Abstract
As publishers and vendors offer more electronic resources, the number of licenses that have to be negotiated and executed increase proportionally. Libraries and publishers often have conflicting goals, making the successful outcome of a license negotiation difficult to obtain. A small group of licensing professionals gathered for a frank, open discussion of the terms and conditions that are important to each side, thereby opening the channels of communication on a larger scale. The discussion panel consisted of three academic librarians and two publisher representatives; the moderator is an academic librarian and a member of the Shared Electronic Resource Understanding (SERU) Working Group, as is one of the panelists. The goal of the session was for each side to come away with a greater understanding of the other's positions on key license points, as well as some strategies for arriving at compromises. The mission of SERU was discussed and promoted, creating more awareness on both sides of the table of the existence and usefulness of SERU as an alternative to the cumbersome license negotiation process.
INTRODUCTION
The process of negotiating license agreements between libraries and publishers appears to be fraught with difficulties, especially to those librarians or publishers unfamiliar with the wants and needs of the other party. Although individual publisher representatives and librarians may occasionally exchange viewpoints on licensing outside of the negotiation process, in general there has been little public discussion of what each party wants or requires from a bilateral negotiation. This lack of discussion may lead to mutual misunderstandings, whereas a frank and open exchange of thoughts could illuminate what each side finds most important in a transaction that requires a license agreement and lead to greater understanding and agreement on key issues. The three librarians (Selden Durgom Lamoureux, Lisa Sibert, and Micheline Westfall) and two publisher representatives (Vida Damijonaitis and Brett Rubinstein) who comprised the panel for this program agreed to sit down together to discuss the license provisions of most importance to both parties. As questions were posed by the moderator (Clint Chamberlain), the panelists were free to respond in any order they chose as well as to respond to each other.
PANEL DISCUSSION
Question: Have you noticed that licensing, as a process, is improving? How long is the process, from the time you receive the license to the time you complete it? Is it becoming more streamlined? Frustratingly static? Some improvement but not enough? What is your sense of trajectory for this process? Is it getting easier?
Rubenstein: I have been licensing content to various types of libraries for four years, and generally, no, I do not think that the licensing process is improving. In cases where we have worked on numerous licenses with a specific institution, we have certainly imported agreed-upon clauses from prior licenses, but this is not different than other efficiencies gained by ongoing relationships. The time it takes to complete a license varies from a same day signature (the exception!) to as long as nine months. I would say the average process takes eight weeks, though I once worked on a license for a corporate library which took eleven months.
Westfall: Some days it seems that the licensing process is improving (i.e., for those publishers who are aware that the governing law and other sections need to be modified in order to comply with state laws). Other days, it seems frustratingly static as we struggle with defining terms such as “site,” “authorized users,” “settlement of disputes,” “taxes imposed,” etc. The main difference between these two processes is often dependent on the size of the publisher. Large publishers that have worked through the process with libraries are more willing or able to comply with libraries' needs, whereas small publishers and associations appear to be working through the front end of this process.
For us, the licensing process has not become streamlined. Our contracts and legal office are well-educated about the difference between licenses for electronic products versus contracts for widgets. However, I believe they depend on us to ensure that the terms crucial to the library meet our needs, while we depend on them to ensure that the terms crucial to the state's laws are met.
Lamoureux: My own sense is that it is not. While libraries do have more experience now, the same stumbling blocks exist: a long and cumbersome back and forth between librarian and publisher over what are, often, a set of issues that are critical for both sides.
Sibert: First, a disclaimer: I have been involved in licensing electronic resources for only about eighteen months. I have had some licenses take as long as ten months, while others take only a matter of days. It is difficult for me to know if the license process has improved over the last few years, but in my own experience it does not seem to be improving. Allow me to qualify that. For large publishers, the process is fairly smooth. They are often familiar with the needs of a University of California (UC) campus, and can anticipate the changes that I will ask to be made. This is appreciated; it makes the negotiation process relatively quick and painless. The smaller publishers, however, are a different story. Often the amount of time and energy that goes into negotiating a license with a small publisher is disproportionate to the cost of the product we are licensing. The smaller the publisher and the product, the more money it takes to license. I find this to be extremely frustrating, and it is an area where I hope Shared Electronic Resource Understanding (SERU) can help.
