These are the words that I picked from the material I received from the program organizers that I’m supposed to talk about: providing adequately for information needs across a statewide higher education community. That being Ohio and that being not only Ohio State with fifty thousand students and a big research agenda, but Belmont Tech with fifteen hundred students out on the east end of Ohio with a technical college orientation; and everything in between. That’s about thirteen so-called universities, twenty-three community and technical colleges, and about thirty private liberal arts colleges, mostly of the small variety, fifteen hundred to two thousand students. In grand total it’s about three hundred eighty thousand student FTE. That is the community we’re trying to serve collectively.People today, so far, have talked about fundamental changes in scholarly publishing, and we’re talking about how to make use of the material now with whatever players are out there. As you’ll see, we will dance with devil. So this is the cheap and superficial portion of the program. And I apologize for that. I do think, though, that what we’re doing will provide some fundamental changes because of how we use the information. I think how we change the use of it is very dramatic, and I think ultimately it will bring fundamental changes. But I would agree very strongly with Bill Arms that we do not have what we’re doing programmed out very far into the future. Experience is the only way we’re going to figure out the new equations, and I think that what we’re doing will add to that dramatically. But there’s no way to think this one out in a complete sense. So, we very much believe we should experience it, let’s change things, let’s get our hands dirty and see what happens. So I want to walk you quickly through what’s motivating us to move ahead, to "dance with the devil," and, more importantly, look at six months worth of data we’ve analyzed on use of our Electronic Journal Center. I think you’ll begin to see some dramatic, new changes in patterns of use of information that are far different than what we’ve historically seen in a print environment.
Let’s start with the assumptions that really are the underpinning of what’s motivating us, and the first one, of course, is we didn’t, we don’t, we will not have all the journals we need. We don’t believe that just because we bought a journal that its clearly what we need. We don’t believe that we’ve ever made those decisions correctly, or had the money to buy all the things that people could possibly use. One thing we do know is that the use of information is extremely elastic. If you make it very easy to get to, people will use it in quantities that are far beyond what they’re used to. We’ve been doing patron-initiated non-mediated book borrowing since 1994 across our institutions as we’ve expanded our system. Before that time we used to do thirty thousand book loans a year among Ohio higher education institutions. We now do five hundred fifty thousand requests. The patron can easily find it, push the button, and we deliver it very quickly. It’s that kind of elasticity that we see in physical material. We make it easily available, and certainly expect to do so with electronic material as well.
If you the past as a guide to the future you’re not going to do the right thing because the future is far different from the past. When you start changing all the variables of information access, you have to assume a radically different ballgame. Additionally we have the tools to do it. Certainly the tools are good enough and we can always make them better. We always have rising expectations. But we have systems today that will greatly ease access. We have demonstrated this to our satisfaction and just say to ourselves, "Keep going."
Why are we so adamant about moving ahead? This chart(by the way you have all these things in your handouts that you picked up at the back of the room) looks at thirteen priority publishers that our people identified as ones we should be trying to get access to in electronic form. This cart looks at our university community. You don’t necessarily have to know who all these are specifically but just look at the percentage of titles that we own of these thirteen priority publishers. These represent over two thousand seven hundred titles. We don’t have diddly, and we have less of it every year. Only Ohio State can claim over half of all titles published by these thirteen publishers. These are for-profits, societies, a variety of different types of publishers.
This next chart is the slide I pulled out of a presentation I made to the university provosts a couple years ago. It looks at our Elsevier print subscription trends, actual and expected. In 1994 we had over four thousand print subscriptions in our universities. By 1997 we were headed down to three thousand six hundred, and we were going to be somewhere "out there" on the same downward trend into the future. So we were looking at a continual decrease in the number of print subscriptions that were accessible on our shelves. In the meantime, we lose eleven per cent of our titles we pay forty-two per cent more. What a deal. We’ve projected that we were going to spend five to ten per cent more just because of the resistance to cutting and slashing and burning your subscriptions. We were going to find a way to keep enough money in the system to spend more and get less. It was inevitable that we would spend more money on Elsevier if we were to continue to act individually, and get less for it.
