An LJ Round Table with the Aggregators

In our most recent Academic Library Budget Survey (LJ 9/1/01), one trend in particular stood out among librarians surveyed: the rapidly growing popularity of aggregated article databases. It is a trend mirrored in public libraries, which also report a significant up-tick in patron usage of databases. But as libraries and distributors of aggregated articles and journals have scaled up to meet user expectations, challenges and questions abound—for libraries, for publishers, and for patrons.

With these questions in mind, the editors of LJ invited a host of key players to a round table discussion of aggregated databases at the recent 2002 American Library Association Midwinter Meeting in New Orleans. Moderated by Yale University's Ann Okerson, the event marked the first time that three of the major "aggregators" (EBSCO, Gale, ProQuest) had sat down together to share their thoughts on the rapidly changing information landscape.

As Okerson noted, "The expectations of our readers have changed in the last five years [because of] Amazon. com. Our readers expect [that] kind of speed of service," for both articles and books. Of course, such expectations are not met easily. And for libraries, managing resources, providing service, and offering access in the age of the digital database has never been a more delicate balancing act.

In a brisk 90-minute session, our panelists cut to the core of some of the most vexing issues facing librarians, aggregators, and publishers today, among them embargoes, exclusives, and journal cancellations. While more information is available than ever through digital databases, publishers, concerned with losing traditional subscription revenues, have implemented a range of safeguards, including embargoes of current content and exclusive relationships with certain aggregators. But have such practices, as some librarians assert, led to a fractured and less efficient information landscape?

Our panelists also discussed the trend toward consortial buying. At the beginning of the round table Okerson stressed the importance of consortia in the current collection development environment. "We learned that as a group of libraries together, we could achieve a lot more than we could behaving as a single library for a single resource," she said. But how does consortial buying change the way libraries and information providers operate? Does consortial buying lead to a homogenized library environment, where more libraries have access to the same materials, leading providers to seek competitive advantages through other means, like better interfaces and indexing?

No discussion regarding aggregated databases would be complete without touching upon a major issue affecting their comprehensiveness—the Supreme Court's 2001 ruling in the landmark Tasini v. New York Times case. Contrary to publishers' arguments, and the opinion of a lower court, publishers, the Supreme Court ruled, must have the explicit permission of freelance writers before including their work in electronic databases. The response from some major publishers, including the Times, has been swift—pulling content from their databases where permission was in question. Is the Tasini case a roadblock, or just a pothole on the information superhighway?

After more than an hour of give-and-take on complex issues, Okerson asked our panelists for their views on the future. They responded with some surprisingly upbeat thoughts. Are there challenges ahead? Absolutely. But with cooperation, and the kind of insights shared at this forum, there is much to be upbeat about. What follows is a distillation, in five sections, of the lively and wide-ranging discussion.

Okerson: I have always said that the people who understand librarians best are often not librarians but those who work with us and observe us. And I think that librarians deserve to have anthropologists or aggregators or content providers study them and tell us what they think we're doing. So let's start by going around the table and characterizing or encapsulating your view of how libraries or the market is behaving today.

Bell: The question brings to mind an anecdote. I recently met a dean of the graduate school of a large university who was the chair of the search committee to find a new library director for his institution. As we were conversing, he said to me, "I had no idea how complex and large your job is." This complex environment pertains to any kind of library. But the bottom line is that we're trying to create access—to connect people with the information resources that they need to do their research, to be a lifelong learner, or to be whatever it may be that they want to do. And libraries need the support of publishers and aggregators to help us achieve that goal.

Brooks: I think librarians are struggling with a lot of issues right now. For example, journal subscription prices are increasing an average of seven to eight percent per year, and as a result, [librarians] are looking for a way to provide more information to their patrons and students and at the same time not bust their budgets. Librarians are certainly in a tough situation. I guess what we've observed is some librarians coming up with very creative ways around issues and some librarians making some unknowing mistakes that come back to hurt them later.

Smith: Before I went to work at LexisNexis, I was an academic librarian for 18 years. I would say that librarians today are sort of placed in an untenable position. I think that as a group, librarians are unfortunately extremely unrealistic about the economics of the world. Having been on both sides, I can say that. I think librarians are trying to maintain a semblance of what they knew in the paper world in the electronic world, and I think that there's going to have to be some hard choices made, particularly in the academic world, where there's more and more faculty and more and more expectations.

Melkin: I was in publishing for 20 years. And now coming from an aggregator perspective, dealing with publishers whom I talk to, I have a sense of publishers' views and views of my own about librarians. I agree, it's a very difficult situation for librarians. I think everybody's trying to find out how to best deal with this situation, but I think there's a great appreciation of the bind that librarians are in.

