

For this study, the researchers decided to avoid some comparison problems by ignoring—well, almost ignoring—the number of circulations or access per item and considering instead the percentage of items that were accessed or circulated. That helps—but the earlier notes identify a problem that continues with this study: Print books used in-house but not circulated (a type of use that closely resembles typical short-term ebook use) are counted as “unused,” while their ebook equivalents are counted as accessed. On the other hand, this study does precisely match ebooks and print equivalents, discarding items only available in one format and looking at just under 8,000 titles in all.

The results are interesting. Of all the books available in both forms, 39% were used in both forms; 34% were used only in ebook form; and 27% were used only in print form. (Just below that chart is the single case in which the authors couldn’t help but note that access per ebook outnumbered circulations per print title—but if in-house use of print books is equal to circulation at Duke, a number that isn’t known, then “outnumbered” runs the other way.)

There’s a lot of other data here, clearly and fairly presented. Recommended as a careful comparative case study.


Cites & Insights: Crawford at Large, Volume 5, Number 4, Whole Issue 60, ISSN 1534-0937, a journal of libraries, policy, technology and media, is written and produced by Walt Crawford, a senior analyst at RLG.

Cites & Insights is sponsored by YBP Library Services,

Hosting provided by Boise State University Libraries.

Opinions herein may not represent those of RLG, YBP Library Services, or Boise State University Libraries.

Comments should be sent to Cites & Insights: Crawford at Large is copyright © 2005 by Walt Crawford: Some rights reserved.

All original material in this work is licensed under the Creative Commons Attribution-NonCommercial License. To view a copy of this license, visit or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.