Damijonaitis: The length of time to process a license can vary greatly. A license that is returned to us signed, with no requested changes, may be completely executed within two weeks. On the other hand, some licenses have taken well over six months to complete. While negotiations may be time consuming in some cases, often the time delay is due to the length of time it takes for either side to respond.
As an organization, we are constantly evaluating both our internal licensing procedures, as well as looking for ways we can improve the license itself. Despite this, the licensing process remains cumbersome. Part of the problem is the increasing number of licenses that need to be processed each year. The sheer number of licenses are overwhelming for both libraries and publishers.
Question: What are the risks you want to avoid/mitigate by using a license, and consequently, what are the most important parts/aspects of a license to your organization?
Rubenstein: I would say the primary risks for publishers are a reference to copyright law and protection against commercial redistribution. There are, however, ancillary benefits to the general practice of licensing which I think are important: (1) as a commercial entity, licenses are important for valuing the business and forecasting revenues; (2) in a fairly mobile industry, where both librarians and publisher representatives often change positions, well-written licenses serve as historical documents to explain the history of the relationship to newly introduced stakeholders; and (3) in case of a breach of terms, licenses can serve as a “resolution map.”
Westfall: We do not (ever) want to lose access for our users, nor put our institution in jeopardy (for indemnification, for example). The most important parts of the license are authorized sites not being limited to “one contiguous set of buildings”; authorized users must allow walk-ins; access terms via IP address (instead of limited to username/password), so that our distance education students can access the resources; allowing for course reserves, interlibrary loan (ILL), and Blackboard use; governing laws, taxes to be imposed, settlement of disputes, etc., must adhere to the laws of the state of Tennessee; no indemnification by our university or state; and allowing for post-cancellation rights and/or perpetual access.
Lamoureux: As a librarian, it seems to me that licenses were primarily written by publishers to protect their business interests. I would be happy to revert back to copyright law, which guided our use of print, and let the business terms be discussed outside of a license (in a purchase order, for example). Having said that, when a license is required, I want to be certain it does not conflict with institutional policies, practices, and our ability to comply. I do not want to make promises that I have no power to keep, such as promising the library will indemnify the publisher (something I am prohibited by state law from doing), or promising to see that users never breach any terms (something that the library cannot control). I also do not want a license that exposes my institution to unreasonable expense (if we have a license, it must include some statement that the publisher is not infringing on third-party rights; we do not want to be responsible for the cost of litigating infringement claims). On the side of what I want (as opposed to what I want to avoid), I want to ensure the use terms make sense in the university and library context (for instance, the ability to place a resource on electronic reserve, the ability to ILL appropriately, and the ability to use for scholarly purposes and teaching).
Sibert: As a UC campus, we are bound to include certain terms in every license we execute. The most important is indemnification. We have to be indemnified against third-party claims of copyright or any other property right infringement, and we also have to be indemnified against alleged breaches by the other party. Another is governing law; we cannot accept governing law in any state other than California. These are both deal breakers. We also find it extremely important to include specific use permissions in our licenses, to avoid ambiguity when possible. The most important use permissions for us are scholarly sharing, ILL, walk-in user access, perpetual access, and archival rights. If the publisher wants to remain silent on the use permissions and include a statement about fair use, this will sometimes suffice, depending on the product. ILL can be important enough, depending on the product, to be a deal breaker if the publisher expressly prohibits it.
Damijonaitis: In addition to reiterating copyright, the license serves two main purposes. First, it defines the permitted uses of the content and specifies the scope of users that the institution has agreed will be allowed to use the content. Second, it defines the cure for breach of the terms.
Question: Has the license proved to be a good way to address these risks? Does it make the resolution of breach easier? Or, given the long history of cordial relationships between library and publisher, could breach resolution be resolved even without a license?
