The perspective in this paper grows out of a year in an institution that currently owns or accesses well over 200 electronic databases, including stand-alone CD-ROMs, networked CD-ROMs, locally loaded tapes, and databases accessed from information providers through electronic networks and services. I characterize at least half of these as significant (meaning that they are either expensive or have sizable user base on campus or both). Most of these resources are governed by license agreements which, for electronic information, are the prevailing legal means for specifying how we and our customers may deploy and use the information. These days, we are adding such resources at least weekly, it seems. In passing, I observe two disconcerting things:
(1) we do not have copies of all 200+ of these licenses; some never made it as far as the central library license files and others were shrink-wrap (prefab) agreements whose revelatory packing was not preserved. We did not do as good a job of paying attention to licenses in their early years as we do now.(2) the licenses we have signed are all somewhat different to each other and the uses they permit are not consistent. My greater concern here is that we need to make some sense of these different terms as we present them to our student, faculty, and research users. We are remedying that in part through developing a database of our e-licenses and attaching permitted uses where we know them. This resource will be made available to our library staff.
At Yale, license negotiating and signing are the responsibility of the office of General Counsel except when such authority has been delegated to a specific position elsewhere in the system, as it has been, for Library content, to the AUL/Collections. Such delegation of authority requires that there be mutual understanding, regular communication, and accord between Counsel and the Library. Signing means not only arranging the best possible deal for these so-called "products" but also representing the University's interests. The licenses the Library signs can carry many consequences for the University, and as a steward for institutional interests that go beyond the Library, I must do the job right.
What generally do we do when we negotiate licenses?
What the publishers and I are engaging in is a mutual dance around a prospective transaction quite unlike what happens when one shops for, say, a winter coat. When buying a typical commercial product, we may all negotiate with ourselves -- what we can afford, do we prefer this one or that one, what use will we make of it? In that sort of transaction, the salesperson has an unambiguous role: to praise the product, perhaps to inform the buyer. But in a normal daily buying transaction, say for a winter coat, one seldom asks the salesperson to take a lower price than the one on the tag. Let us think for a moment about the circumstances under which we do negotiate in our society.
Home-buying or selling is very much like that, and good deals may be had, depending on the current marketplace, one's knowledge about it, and one's skills in presenting attractive offers and counter-offers. Automobile-buying is a substantial transaction, though there the tradition of negotiation seems to be what persists; today some dealers try to gain market position by promising not to negotiate with the customer, as if car negotiations were slimy or unsavory and what the customer really wants is a true price. Generally, negotiations occur when two parties each bring a substantial piece of economic clout to the table. In an institutional setting such clout generally involves bulk buying: for example, 200 workstations for the Library or a one-year contract with one's subscription agent (note that one does not negotiate for a single subscription). Licenses for sizable electronic resources are among these "substantial" transactions. Electronic resources are often large and expensive, and even a single research library still represents a significant customer to a given publisher. The negotiations scale up even more when our library and some others negotiate as consortia, bringing large dollars and desirable customers to the table.
When neither party has the confidence that there is a correct sticker price or a standard contract, there is more negotiating to be done. That is clearly the case here. The products are new ones, the ways buyers or licensees can use them are new, the costs incurred by both sides are to some extent unknown, and there is no ready way of comparing one product (say, the license for Encyclopedia Britannica) to another (say, the license for 120 of the journals published by Academic Press) -- and hence no way of being sure which features of the last deal are relevant to this one.
In the area of electronic texts, publishers come to the table with real anxieties that their products may be abused by promiscuous reproduction of a sort that ultimately saps their product's marketability, while libraries are fearful that restrictions on permitted uses will mean less usable or more expensive products. In such a situation, both sides are reluctant to take the other's first offer, and so the dance is necessary. I would also emphasize a feature of electronic resource that is perhaps surprising at first: the discussions and informal negotiations that must occur within the library and university community. The "print" or "traditional" model of selection and acquisition (send a book request to the Acquisitions Department and a decades-old process kicks into action, eventually leading the desired item to its appropriate place on the shelf) does not apply. It turns out that many parties inside the library have a stake in the resource that is to be added and the deal that is to be struck. Our Library, for example, speaks now of a "systems premium" being attached to each new electronic resource; this premium is the cost of analyzing a new resource and taking the steps needed to bring it effectively to our own community as part of a seamless web of university-provided information resources. We cannot yet add any important electronic resource without bringing together a small team of collections, public service, technical service, and technology specialists to assess, identify funding for, and incorporate the proposed new resource.
The traditional means of regulating use of information resources sold to libraries is copyright law. In that world, the Library purchases a physical artifact and owns it. the US Copyright Act of 1976, while not explicit about every use that libraries and readers can make of the copyrighted information, nonetheless prescribes use in a way that is relatively easy to understand, and the means of resolving disputes has been established through the existing judicial (court) system. The US Act says that copyright owners (or those to whom ownership is transferred) may reproduce works, make derivative works, and may exercise the rights of ownership for the duration of the copyright period (50 years plus the lifetime of the author, for personally created works). Nonetheless, Sections 107-120 outline limitations on the owners' exclusive rights. In particular, Section 107 (fair use) allows copying for certain purposes which include teaching and learning and scholarship; and Section 108 allows libraries copy in order to lend and preserve works they have bought.
