The 139th ARL Membership Meeting

Jurys Hotel, Washington, DC
October 16-18, 2001

Report from the Scholars Portal Working Group
ARL Business Meeting
Thursday, 18 October 2001

Jerry D. Campbell, Chair, Scholars Portal Working Group

Mr. Campbell (Southern California): I would like to give a report on the ARL Working Group on Scholars Portal. The working group, by the way, is made up of all of the libraries who expressed interest in joining up, the quid-pro-quo being that since our philosophy has been that we want to buy it, rather than build it, there is no outside money supporting it. So this is all a kind of pay-as-you-go effort, and we are pleased to have as many participating as we do.

Now, let me say just quickly what the Scholars Portal is and in doing so remind you of a poignant, sweet, little country song called “You Are the Reason our Children Are Ugly.”

(Laughter.)

Mr. Campbell: As you know, we, along with all of our partners in the scholarly publishing world, have built a rather complex set of digital sources, and it becomes very difficult to negotiate that in our institutions. And so the simple way to say what this Scholars Portal effort is about is to try to find a Google-like approach where one gets a simple, easily manipulated screen that says what do you want and you type in your request.

And the system searches across all of the disparate databases, including locally created databases, commercially licensed databases and the Web and Internet world as well. And included in the database searches would be the online catalog and other keys to local sources of traditional materials as well. The question we faced was whether we could find some technology that would do that for us.

In searching that out we have looked at about 30 vendor possibilities. I think we might say that there were probably three of them that offered some reasonable percentage of what we were looking for.

One vendor, to the best of our ability to calculate it, offers about 80 percent of what we thought we needed. And it turns out that vendor is interested in collaborating with us to try to build the other 20 percent, believing that that would be a great boom for its business.

And so, where we are at the moment is attempting to negotiate a contract with the vendor we believe offers the products with the richest range of functionality. The interesting thing about it is that all of their products together equal what we think is about 80 percent of our requirements. All of these products exist. It is simply that the vendor never put them together in one package. They are all working one place or another. In some cases, in twos and threes, but the products were not combined in the way we want.

So we hope what we are talking about is a relatively low-risk effort here. Certainly we say to ourselves, for encouragement, one vendor’s products ought to work together.

We expect to have a contract crafted by the end of December. Mary Jackson is just doing a tremendous job representing the working group in trying to create this contract.

What we are discussing is what we would call an ARL or research library package, that is, the specific combination of the vendor’s products that we want. That combination might be then made available to anyone. Certainly others could license other combinations from this vendor in the future. But what we are looking at is trying to create a set of functions that we believe would be really good for research libraries.

The reason we are not naming the vendor is not to be coy, but the first time we name it and it gets any publicity, we lose a certain amount of negotiating power. We want the vendor to understand that we are not talking publicly—that we are not using its name—and we are willing to walk away from the negotiating table on a moment’s notice if they fail to give in on such insignificant things as price and functionality.

And so, we are trying to drive a deal that we believe would really be suitable and preferable for all of us.

We are looking at what we think is a two-year project and we think we have half a dozen libraries who are willing to pay their way in the experiment. If we don’t, I may be calling around to get one or two more. So hold on to your pocketbooks if I call.

But what we hope is that during the first year we will seriously test the implementation of the products that they have. If it is as easy as they say, it is a 30-minute project. If it is as hard as we think it probably will be, it is going to take a better part of a year.

What is hard about it, to some degree, is that we are going to have to make some decisions that in the past we have been able to finesse. Those decisions include such things as if you search across multiple databases and multiple information types, such as online data and traditional books, in what order do you display the data sets, the return? Do you prefer one over another? Do you sample some or all of them?

This can and probably should vary by library. But we are going to have to make decisions about that, and we believe those decisions are going to force us into some more time commitment than maybe the vendor thinks might be the case.

We also then anticipate having some time in the project to work on the other 20 percent of the functionality we would like. This vendor is very interested in collaborating with us at its own cost to develop this functionality. Such functionality might include things like building interfaces to permit the Scholars Portal to link to multiple external systems or whatever.

For instance, we want to interface the portal with a campus system like Blackboard or whatever your campus might be using for a virtual classroom system. Those interfaces could be really important in terms of linking students to reserve materials or other such things that might be identified in the course.

By the way, the name Scholars Portal is much in debate, and we really don’t care what it is called. We just want it to be called something that is reasonably accurate. So that is where we are at the moment and how it is moving. We are guardedly optimistic about it, not trying to underestimate the difficulties of implementing.

We hope, if some of you are interested in implementing the system, assuming that we prove it can be implemented well, that you would notify Mary Jackson and say, “I would like to consider joining as a project participant.”