Rubenstein: My experience with copyright infringement is extremely limited. In cases where Springer discovers pirated content (often due to not employing digital rights management on our platform) we have certainly taken legal action against websites or other points of distribution. We have not ever linked those activities to our customers or had to take similar action with a library due to an egregious violation of our terms.
While I do think that librarian–publisher relationships are cooperative and in almost all cases cordial, I do not think it is responsible or practical to assume these characteristics will persist, or that they can serve as a replacement for the reassurances present in a negotiated agreement.
People acting reasonably or logically is certainly appreciated and expected, but planning on it is most certainly a mistake.
Westfall: From the university's point of view, the license probably provides a sound “legal” way to address these risks (especially those parts that are governed by the Tennessee Code of Laws). Personally, I think that the history of the cordial relationship between publishers and libraries is usually sufficient to resolve most breaches. It is unlikely that a publisher would sue a university or vice versa. When we are dissatisfied, we generally do not renew the service. Further, the practice of the academic community in general (led by the Scholarly Publishing and Academic Resources Coalition (SPARC), Create Change, and librarians), as well as by federal legislation (such as the National Institutes of Health Public Access Policy), are likely to be more productive responses.
Lamoureux: Not in my view. The things I want to ensure (ILL, course reserves, scholarly sharing, and use for research and teaching) are already covered in copyright law. The things I am anxious to avoid (indemnifying the publisher and exposure to risk of infringement) are not issues if we do not have a license. I think publishers and libraries have a long history of working in a non-litigious environment. The libraries I am associated with would, even without a license, work to resolve a breach. If the collegial history between library and publisher is not sufficient, there is always the fact that publishers control access.
Sibert: It depends on the nature of the product and the amount of liability that could be associated with the use of the product. For instance, medical and business databases are more vulnerable to misuse than some others, so we might prefer to address risk, liability, indemnification, use permissions and restrictions, etc., in a formal, negotiated license. I think if you are going to go to all the trouble of using a license, it is best to include breach cure language for both parties to ensure that the responsibility to address the breach is clear so you can work together to bring about a resolution. For single journal titles from small publishers, however, it may not be worth the time and effort (which equate to money) of engaging in a lengthy discussion about use permissions and restrictions and breach cures if each side can be mollified by invoking SERU.
Damijonaitis: While there are other ways to define the permitted use of content, a license is the best format to define the cure for breach. Licenses, like all contracts, are developed to protect an organization. Whether the relationship is between library and publisher, or any other entity, the large majority of business relationships never have a problem. Licenses or contracts are there to address the small number of situations where issues cannot be resolved through other means.
Question: For libraries: What is reassuring to you about having a license? Is the license the best vehicle for establishing that?
Westfall: The only reassurance is that this is a type of safety net, should a publisher become too limiting or not concerned about the library's commitment to its users. I think that most public institutions consider licenses as the best vehicle. However, I think that new approaches, such as SERU, could provide a modified, improved model. After all, we did not need licenses for books and journals in the print environment. If we can determine ways to get electronic content on library-owned (or library-trusted) servers, licenses would be of no benefit to the library.
Sibert: Libraries have less control over electronic content than we do with print, and the possibility is always there that access could be taken away, for valid or invalid reasons. A license acts to ensure that access is continuous, which is really fundamentally the assurance a library needs.
Question: What tend to be the most difficult points to negotiate? For libraries: Are there certain issues that are non-negotiable? For publishers: Do negotiations with librarians ever help shape the content of future iterations of licenses?
Rubenstein: Absolutely. In cases where there are ongoing and active relationships, we certainly save time by importing agreed-upon negotiated terms from prior agreements. In addition, when change requests become common, and if these changes are something that we have decided to approve, we will simply update our standard licenses. I think a good example of this is allowing for walk-in users as part of the authorized user base, and it is something that Springer is beginning to examine with respect to alumni access. In both cases, and I would assume that this is common across content providers, we are charged with weighing the costs and benefits to our customers, end users, and authors against the risk to our shareholders.