While many believe (as I do) that the principles behind the copyright law can be extended to the electronic environment, the different stakeholders have different views of the underlying principles -- and the law itself has not been altered to reflect the boundaries of copyright in the presence of technologies that can permit unlimited copying. The United States has seen over the last three years a so-far abortive movement towards regulation of electronic resources through modification of the Copyright Act (whether one thinks of the recent US Government's legislative proposals as merely "tweaking" the act or as a more drastic set of changes), paralleled by the elaborate and so far mostly stymied CONFU (Conference on Fair Use) negotiations.
What I have realized during the current licensing activities at Yale is that the license negotiations we engage in are doing legislation's business by other means. Instead of waiting on Congress and allowing terms to be dictated to both parties by law, we (publishers and institutions) are making our own peace, thoughtfully and responsibly, one step at a time. Crafting these agreements and relationships is altogether the most important achievement of the licensing environment.
What specifically do we negotiate when we negotiate a license?
Let me here note a personal irony: Where a few years ago, I was part of a movement among librarians calling attention to the boom in serial prices and agitating for restraint, I find myself spending hours working with publishers on every aspect of our prospective transactions and the greatest achievements do not, in fact, come about in the pricing area at all. Even as there can be a great deal of movement in important areas such as technology, delivery, and permitted uses, rarely is there noticeable give in the pricing model or actual price to be paid. If it is not in the pricing arena that our principal negotiating energies are spent, then where does the effort bear fruit?
The license answer crucial questions of the following sort:
These issues are central to our mutual anxieties, but they are only part of what belongs in a license/contract.
Some of the first licenses that producers offered to libraries -- and many libraries signed -- read like some of the leases that a big-city apartment owners hand a tenant to sign: most of the rights seem to be the owners' and most responsibilities the tenants', with disputes to be adjudicated by the owners' lawyer. As the licensing and signing authority for our library system -- and a librarian serving many diverse customers -- I must ask hard questions about the publisher's performance: his ability to assure us that an online resource will continue to be reliably available in a useful and consistent manner. When there is actual artifact -- a CD-ROM for example -- in hand on terms other than ownership, the license must be clear about when and how customers can or cannot keep what they have paid for. Scheduling upgrades and the disposition of old versions must be covered. Often the question of future payments comes into play. If the customer does not want, say, version 2.0, may he keep 1.0? What is the compensation or replacement policy for defective disks or server downtime?
Often there is a preliminary set of questions about how the product will be delivered to readers. What technology infrastructure is involved and what are both parties' expectations about it? Are they compatible? Questions of authentication may become relevant. Let me interject a small but true story here of the university whose library agreed to make a given body of scientific information available to those with institutional network ID. The permitted use terms of the product were excellent; the information in question was of high value to this particular faculty because of the faculty's strength in certain research areas. But one condition of that license was that there be no pass-through, by any means, of the information to the industry sector that might also find the resource to be useful and would be expected to pay a higher, non-education price for it. The university's strength in this area had naturally led to symbiotic working relationships (via contracts, sponsored research, collaborative projects) with companies in such related industries. In fact, the university's research and administrative offices may offer network access to some of the researchers and officers of this industry, to foster deeper cooperative relationships. Such bits of lagniappe can visibly threaten a license and the licensee therefore has to develop security mechanisms to define, layer, and control access within its community, explain to those who are excluded just why they are excluded, or negotiate (likely at a higher price) access for its research affiliates.
Authenticating access to any level of specificity is not a trivial pursuit, though increasingly as universities' finances, student records, and human resources files are moving online, there are multiple reasons for the considerable investment that sophisticated security systems require. The Library needs to be part of institutional discussions and track the implications of every modification and change. The negotiator must talk both to campus technologists and to the licensing producer about such matters and make sure that the required assurances can in fact be delivered. Other technological questions abound: how will, say, a reference work licensed on tape or CD-ROM bet networked on campus? What kind of training/documentation should be provided (by publisher or library)? What kind of support exists for users (an 800-number from the publisher for help with a proprietary software interface)? Can use be measured and how?
Copyright law gives the library, as owner of a physical artifact, the right to keep materials forever -- or as long as it chooses to do so. The volatile nature of electronic information makes archiving a profound -- and expensive -- concern and the language of licenses, which tends to remove access once payment has stopped, intensifies the fears that electronic information will be lost. May or should the library retain copies for the future? If not, what assurance will the publisher give that it will do so? Is the form stable enough that even if libraries have the right to indefinitely keep information, it will last so much as a decade? Libraries have been the aftermarket stock managers for generations of publishers. If they are to continue, they need to know the terms by which that can be done. If publishers resist, then libraries need to plan for the technological systems necessary so that information paid for in 1996 will be accessible in a few years. In spite of library efforts to secure archival permission and formats, for various reasons this is the part of the license that all parties address least well. Even where libraries secure for themselves a CD-ROM at the end of an online subscription period, or attempt to license content only (not interfaces), or require that content be made available in standard formats (such as ASCII or PDF) and standard tagging (such as html or sgml), we do not have a fully-developed sense of what we must to maintain and present that information ourselves.