We looked at what they wanted us to pay for a collective license that would give everybody access to every single title, not just the ones they owned. We said here’s what we expect to spend as individuals, and this is what they want us to spend for a collective license. Gee, that’s not so far afield. When I tell a library director who has twenty per cent of the Elsevier titles that he can now have all of them, that comparison isn’t so bad. It’s darn good in terms of the cost-effectiveness of that expenditure. When we look at it internally, as a consortium, we can actually pick up some of the cost centrally, leaving what the library actually has to pay in the lower end of its expected spending range. And in return, while we as a group spend what we were going to spend anyway, each university gets twice as much, three times, four times, five times as much information access as it had the day before.
What are we trying to accomplish in this type of scenario? Step 1 -- and I say "Step 1" because a lot of people are trying to make this a simple one-step-transformation to solve all our problems, but it’s going to be multi-step. Our first step consists of these things. We get more information and we get more information for each dollar spent by creating an environment of increased access. We make every dollar we spend more useful, more valuable. We are now in a position in Ohio where we say to our provosts or the board of regents, "Give the library system more money and we will buy more information." We no longer have to say, "Give us more money next year so we can buy less." We’ve changed that fundamental, unhealthy, year-to-year groveling process to one where we can actually stand up and say we can do more with the money that you give us. We want to begin to gain control of costs. It is not a simple process. If it took three hundred years of doing what we’ve been doing, you can’t expect it to change overnight. You have to start the process somehow, so we’re saying we want to begin to gain control of costs, begin to stabilize the economic equation. And we do that through group licenses where the cost is known and predicted and controlled. The license has a maximum amount to it in return for the increased access that we get.
There is not a single solution. I can’t go to the vendor today, to a commercial or non-commercial publisher, and say I want more for less and I want it immediately. But I can begin to take an equation that you have all experienced historically and begin to bring it under control. And that’s what we’ve been doing with our group licenses that are over multiple year terms.
We’re all flying blind; we’ve been flying blind for a long time. The best statistics we have are the kinds of things that Ken just brought up in terms of what did you observe their use to be. We don’t know a lot about the dynamics of information use, and we know zero about their use in an electronic environment. We know that they can’t be the same as print when you change all the variables in terms of speed of access and ease of access. So, we have to get an environment where we begin this new experience to have a better understanding of what we need, what’s the real equilibrium of electronic information access. Then we gain a better sense based on experience rather than making it up as we go along.
If we can do some of these things, we can move to Step 2, which means using our improved electronic access and our increased ability to measure, to get a better measure of use and value; and maybe come up with a better, more rational set of what information resources we really need over the long haul, at an affordable price. I can’t tell you that Ohio needs all the Elsevier titles. I can’t tell you anybody needs all of them. I can tell you this: ask me a year from and I’ll have a much, much better idea of what the answer to that question is because we now have all the Elsevier titles accessible electronically to all our institutions. As we move forward we’ll be better prepared experientially about what our next step ought to be. I think this may mean there is going to be a fallout, there is going to be a change, there are going to be some winners, there are going to be some losers. But we will at least be able to do this -- at least from our perspective -- with much more information available to us about what we really need to do to help move the marketplace in the right direction.
So what do we know so far about what this type of attitude and set of objectives will lead to? I want to talk about our EJC experience. EJC is our Electronic Journal Center -- a software and hardware site maintained by OhioLINK. We are our own journal aggregator. The EJC consists of a growing number of publishers amalgamated to provide direct retrieval as well as linked retrieval from other resources. It is one that we manage and maintain on behalf of the Ohio libraries. Now, caveats you have to remember. This thing is only about seven months old and we’re only looking today at six months of data. The first six months of data of its little life! It’s the wrong six months. April through September is not really the period you want to measure. Ask me six months from now and I’ll have a heck of a lot better information for you. It’s only two publishers, Elsevier Science and Academic Press, but they total 1,325 titles. We did go after the eight hundred pound gorilla first!