Westfall: Non-negotiable issues include liability, indemnification, governing law, settlement of disputes, and imposing taxes. If these issues are not able to meet the laws of the state of Tennessee, then we are not able to sign the contract (or buy the resource).
Sibert: We also cannot yield when it comes to indemnification and governing law. These are deal breakers. As a publicly funded institution, it is extremely important that walk-in users be included in the language defining authorized users. This can be a deal breaker, depending on the product. ILL and scholarly sharing are extremely important, as well. These are perhaps the two most difficult points to negotiate. It is difficult to convey to some publishers the importance of ILL to a public institution and scholarly sharing to the spirit of research.
Damijonaitis: Jurisdiction, indemnification, and arbitration are the most common issues and clauses that we are asked to adjust. While we have never had to walk away from a relationship due to issues in the licenses negotiations, there are certain requests to which we will not agree. We will not allow an institution to host any of our content, consultants must have access to an institution's intranet in order to have access to our content, and licensed content may not be used for any commercial purposes. In the end, we are constantly working on revising and simplifying our license. Over time, negotiations with librarians influence the language that we utilize in new versions.
Question: Do you need to go through the same signatory process with purchase orders as with licenses?
Westfall: We do if the purchase order includes terms and conditions.
Lamoureux: Not in my institution. The Acquisitions Department has the ability to sign purchase orders, whereas no one in the library has authority to sign a license.
Sibert: It depends on whether it is a true purchase order or not. If there is no reference whatsoever to a license or any other terms and conditions, then I can usually sign it. If there is reference to license or terms, it has to go through our usual license signatory process.
Damijonaitis: We currently do not have any kind of signatory process for purchase orders. However, because of the size of certain licenses, this may change if a decision were ever made to replace licenses with purchase orders.
Question: Would a return to copyright law combined with a purchase order that describes business terms feel sufficient to you? Would it carry the same weight as a license?
Rubenstein: I would say that the combination of a purchase order and copyright law could provide protection against the main risks we address with licensing, but this would also remove many of the details in a license that could be contested. In the case where a publisher changes their policy on, say, authorized users, when would those changes go into effect? If this is contested and no resolution is achieved, does the library have any recourse?
With regards to SERU specifically, Springer has begun employing the SERU plus purchase order process and it is something that we will continue to offer and assess. We certainly feel it is to everyone's advantage to reduce the burden of licensing where both Springer and our customers feel comfortable, but we have to always be mindful of any potential risk.
Westfall: In my opinion, it would be sufficient. However, I doubt that our university legal office or personnel would agree with me.
Lamoureux: Yes, it would be sufficient. We have used copyright law for those electronic resources that do not have a license, and we have also used SERU with publishers who have supported it. There has not been any negative impact in either of those instances, and the entire acquisition process is super fast.
Sibert: This is a difficult question to answer. On the one hand, I want to say yes, that would be preferable, because it would ease the burden currently placed on libraries when acquiring electronic products. However, as a representative of a UC campus, I cannot say this would be appropriate in every situation. There are certain things that are really important to the UC campuses that can be deal breakers when negotiating a license, and they may be important enough that they need to be documented, especially for the larger purchases. Indemnification, use permissions like ILL and scholarly sharing, and perpetual access and archival rights are a few. However, it is safe to say that some combination of a purchase order and invoking SERU would be preferable to negotiating a license for an inexpensive product from a small publisher.
Damijonaitis: This is an option that we are currently investigating. While we do not feel that a purchase order offers the same protection for us as a publisher, or for the library as a customer, we need to decide if the organization is comfortable assuming the risk.
Question: For libraries: What do you do when the publisher does not have a license?
Westfall: We document this in our license spreadsheet and resource file folder. If the publisher does not require a license, we do not either. We do ensure there is no reference to a license on the invoice by ensuring the invoice does not include any wording, such as “per … terms and conditions,” “license,” “agreement,” or “contract.”