Who will resolve any disputes that may arise? Where traditional library purchases under protection of copyright statute make it clear that violations or disputes are prosecuted through the court system, likely in the jurisdiction in which they occurred, the contractual relationship of a license allows the licensor -- and licensee -- to specify a process and a jurisdiction. This can be a touchy issue, particularly when a publisher is located in a different country or on a different continent. In such a situation, a licensor will naturally be inclined to suggest that the governing law be that of his or her own locale, while an American university library would see the prospect of being engaging lawyers or going to court on another continent as a forcing move. If a licensee must travel Luxembourg to defend himself, then he will incline to capitulate, his position in any prospective discussions significantly weakened. Resolution of this issue can often be a way to show good faith on both sides, good faith in the mutual hope that no such formal process of litigation will ever be needed.
Not infrequently, several products must be identified and several licenses negotiated in order to present one product and this compounds or triples the amount of licensing that is required. For example, our Oxford English Dictionary is content-licensed with Oxford University Press; the display and searching software is licensed from OpenText Corp.; and the software that optimally presents the complex characters such as diacritics, is licensed from the Humanities Text Initiative in the University of Michigan Library.
What does it take to negotiate licenses successfully?
1. Broad knowledge and understanding of both what the library is trying to achieve and of what the publisher requires.
This includes the routine business of knowing one's library users and collections and understanding the technology base of the institution. Beyond that, the presentations in this conference signal a new and vitally important area of expertise for both libraries and publishers. To help our colleagues in the library and publishing communities move forward with electronic content licensing, and under a grant from the Commission on Preservation and Access and the Counsel on Library Resources, Yale has been coordinating and preparing an online WWWeb primer which will be available sometime during late December and will be widely announced. Sequenced to flow like an actual electronic content license, it will define licensing terms, discuss their intent, offer examples, and point at pitfalls for the library-customer. Additional features include links to relevant Web resources, a bibliography and a licensing discussion list. Rodney Stenlake, Esq., a contract lawyer and Yale's consultant to this project, has been working with Georgia Harper on the resource. Both of them are here at this conference.
2. The normal virtues of a librarian: patience, good will, and a sincere desire to succeed.
Sometimes the process can be lengthy, sometimes discouraging. It can test the patience of either or both parties. Success, on the other hand, achieves the terms one needs for one's users and brings a real high, as does the knowledge both parties have gained and the sense of a mutual partnership and accomplishment that can be realized.
3. Authority is important.
Negotiating licenses is not simply a matter of knowing one's way around all the issues and knowing the people who have the requisite technical expertise. The negotiators for both the producer and the university need the authority to take their institution's good name to the table, make that deal, and be the "face" that remains accountable. The library negotiator needs to have the confidence of people inside and outside library: staff in collection development, public services, technical services, and systems. He or she needs the confidence of university's information technology personnel and legal office. That authority should be vested formally, either as position or specific individual. This last point requires some elaboration: Ultimately, any agreement a library makes must be acceptable to the overall institution. Just how that approval is secured will vary from institution to institution. Quite apart from the formal ratification by Counsel's office (which may not be necessary in some libraries), the negotiator will want to enter into discussions -- and the producer wants this as well -- knowing with reasonable confidence that what is negotiated will stick. We have had a very happy relationship with our attorneys in these matters, and I cannot urge strongly enough that librarians invest time to assure harmony in that relationship.
The negotiations we have been involved with over the past year have been time-consuming, sometimes exhausting (we take each phrase very seriously) -- and sometimes exhilarating. From time to time, I find myself with a box of software, ripping off the shrinkwrap, glancing at the fine print, and installing a new tool without much fret at all. The terms seem reasonable. When a market matures, when electronic content licenses become so standard and predictable that both the producer and the licensor are reasonably confident that the terms will work and when the shape of doing businesses becomes familiar, then we will know we have succeeded in making an important part of the infrastructure for this relatively new marketplace of electronic information. Already, some licenses presented to us are more realistic at the outset, and we move a little more readily to mutual resolution of differences on them.
Increasingly as I negotiate licenses I realize that I am working hard to adjust the language to give protection for the rights of both producers and users, and that what we achieve as permitted use in our most successful licenses resembles the rights our library users have for traditional materials under the Copyright Act. In negotiations, we frequently talk about "fair use" or its equivalent as one of our goals.
What I suspect will happen over time is that the concept of license as a focus of the electronic environment will itself will seem less and less relevant and necessary, that both publishers and users will see themselves adequately represented and protected by some combination of copyright statute and prevailing licensing terms. The paradox and the hope is that by working very hard create comprehensive and effective electronic content licenses today, we will render the exercise obsolete or automatic over time. To accomplish this worthy goal, we must make the very best licenses we can right now.
Visits to this page since May 2, 1999