Frankly the currency and quality of Elsevier data have been sub-par (to be kind) during this period. They have yet to have their production system transformed to support an electronic information environment. So, in this six-month period we’ve been running four to eight weeks behind print availability. We have missing publications. Everything they’re delivering has been scanned image so photographs are unreadable. Of course, they’re changing that. They’ve been transforming their production system to deliver true PDF, to be concurrent. We’re the beta site for their new production system. But, in this initial period that has not been the case at all. We just turned it on in April, after waiting for a number of improvements in the Elsevier data that we knew just weren’t going to happen. So we said we’re going to declare victory and just open it up; there’s enough there that people are going to find this of value. Other than what the campuses might have done in terms of sending email out to faculty and students, we didn’t really do any additional promotional activity or prepare support materials until this fall. We just turned it on and let it rip. We’re in a shake-down cruise anyway. We’re not going to really begin to promote this until we get some of the bugs out and we expand our publishers’ base. Also during this time there were no links yet from any of our A&I databases or our catalogs.
One of the things that we do when we put a title into our EJC catalog is that we supply all of our local systems with a catalog record -- a hot link -- so that if you happen to do a search and retrieve that serial title, you’ll have a link into the EJC. We’re also developing the linkages out of citation databases to the full text, as is the California Digital Library, I believe. Last month we turned on the link from the ISI "Web of Science." We maintain the three ISI citation indexes on our system. Now if you’re in that system and you find a citation that we have in the EJC, you’ll get linked. Conversely, if you’re in the EJC and you’re at an article and you want to find out who cited it, you can go back to the ISI system. But those, again, were not present during any of this period of time.
We’ve been counting weekly article downloads. These are PDF article downloads from our EJC. From April through September, we were running from two thousand to three thousand articles per week. As we moved into our first academic term, we started to climb very, very rapidly.
It doesn’t even begin to represent the kind of heavy use that you’ll see as awareness builds, publishers grow, and as we make links, ie., all the things that are going to encourage and stimulate use. But, nonetheless, this body of use represents sixty-six thousand articles downloaded in the first six months.
I don’t think it’s the absolute level of use that matters at all at this point. What I think is important is the distribution of use across the 1,325 titles because it’s too early to comment on what the volume level really means. Most of the iniital use is in the universities. We have limited use in the two-year colleges. We have about nine or ten private colleges that are accessing it, and they’re making limited use. Most of the activity is where you would expect it to be at such an early stage -- the university community.
In the first six months, 1,270 out of the total 1,325 titles had at least one article downloaded. In October we picked up about twenty titles, so in the first seven months 97.5% of all the titles were used at least once. We don’t have a 20/80 rule; we have a 40/80 rule; 40% of the titles account for 80% of the articles, and of course you have the extremes. You have a little over 1% of the titles doing 10% of the articles, and you have 43% of the titles doing only 10% of the articles. What I think we’re beginning to see as we analyze the accumulating data -- we accumulate this data off our system by week, by month, by quarter, by aggregate period of time, and by school -- is that this line is going to go flatter. Use is going to become more distributed across a broader number of titles. We’re going to see it less concentrated as time goes on.
Now, this of course represents the aggregate experience of our whole community. Ohio State more or less stands alone in terms of the degree to which it has in print the majority of titles in our EJC. But do they have a very different pattern? No, they have a very similar pattern. Ohio State, which I think has the highest number of Elsevier titles of any of the Big Ten institutions, has a very similar distribution which says that even though they have a significantly higher percentage of the Elsevier titles, even they don’t have what everybody will use if you give them a chance to find it.