Price: I'm going to use an analogy that actually came from Gale. The argument was that you're in a canoe and you're paddling down the river and everything's nice and smooth. But occasionally you come to rapids and you have to paddle like heck. Sooner or later you're going to come out the other side. Well, we're in the middle of rapids right now, and there's no end in sight. The Internet has changed the whole dynamic of collecting material, accessing material, and the per unit cost to deliver it.

Barnes: To use Vince's analogy, I think we might be a little bit further down that river. Certainly 2001 was a hard year for a lot of reasons. But I think there was another event that occurred in 2001 that kind of vindicated our industry and vindicated what librarians have been saying for a long time—the flame-out of the dot-coms. There was a lot of craziness in 1999 and 2000. Business models that didn't make sense. There were calls for the end of librarianship, and, in the end, I think a lot of us feel pretty good that we're still here. Our business models still work. Librarians and libraries still work.

Bell: Using aggregator databases as a way to save funds is a totally shortsighted management approach. And I think a lot of libraries got burned early on with LexisNexis. When we first started getting access it was a bonanza. But we learned very quickly that you don't have control over all those titles in the database. Some titles drop out, new ones come in. I think there's two different things to look at: an aggregator database [like EBSCO, InfoTrac, or ProQuest] vs. an e-journal collection [like Project MUSE], where I know that more than likely they're going to maintain those publications as long as they're published. So I can't go along with this idea of cutting back, saving money by looking at your aggregator titles and saying, these are the ones that we should target for a cut. But I do know some libraries that are doing that.

McCracken: [The University of Washington] went through a serials cancellation project last year and that was a big learning curve for us as librarians: when you're accessing an aggregator database you're buying a service, it's not a guaranteed list of titles. When we made decisions, we said that things like Project MUSE or services like Ingenta may be an area where we cancel the print subscription. When it was in an aggregated database, that was not enough reason to cancel.

Okerson: What's interesting, I think, is that as more institutions around us are experiencing cancellations and cutbacks, they think it's okay, because there are libraries of last resort. Like the Library of Congress with its copyright program, or schools like Yale or Harvard. As long as these libraries have holdings, they think their libraries will be okay. In the short term, the institutions that cut their print and keep the electronic save money, putting a kind of heavier responsibility on an increasingly smaller number of libraries to be their national and international support and backup. And I think it's a risky position for us to be in, because I don't know for how long or to what extent we will be able to deliver on that promise

Melkin: I've had this situation in front of me because the publishers who work with us, whom we host as full text, ask me if I think they're going to lose their print subscriptions if they are in the aggregator's database. My advice has been that publishers have to try it and see. They can make a change. They can put in an embargo if they're concerned about the current year's content. It may be an incremental market for them, and [embargoes] let them feel at least in the short term that they protect their subscription revenue.

Price: But what does the publisher see? If a publisher sees budgets are squeezed at libraries, and libraries are therefore reevaluating all the resources they buy, especially in print subscriptions, and they drop a title, how is the publisher supposed to tell what caused that to happen? Were aggregated databases a large or a small factor in that? It's important for publishers to figure that out, but there's no solid way for them to determine that at the moment. We try to present the ProQuest databases as a service, rather than a collection—our argument is libraries have to buy an item at least once. And as a service, accessing a title within a service is not buying—you don't own it, you rent it. But if you feel the need to collect it, then you have to look at an e-journal [or other format] and subscribe.

Bell: What concerns me greatly is what use are databases going to be for full text if there are just huge numbers of embargoes. We need to rethink the whole way this is working. It seems like a defensive posture to create embargoes. The psychology being that if we put the embargo in place, then a library can't cancel that title because if they're depending on the aggregator database to get access to it, they're going to disappoint users by saying, "Oh sorry, you'll have to wait six months to see that next issue." But if a library needs to cut titles because it has a significant budget cut, the presence of an embargo is not going to stop it from canceling a subscription.

Barnes: A lot of issues result from a lack of understanding of why librarians make the decisions they make. Often librarians when quoted in a trade publication, or when talking to a publisher say, "I canceled your title because I can now get it in this aggregated database." The publisher makes a mental note of that. They hear that five more times, and they go, "This is bad." Aggregated databases are bad and how do we solve that? They embargo titles. But that's a short-term solution, it won't last for long. There are new business models that are coming out that will be announced shortly that will put a different spin on things. I don't think embargoes are a long-term solution. [Later in the discussion Barnes announced an alliance between Ingenta and Gale that allows Gale InfoTrac users to search also Ingenta and get the full-text current journal, if a library subscribes to it.]

