Visits to this page since May 5, 1999
We live in an information world that demands all of us (users, authors, publishers, librarians, and many other players in the information chain) change traditional practices and interactions in many ways. One of the major changes comes about in the way information is owned. In the print information world, purchasers buy an object and own it outright, its re-use via copying governed by national copyright laws and overarching international intellectual property agreements. In this traditional world, concepts such as fair use or fair dealing allow copying of works in ways that advance education, study, and scholarship. For the world of electrons, society has not fully clarified the the implications of copyright law. Therefore, it is current practice for publishers or producers to lease or license information to customers, the use of that information then governed by contracts and contract law.
How these two constructs (copyright and contract) interact is not my topic here. Rather, the organizers have assigned me a simpler task: to share with you in the STM publishing community some academic library responses to the licenses we have been asked to sign, now that we librarians have a few years' experience as licensees of information from you. In my position at Yale University, it is my responsibility to sign contracts for the library system's electronic information and where those contracts are not appropriate, to negotiate ones that better meet our needs. We estimate that the Yale Library uses just under 200 electronic products and services ranging from diskettes to single-user CD-ROMS, networked CDs, tapes loaded and networked on campus, and electronic resources we access from a producer's site. At a guess, about 1/4 of these are significant resources, either because of high price or their utility for a large percentage of our campus users, or both. For each significant resource, a license has been signed, generally (because the initial licenses offered to us can hardly ever be signed without changes ranging from minor to significant), after negotiations between the library and the producer or seller. The same library concerns are addressed in contract after contract, and the same kinds of language are adjusted over and over. The good news is that it is rare that we cannot agree on language with the publishers or producers (who have been for the most part a pleasure to work with); the bad news is that such discussions take time, they cost money, and, what is worse, they delay implementation for students and staff.
My purpose here is to discuss with you the parts of a license that most frequently delay library negotiations and to propose that it is time for us, as libraries and STM publishers, to discuss license terms, find areas we can agree upon, and propose at least some standard types of language for at least some parts of e-journal and database licenses. We probably will not agree on everything, and there are some contract terms or sections that may never become standard, but if we can address our differences now and resolve them, we can use the saved time to get on with the rest of our busy agendas.
In optimism, I note that while it is apparently impossible, at this time, for libraries and producers to agree on the meaning of fair use in the electronic environment (a conceptual and theoretical debate of surpassing importance in which 70 U.S. CONFU discussants have made little headway over two years' time), we do have great success together in negotiating workable, effective electronic information licenses. It is in the practical arena of licensing that we are together overcoming the obstacles that we cannot surmount in our discussions about electronic copyrights.
Electronic content licenses range from short (one page or less) to long and complex (15 - 20 pages) and being longer does not necessarily make the license more effective. For the academic library customer, ordering an electronic resource bears no resemblance to the "traditional" Acquisitions mode (send an order to acquisitions to be keyed; mail out purchase order; order is received; and sent through normal processing routines). Rather, a request for electronic information represents a library partnership that includes -- and regularly involves joint consultation between -- collections development officers, public services staff (who work directly with users to assess needs and functionality of the electronic resource), systems/automation staff (to resolve computer and network requirements), and legal authorities or their designates, to be sure that licensing language meets the institution's requirements. Externally, the academic library is involved in several partnerships to make any license happen: with the publisher or producer, the vendor, other library institutions (if a consortial purchase), and not infrequently with the creators of software that runs and displays the electronic resource (which, in turn, can involve the academic library in multiple licenses to utilize any one product).
From the Yale Library's point of view -- assuming the existence of a desirable electronic resource for our users -- five general components of a license need to be "right" before it can be signed. If any one is not suitable, the license cannot be signed. These five components are:
1. Definition of use and users must suit the academic and library environment, so that members of the licensing institution can pursue their usual academic activities.
2. Legal liabilities and responsibilities must be reasonable and balanced.
3. Producer's and library's understanding and functions of technology must be in synch.
4. Business model has to be understandable (and affordable).
5. Archiving (and perpetual access) need to be addressed.
In an educational license, the definition of user must include all those that will use the resource. For example, in an e-journal license, we seek coverage for students, faculty, staff, visiting researchers, and others working on university projects. Some institutions have policies that make it desirable to include other categories of users as well. For example, Yale allows retirees to keep their Net IDs for six months; emeritus professors probably keep NET IDs as long as they wish to. We also expet the authorized user group to have access from any computer where they are working. In addition to the campus's authorized users, academic libraries are available to members of the local community and it is not uncommon for, say, high school students to visit the Yale Library to research essays or projects. Just as we permit our visitors to use print resources, so we expect that they shall be able to use thee- resources on our computers. Thus, a desirable academic library license includes walk-ins to the campus libraries.