This looks at the number of articles downloaded by institution. The absolute numbers aren’t so important but you can see that Ohio State jumped on this very early. Even though the other institutions are relatively low, they are happy with the results. For each school the real thing to look at is the relationship of articles downloaded not held in print compared to articles downloaded that were held in print. And you can see that, except at one end of the spectrum, you have most institutions where the majority of downloads are journals they never had on campus before. It’s not just a convenience product. And when you look at titles, it’s even more dramatic in terms of the number of titles that were downloaded off of the EJC in this early six-month period at each institution. When you look at percentage of titles downloaded not held in print, everybody is over half, except for Ohio State where its 38% of the titles that were downloaded. To us this says there is a dramatic expansion of accessibility that was impossible to deliver in a print environment at our institutions. When you look at it on an article basis, it’s the same thing. It’s a little bit lower which speaks well of what the librarians have been buying -- it should be lower. But, again, it’s the same relationship: at almost all the institutions, except for those at the top with the most print subscriptions to begin with, over half of the articles downloaded they never had access to before.
Think of the interlibrary loan costs it would have taken us to support this, the costs of the "rule of five" that might have been paid. I’ve actually analyzed that, and it’s thousands of items that we’re not subject to pay for or subsidize. The good news, at least if you’re a librarian, is that the average article downloaded per journal title that’s held in print is higher than that for the ones not held in print. People use what is familiar to them in the first place. So they do use what is on campus as a convenience tool a little bit heavier than they use the stuff that isn’t. We’ll see over time how that plays out but at least it says that buying practices haven’t been random, certainly people are subject to habit in terms of the titles they go to.
In conclusion, we are quite pleased with the breadth of use at each campus. We consider it to be a dramatic expansion. This is what’s happened in the first six months; what’s going to happen over the course of a full academic year? What’s going to happen as we continue to grow this site? We think it suggests that the leveraging that we’re doing--- with essentially the same amount of money that would have been spent with maybe a little bit thrown in from the central site where necessary--- delivers a tremendously better value in terms of the amount of information that can be consumed by the user. We’re going to continue to add publishers on similar first generation licenses as we have with these several. We’ve already added the Project Muse collection -- we turned that on this past month -- we’ve got a couple others that we’ll be adding with both society and commercial publishers.
A lot of effort is underway to grow this site significantly over the course of 1999 and I would say that from my experience talking with a wide range of both commercial and non-commercial publishers. I don’t think you can paint either of these two groups with a black or a white hat. I think you have to look at the individual publisher. I find that there are many commercial publishers with whom it’s easier to do business with than with non-commercial publishers because of the fit of my business model and their business model. I’m happy to throw oil on the fire when it’s appropriate but I don’t think we should make any blanket statements about the publishing community because it really does vary publisher by publisher. We will continue to create robust links. Obviously we have started to do that with our ISI links. We have twenty-six journal citation bases under one search engine that we control, so we will just add these types of robust things. We will do all the things that are written about and talked about in the library press about how to make all of this accessible to everyone. We’re certainly well along that evolutionary path.
We’re already beginning on the second generation licenses with Academic Press and Elsevier, and the big question is are they going to be different or similar? One of our key objectives in second generation is more cost control. At this point, we don’t know enough about the use of information to say we know what the new equilibrium is. Do we know which titles we don’t want, if any, or do we know we want all of them? The default is we want to keep getting all the titles from all the publishers that we can make licenses with, and we want to make them available across the state until such time as we can figure out where the dust is going to settle.
In the meantime, with more economic bargaining power as a statewide group we’re going to try to tighten the cost control issues. If they want us to continue to stabilize their economic equation, they have to make it more palatable for us to lock up that kind of money because the big downside for libraries in this is that you lock up all their money for a given publisher. They don’t have title by title control to save funds. Under the group license there is only a modest reduction when you get rid of the print copy. What we’re doing is we’re getting a license to intellectual content. The medium is a secondary issue. For us to lock in this kind of money across all of our institutions on a publisher by publisher basis, we’re going to tighten the screws a bit.