Lamoureux: We document the fact in a paper and online file, and proceed with purchase. It is fast and easy. (The advantage of SERU over simply using copyright law is that SERU outlines best practice, describing expected behavior on the part of publisher and library, that supplements the print-centric copyright laws.)
Sibert: Cheer and document. If the product price is small and we are not concerned about great liability, that is. In these cases we are happy when there is no license. It allows the time from order to activation to be minimized, which means I can turn on a product for my users much quicker. It is all about access and if foregoing a license gets us to that end quickly, it is a good thing. These are the situations where, if there ever were any suspicion of improper use, we would rely on copyright law. We are trying to invoke SERU where appropriate in order to at least have a good faith understanding among the parties about expectations.
For high-priced products, it is a slightly different story. It is a matter of weighing our liability against the amount of documentation required to protect our rights to use the product. In some cases, we will request a license even if the publisher does not require one, and we will go in with our license template to start the negotiation.
Question: Will a one-size-fits-all procedure ever work?
Rubenstein: I doubt it. In my opinion, the larger problem with a one-size approach is that it will invite a one-size mind-set. Nearly all of the licenses we work on are customized. In many cases the pricing is the result of a variety of factors: standard pricing, historical spend, consortia discounts, etc. In other cases the terms are specific to a library or group of libraries, as mandated by their administration or state government. I do not think that the differences across agreements can be accounted for with a one-size approach, as many of these terms are essential. This is, of course, just my assumption, but in general I am extremely wary of anything that invites less flexibility or open-mindedness from either publishers or libraries.
Westfall: I seriously doubt that one size will fit all procedures. However, if publishers are willing to agree to it, we would be willing to give it a try.
Lamoureux: Probably not, although SERU has been more adaptable than I initially imagined. We have been piloting the use of SERU for multi-year deals and e-books; that is something I had not anticipated. In the case of multi-year subscription periods or any transaction that has unusual business terms, it is important that the purchase order be clear about expectations.
Sibert: Probably not. Again, it is a matter of weighing the costs associated with the license negotiation process against the cost of the product, and also considering the amount of liability we could incur with use of the product. There is no single solution that will be right for every product, for every purchase or subscription. SERU gives us another option, though, and I am really happy about that. I am trying very hard to invoke SERU whenever and wherever it is appropriate. So far, I have not been very successful. I have had one publisher agree to SERU in lieu of a license. Other publishers have been less responsive. I am not sure if it is that they are not aware of SERU and cannot take the time to learn about it, or if there really is some resistance to it. I am going to keep trying, though. It is an important initiative.
Damijonaitis: I do not think there will ever be a one-size-fits-all procedure that will work for all parties involved, but I do think that over time publishers and libraries will be able to come to resolutions that will resolve difficulties and save staff time on both sides. Even if we only improve the process by 50 to 60 percent, that would have a huge impact on the resources of all parties.
Question: ONline Information eXchange for Publications Licenses (ONIX-PL): Any familiarity with it? ONIX-PL standardizes the elements of licenses and allows libraries and publishers to identify whether or not a given element is present, and where in the license it occurs. It makes it possible to automatically load/map a license into electronic resource management (ERM) systems. Would having this ability be a plus? How can you envision working with an ONIX-PL-ified license?
Rubenstein: ONIX-PL grew out of an effort to allow publishers to express their license terms in a standard format, with consideration toward the data requirements laid out by the Digital Library Federation (DLF) Electronic Resources Management Initiative (ERMI); creating one standard machine-readable format that could load license terms to ERM systems owned by libraries. ONIX-PL includes not just the specifications, schema, and dictionary of elements for expressing many different kinds of license terms in XML format, but also an XML editing tool called OPLE, specifically designed to help publishers edit their XML licenses. Nothing about ONIX-PL limits a publisher as to what may be expressed in their licenses. An effort was made to test the schema with a number of different publisher licenses. The end result of the ONIX-PL process is a license that a library can load to their ERM system, which may be tied to various public displays to end users to indicate what kinds of uses of the licensed electronic content are permitted.