Normal university purposes and functions should be permitted. For example, we ask that authorized users be able to "copy" electronic as well as print resources: that is, download onto their computers; store the downloaded material for the duration of research or study; display and print materials; and quote or cite in essays or publications -- in the course of university teaching, research, scholarship, study. Users should be able to share materials with colleagues in other institutions for purposes of normal scholarly communications, not systematic transmission or transmission for sale or re-sale. Additionally, we ask that the electronic work be available for other normal academic library uses: electronic reserves and Inter-Library Loan. ILL is perhaps the most controversial provision libraries request in electronic content licenses. In the United States, the Copyright Act is surrounded by voluntary guidelines, many from a consensus process of 20 years ago called CONTU. One of CONTU's provisions was a so-called "suggestion of five." It encourages libraries that copy more than five articles from any journal title of the last five years to either seek permission or pay royalties to the publisher of that journal. As we see it, this guideline can be moved into the electronic license, giving libraries some elbow room to share works and at the same time defining a point at which additional fees could be paid.
I offer two examples of unacceptable use language in electronic contracts that Yale Library subsequently re-negotiated:
A. "No part of the information may be reproduced in hard copy, machine-readable form, or other form without advance written permission from XXX."
This was changed to: "No part of the information may be reproduced in hard copy, machine-readable or other form without advance written permission from XXX, except for legitimate fair and personal use such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research. [NB: Language from Section 107 of the US Copyright Act]
B. "Under no circumstances may a Permitted User store any record or records electronically for the purposes of resale, redistribution, or performing ongoing searching, except that a Permitted User may download records for temporary storage and internal use. The Permitted User shall erase the records once the temporary use has ended."
Amended by letter: "We have no means to tell the Permitted User to be sure to erase every record once the temporary use has ended. Additionally, there is not and cannot be any precise definition of what temporary means. For long-term scholarly projects, temporary can span a career. Therefore, we ask that you remove this from our agreement."
Some examples of desirable provisions for academic libraries include:
A. Notice of termination obligations need to be balanced and even-handed for producers and customers. For example, some licenses ask that the Library give 6 months to a year's notice of cancellation, while allowing the provider to terminate (with cause) on comparatively short notice.
B. Performance standards need to be specific (beyond, "provider will do his best to offer the resource reliably") and include compensation for failure to perform for certain times.
C. Reasonable locations must be offered for dispute resolution or arbitration. US institutions cannot easily go to arbitration in foreign countries. Some state institutions require that arbitrations happen within the state.
D. Confidentiality is a concern. Licenses usually require that all terms remain confidential between the two parties. In fact, while some terms (such as a special price arrangement) might remain confidential, most should be standard, well-understood matters. All of the library staff and authorities in a college or university -- and perhaps even the users -- need to understand the terms that bind them. Librarians are accustomed to sharing with each other information about terms for the materials they purchase.
E. The biggest issue is library or institutional responsibility and liability. Producers often expect a library or university to be responsible for the behavior of any and all users and state that an individual's infringement may be cause for cancelling the license. It is simply impossible for an institution to guarantee the behavior of a user. A reasonable license will ask that where violations of the contract occur, the library and licensor shall work together to remove the problem.
The licensor and the Library have to agree about technology expectations and the licensor has to request terms that the Library can meet. Some stumbling blocks have been:
A. Security: the provider may expect a greater degree of security than the institution can provide. For example, the licensor may desire the institution to hand out passwords to authorized users only. In general, it is not workable for institutions to hand out passwords for particular resources; the size of the user population is often too large to do this on an individual basis, and posting the password near the machine would be counterproductive to the publisher's concerns. Our preference for widely-used campus information resources is that the producer validate all the University's IP addresses and permit access from those addresses; or that we are able to use our own existing systems for validation.
B. Technically impossible requests: As an example, I offer the following language: "Yale shall faithfully reproduce or if omitted add the copyright symbol and clause of XXX producer to any printed or downloaded part of the e-databse as follows:...." Unless the producer has embedded the copyright symbol in the online version, this requirement is not realistic.
C. Vague and imprecise language: For example: "Yale will ensure that its items of equipment or software which it proposes to use to access the information shall be capable of communicating with the system to enable the full functionality of the information for its IP users." The intent of this clause may be that the provider not be responsible for creating communications technologies for for the institution. But effectively it requests that the host institution provide unspecified technologies to make the information display in some unspecified way.