Brooks: Clearly, embargoed access to a journal is inferior to the e-journal. But the whole goal of databases is to bring new journals into the library that you didn't have before. If a journal is really important to you, you should certainly be subscribing. And most database vendors are linking from the database to the current journal, if [the library] buys them. Publishers make only a tiny fraction of their subscription rate from their databases. Anyone can figure out the math. If you figure a $2000 journal is available for $20 through the database, the economics there makes it impossible for a publisher to see the databases as at all similar to their subscriptions. And because of the huge number of journals that are available in these databases, it would be almost impossible to learn which titles a library is planning to cancel, and say if you cancel this one, the database goes to this price. So publishers came up with embargoes.

Price: That brings up the other attribute of databases that typically isn't discussed. Aggregated databases are also finding tools. They're indexes to help deliver the user to the material. Now, if it happens to be in full text, users are much happier people. If it's in the stacks at the library, then that's great, too. If they have to do an interlibrary loan from Yale, then they have to wait. But the point is, did they find the articles they needed by using the tool?

Brooks: The Tasini decision has had a big effect on what's available in aggregated databases for general magazines and newspapers. But it has not really affected academic journals. I don't know of any academic journals that're calling us, saying, "We need to take our articles out for these 500 authors." We are getting those letters, however, as our friends at Gale and ProQuest are, from the publishers for general magazines and newspapers asking us to remove very large numbers of articles. And that is something where, if libraries are relying on electronic-only for those magazines, they're not going to get everything.

Okerson: Is it hard to pull that control or is it easy?

Brooks: Well, from EBSCO's perspective, it hasn't been hard because we prepared for this before the Court's decision. Our programming is set up so that we can go in and just remove all the articles from those authors. We would rather not do that. But at this point there's no solution. And articles are just disappearing at the direction of the publishers. The ideal scenario would be one where we could keep content and pay the authors directly. But at this point if you ask the publishers, they're not interested in that. They just want to take these articles off. They don't want to deal with the headache, so there is this problem that's come out of Tasini.

Barnes: The other side of that is, for a lot of the publishers that are concerned, the library market is not their primary market. We see it, as Sam mentioned, especially prevalent right now in newspapers. I won't name the publisher, but one very large newspaper publisher had signed up, and we were ready to put them into our databases, and they pulled all the newspapers back. What are you going to do? We already had sales to libraries that had been asking for that material. But they were skittish. They just don't know which articles they can republish. It's not that it's technically hard to pull information. We all have the capability to do it. But it's disappointing that the only solution right now is one of butchery.

Bell: In some ways, though, the publishers are punishing the author. Authors in most cases, I would think, want their articles out there. I think mostly it's totally inconvenient for publishers to have to go back to all these authors. So instead, I think we're seeing a lot of date ranges where a particular publisher [will say] "from this day forward, all the articles are safe because we've had the authors sign contracts saying that the articles are turned over like they are in academic journals. But everything before that, even though you have a lot of big back files for these magazines, they're gone," and it's across the board for all aggregators.

Okerson: I wonder if Diane or Vince could give us a sense of percentage of content that you may have had to pull. Is it a little, is it a lot?

Price: It's little in the grand scheme. And Sam's exactly right. It's the general publications and the newspapers. Scholarly databases are only slightly impacted. Newspapers, it's all over the spectrum. There are some where, honestly, it appears they kept very careful records. They give us their list of authors. It's a relatively finite number and it's low single-digit percentages. There are other publishers that say "anything in the author field that doesn't say 'staff writer' remove it." And that is way overkill because big chunks of content are coming out.

Okerson: Is that costly?

Price: It depends. If it's programmatic, if they say, it's these authors, it's not so costly. It's when the publishers don't really know that it gets difficult. Removal requests may come in dribs and drabs, and worse sometimes, they'll say, "Take this stuff down" and then they'll call and say, "Oh, you can put that stuff up again." That's just difficult. Publishers can't figure out who all the authors are and they don't want to make the effort. Probably because it's a lot of authors over a lot of years. It's expensive.

Melkin: Just to emphasize again how far-reaching Tasini has been, when I started at CatchWord [bought by Ingenta], as part of our setup process with a new publisher, this issue didn't come up. But it comes up at every meeting now because everyone's concerned.

Smith: From [the LexisNexis] perspective, the difficult part is actually explaining this to the librarians who call up and say, "It was here yesterday, it's not here today, what happened to it?" And then you go through the explanation. But the number of articles is very small in relation to the full size of our database.

Melkin: Going forward, most of the publishers are changing the way they do business with freelancers, and the issue will get smaller. Though for older material, it's still going to be a problem.

Price: I think there's little doubt that [consortial purchasing] is creating a large homogenization of databases. The large megadatabases that you have now are directly a result of consortial activity, where the most important criterion was numbers. And I don't necessarily think that's a reflection of quality. In consortias' defense, they have to serve a wide range of users, from schools in many cases to [Association of Research Libraries institutions]. So they're looking for the largest common denominator database for all their users.