It has been suggested that libraries could load an ONIX-PL license at the start of license negotiations to get a good view on what would be permitted, and then make adjustments on the library system itself if different terms were eventually reached. ONIX-PL is not intended to replace a printed license, but instead is intended to allow efficient loading of publisher terms for display to both library personnel and end users.
Springer supports the ONIX-PL program and has had its basic license converted to ONIX-PL format already. Interestingly, the SERU best practices that Springer has used in place of licenses for some clients have also been converted to ONIX-PL form, and could be loaded to library systems.
Sibert: We are in the midst of implementing our ERM system, so I am very excited about ONIX-PL. Anything that I can use to streamline the process of license analysis would be great. I envision ONIX-PL allowing me to delegate some of the pre-negotiation license analysis to my team members.
Lamoureux: I love the idea of ONIX-PL (although I confess it took me a while to realize this is not a pre-analyzed license, just one that has its components clearly identified). Our ERM systems would still need to create the analysis tools, but with a standardized vocabulary, communication about the license is automatically enhanced. I can see this helping to illuminate what is in any given license, make comparisons between licenses, and pull reports about licenses. It would be extremely useful in reviewing and revising a license. It might also lead publishers to greater standardization of what is covered in a license, as well as, perhaps, the language used to describe terms. If a publisher would transfer their license into ONIX-PL, it could be automatically loaded into my ERM system, where I could review, revise, and store it. I could see subscription agents playing a role in disseminating ONIX-PL licenses as well, since they would be easy to store and transmit. The hard work would fall to publishers, but once done, it would be good for years.
Sibert: If we had new licenses parsed into ONIX-PL, we could then distribute the work of reviewing and revising a license to specialists (perhaps an indemnification specialist, a governing law/jurisdiction specialist, etc.). Distributing the work in electronic resource management would be a huge plus.
Damijonaitis: We are just beginning to familiarize ourselves with ONIX-PL.
Question: Are licenses doing the job you want them to do?
Westfall: Licenses are a huge burden for libraries. I am not sure that licenses are necessary in order to ensure that the needs of both publishers and libraries are met. Over the years, publishers and librarians have been discussing their respective needs and for the most part have reached a mutual consensus or understanding. Often it seems as if licenses are only a bureaucratic necessity.
Sibert: For large purchases and big-ticket items, licenses are doing their job. They are protecting our institution and providing guidelines for acceptable use. For small, inexpensive items like single journal titles from a small publisher, they are not serving a valid purpose. Other methods like SERU, copyright law and/or purchase orders would be preferable to the time and expense involved in negotiating licenses for these small products.
Damijonaitis: While licenses have served their purposes, considering the minimal risk involved, I think there are alternative options to licenses that would work for both publishers and librarians.
CONCLUSION
Librarians and publishers may have differing goals when engaged in license negotiation, but during a reasoned discussion such as this one, it becomes apparent that the two parties are not as dissimilar as one might suppose at first glance. Although each party looks to protect its own interests, it is also clear that librarians and publishers generally operate in a relationship of trust and mutual goodwill. Each side of the negotiating table was represented in this discussion, thereby providing both participants and attendees with a greater understanding of each side's positions on key license points. SERU was also discussed as an alternative to licensing. Ultimately the panelists reached the conclusion together that although we may not be able to discard licenses completely in all transactions, there is no reason for the negotiation process to be adversarial, especially not when a frank discussion of each party's needs and goals reveals that we do share many common understandings.
ADDITIONAL RESOURCES
Digital Library Federation. “DLF Electronic Resource Management Initiative.” http://www.diglib.org/standards/dlf-erm02.htm (accessed July 24, 2009). | |||||
EDItEUR. “ONIX.” http://www.editeur.org/onix_licensing.html (accessed July 24, 2009). | |||||
National Information Standards Organization. “Shared E-Resource Understanding (SERU).” http://www.niso.org/workrooms/seru (accessed July 24, 2009). | |||||