It is vital that the business model be understood by the institution that is licensing the information. Pricing models are already as or even more diverse than what is charged for print and include, among others:
A. Price of print plus add on surcharge for the e-version.
B. Price for e and add-on surcharge for the p-version.
C. Print + e for the same price as print alone.
D. Charge per FTE student on campus.
E. Price related to size of library acquisitions budget.
F. Price related to number of simultaneous users.
G. Document delivery (per article or section price).
H. No relationship to any print product (often the case for de novo e-resources that are not directly related to a specific print resource).
I. Capital cost for buying into the resource plus a small annual access fee.
J. Sliding scales of various sorts (for example, the more users, the cheaper the per-user price).
K. Discounts, particularly for multiple institutions (consortia).
L. Certain categories of users (such as students) get a cheaper price.
Some of the pricing schemes, particularly where more than one variable is involved (say that the product appears in e and p, that the cost will be based on the volume of use and a consortial reduction) can be complicated to understand. The contract must express these significant matters clearly, possibly as an attachment with calculations for the license price over its term.
(An important footnote, though not directly related to our discussion here: The air is redolent with talk of the additional income that information providers can obtain from electronic media, given their flexibility, ubiquitousness, searchability, and so on. In academiia per se, it is hard to imagine where that extra income could materialize, though to be sure some states in the US are, for the moment, providing supplemental funding for heavily used resources to be deployed for a broad range of users such as public, or K-12, or community college libraries. It is not clear these subsidies will persist; and more to the point, most institutions (none in our region of the US) do not have access to such funds. Therefore, should we choose to pay one producer additional fees (higher than current print prices) for his electronic product, we in turn forgo other current acquisitions. This is not in the best interests of scholarship or publishing overall. Libraries are taking some decisions to buy e at higher than p-costs in order to experiment with significant e-resources, but this behavior can be sustained for a short interim period only.
Licenses must address the archiving and access-in-perpetuity issues. It is not thinkable that a library would invest significant funds in resources whose existence it cannot assure well beyond the licensing period. Research and scholarship in many, many disciplines requires access to the discoveries of the past. Any license should either offer a perpetual access to the years being licensed for as long as the institution licenses them (at no to modest cost) or provide the institution with an archival format. CD-ROMS are in general not that preferred format for Yale. Print or tapes that can be regionally or locally networked might be a good solution at this time for us.
Various information intermediaries have offered us their services as negotiators for our licenses. At the moment, this is not a workable offer. All of us are early in the process of developing an understanding of what we need in access and licensing agreements, and librarians' direct experience with licenses before this important activity can be turned over to an intermediary party. Furthermore, vendors who are also themselves information providers find their interests can conflict with those of customers.
In summary, licensing is relatively new; it takes time; and often the same terms over and over (outlined briefly above) are stumbling blocks between information providers and their customers. Thus, negotiations can take time, be expensive for both parties, and delay implementation for the users. At the same time, it is rare for us to be unable to reach agreement with information providers about licensing terms. Information suppliers have been, in most cases, excellent and reasonable, desiring agreement as much as the customers do. I believe we have come to a point in time when we are ready to form a small working group or task force between academic library customers (at least in the US) and a group of publishers to explore the terms of electronic licenses and identify those that we can agree upon and standardize.
Such a working group would be one highly positive way in which STM, through its Library Relations Committee, could work on matters that are immensely susceptible to being sorted out. Also, such a working group would dovetail nicely with an American library licensing initiative. Under the sponsorship/funding of the Commission on Preservation and Access and the Council on Library Resources, the Yale Library is acting as Principal Investigator for a World Wide Web licensing resource for librarians. That is, we are developing an online primer describing e-license terms and signalling to you, our producers, what contract language is acceptable to us. This WWW resource should be available on Yale Library's server by the end of 1996, along with analysis, links to other licensing resources, and potentially a licensing discussion forum. Upon my return, I will propose a meeting to begin work on our mutual task. There is now a solid basis of fine contracts on which to base such an activity and I feel confident that some success could be ours before we meet here next year in Frankfurt.
--In addition to being a member of the STM Library Relations Committee and Associate University Librarian (Collections) at Yale University, Ms. Okerson is also the coordinator of a recently organized consortium of large NorthEast Research Libraries (NERL) whose objectives are: (1) to jointly license substantial electronic resources such as full text journals, databases, and large literary works at advantageous terms and rates; and (2) where access is not readily available from the publisher or supplier, to consider joint deployment of those resources. She is a former senior program officer for the Association of Research Libraries in Washington, DC, in the role of Director of the Office of Scientific & Academic Publishing.
Visits to this page since May 5, 1999