It is not an absolute obligation, but it does give us a list if we need to try to find one or two other partners who might be willing to go along early. And also, for sure, it would mark your place if you want to be in the next stage of implementation.

Let me say just one other thing by way of conclusion. The working group has been clear all the way along that we do not want to end up with a sole source environment, and we even talked among ourselves about trying to identify a couple of competing projects up front and running a project with two products.

We really don’t want to end up with an environment where we have helped create just one vendor. I wanted to be sure to say that. We are sensitive to this, and once we are able to work out moving ahead with this vendor, we hope that what we will be doing will be suitably public; that other vendors who occupy this space can work hard to create competitive products. I think it does us no good if we end up with just one out there.

Also, if any of you who are interested in this effort and would like to chip in to defray the planning costs, it will only cost you 5,000 bucks a piece. That is what the working group members pay, and it would be great if some of you wanted to subsidize us that way. But otherwise, we are going to conclude this on our own. I would be glad to answer any questions that there may be.

Audience: Jerry, can you give us a sense of what it would cost?

Mr. Campbell: Yes. The monies that we put in as a working group are paying for the costs to help us plan the project and work out contracts, for example, a portion of Mary’s time and so forth.

At the moment, what I can say about the costs of the software and vendor services is that we are in the very low six figures. It is less than any integrated library system than I have ever dealt with by a good amount.

On the other hand, those of us who undertake the early implementation are going to try to calculate what it costs in terms of our contributed staff time, because we actually think that maybe a larger portion of our costs will be in trying to set this thing up initially. And we will also share, among ourselves and with you, the kind of models that we derive.

The notion is that there has to be a default. So, if an undergraduate comes up and types in here is what I want, that the system will search a kind of preset number of things and it will display the results in a certain way.

If she feels like she is an expert user, there will be a button that says expert user, which will take one to a somewhat more complicated set of choices. That way a subject expert or other people can actually tailor searches to their own needs. We hope that an individual would be able to do that once and then save the results of that tailoring so that each time she would go back to it she would get the same kind of really focused search results.

But setting that up in advance is going to take some time. It is going to take some understanding of the data that we have and how we might want that to be mixed in a display set.

Audience: (Away from microphone.) Are you going to have a set of expectations or criteria that you are giving to that vendor that maybe we could see to understand what it is?

Mr. Campbell: Sure. Let me talk with you a bit about what would be helpful there, but we will be glad to do that. Yes.

Audience: On the assumption that phases one and two are implemented successfully, what is the time table for the general application of this to the rest of the membership?

Mr. Campbell: Well, you know, it is really—it is probably dangerous and stupid of me to venture an opinion. But let’s say instead of a short time, it takes—let’s say it takes the whole first year really to implement the existing products.

At that time what the first group would start working on with the vendor is the design of those other 20 percent of the features we want, but we could implement them until whenever they came along.

My guess is that at the moment we have a kind of stable implementation, whether that is six months or a year, we would publish that to the group and, assuming that the vendor could ramp up to handling other institutions, you should be well disposed to move as soon as you wanted to to join us. And let me also say, you could make a deal with that vendor on your own any time. But I think prudence is that we want to see how these initial implementations go, and it certainly gives us, up front, much more leverage, in terms of trying to move that vendor into the space where we want them for a variety of things.

Yes, sir?

Audience: (Away from microphone.) You mentioned yesterday and today—a lot of the—the universities—

Mr. Campbell: Yes. I think the question is—for helping a vendor partner design the new 20 percent of the requirements, will we have any equity ownership in that? “We” being the Association of Research Libraries.

Well, the answer is we hope so, and we are attempting to negotiate that. That is part of what we want to have vested in this contract. How that would be expressed we don’t know at this stage. The vendor is clearly aware that we constitute a primary market for their software. We do have some leverage in that and so we are working hard to achieve that. But we don’t know yet at this stage how much give there is there. We are attempting to work out some value proposition for the association; for the benefit of the expertise that members will contribute to developing a product.

Yes, sir. Charles.

Audience: (Away from a microphone.) Jerry, the vendor that you are talking about is probably one of the most important elements. I think we all can guess who the vendors are that you have talked to. All I ask is that you find a way to alert all the customers of the vendor that you are negotiating with them about this or ask the vendor that you are dealing with to contact us.

Mr. Campbell: That is a very good suggestion. We will do that. So, if you are a customer, don’t be surprised if we call you. Blame Charles, not me.

Any other questions?

(No response.)

Mr. Campbell: Thank you.

President Baker: Thank you, Jerry.