Okerson: I believe that consortia in many cases, at least in the United States, are licensing many of the same full-text indexing and abstracting services and aggregations from the same publishers on reasonably similar terms. And I do think the effect of this is moving us, across libraries, toward having content that looks much more alike. Where it's electronic content, I think that's not a bad thing. But 25 years out, as huge proportions of our budget are devoted to electronic, not analog, resources, library collections will look very much alike and the thing that will distinguish them, in part, is the depth of their print and special collections. But to assign responsibility for that trend to aggregators alone is probably a lot narrower than the real case.

Brooks: Consortia purchases of databases have dramatically improved the quality of databases. There are a lot of new, better titles in the database as a result of competition. And unlike journal subscriptions, the rate of inflation for database prices is not even close. They're negligible. That, I see, as a positive.

Smith: Consortia are actually raising the bar as far as library services. I know I've been told by many a librarian, we could never have afforded this if it had not been for the consortium purchase. So I'm sure there are lots of libraries that have materials. It is a homogenization, which sounds negative, but I think it's raising access for libraries, and so it's actually a good thing.

Barnes: What I'll touch on is a trend that has been happening, I'd say significantly in the last year, and that's interlinking systems. It's really bringing everything together. There are metasearch tools that have been developed. We have them and most of the vendors have them, ILS systems have them. This will make getting information much more of a homogeneous experience, so a patron doesn't have go to ten different databases. If you go to any library, one of the things you see and one of the things you hear from the reference librarians is, "I take them to my web site and there are 35 or 100 databases listed, and nobody knows where to start." Well, that problem is quickly getting solved. Patrons will be able to go to one place within the library, do a search, and branch out, much like the web does. All of us work together now, whether we like it or not. We all link to each other's products. And that's going to be even more pronounced five years out.

Bell: I'd hope if we were [to meet] back here in five years I could paint a picture of a library environment where we are able to satisfy our users' needs by having as much ability as possible to connect them to the information that they need. I do have concerns with issues [like] embargoes and exclusives and Tasini and where this is all leading. I'm encouraged by some of the things I've heard today about aggregators and publishers investigating other models that will help make information more accessible. I can't help but think that the technology is going to go far beyond what we've even imagined and that might enable the kind of access I dream about.

Price: I agree with John that the linking is fundamental. Now the question is, what do you actually create as a company, as an aggregator? In the context of the greater collection and the purpose of the library, those who find a way to satisfy user needs are going to be there five years from now. Maybe this will settle down and we'll find out that it's better if all the systems talk to each other. Or maybe not. Things are certainly in a state of flux, but linking is very much how we're going to get from where we are to where we're going to be.

Melkin: The bar's been raised. The expectations are there. Looking forward, I think about the article economy—is the idea of a journal moot? Are articles only valuable individually? I'd say that there will be many more e-journals, there will be more content in databases. And there will be more done with linking. I think in five years we'll know a lot more than we do now about how this affects subscription revenue. In a lot of ways we're in transition now.

Smith: Interlinking and going across databases is clearly the way of the future. That being said, I think what aggregators or publishers, whatever you want to call us, really have to start thinking about is how we can add more value to our products. And I would certainly hope, from an aggregator's or publisher's perspective, that librarians would start thinking more about the issue of information literacy. Is more and more full text really meeting students' and researchers' needs?

Brooks: For the vast majority of issues libraries are facing, I think the next five years are going to be very positive. I think the developments in databases are positive. I think competition is driving that. I think the fact that databases have become so popular is helping. Most of the database companies are reinvesting their sales revenue into their companies. And as a result you see a lot more linking, which takes work. I think in the near future you'll see vastly improved features like subject thesauri in various databases, and that's obviously a positive development. You'll also see much larger back files available, maybe not for general magazines where the rights for the articles is an issue but for the scholarly journals where there really isn't that problem.

Okerson: In five years, I'm going to be interested in the answers to questions that I now think about every day. Will the industry, that is the information industry, the library industry, which has come to a very good place with journals and aggregations, figure out how to get to a comparably good place with full-text books, especially scholarly books? I think we are with books now where we were with journals almost ten years ago. I'll be interested to see, since I'm in an academic environment where scholarly journals really matter, just how far the author movements toward retention of rights, the Public Library of Science, the free science online movement, just how far that goes and whether it will have had a transforming influence on that part of the information industry that's really important to people. I agree with Diane that if we can stop thinking of ourselves as "I'm an aggregator," "I'm a university librarian," "I'm a reference librarian," and think of ourselves as providers of information, then we don't have to worry five years out, do we have a job. Because the job of everybody here is to add value in a certain way. As long as we have good ideas about that and we live up to our ideals and promote our values, we will all be employed.