INTRODUCTORY REMARKS

At Elsevier Science, we welcome the recent initiative taken by the informal group known as ICOLC (International Coalition of Library Consortia). ICOLC issued a "Statement on Current Perspective and Preferred Practices for the Selection and Purchase of Electronic Information". This can be found at http://www.library.yale.edu/consortia/statement.html. We look forward to continued discussion with our library colleagues who worked on the ICOLC statement.

In that context, we would like to note the earlier statement of licensing principles issued by a group of Dutch and German librarians. That statement lead to a series of constructive discussions in The Netherlands and the likelihood of agreement on new experimentation. We hope the ICOLC discussions will have a similar positive note.

Indeed, journal publishers are pleased that librarians are actively engaging in the negotiation process. All involved recognize that we are in a period of transition in which publishers are providing more services and information resources than the mere analog of the print journal. Libraries will act in the end as consumers in a free market and reward (by purchasing) those services that are perceived by them as providing true value for the library users.

In such a period it is to be expected that there will be many experiments as to authorized uses and fees. Such "competition" should be viewed as healthy and as leading to optimal solutions for users. Indeed, publishers, because of the nature of anti-trust and competition laws, cannot collectively decide on a single approach or set of policies for the library market.

Publishers view contracts and licenses as a robust and clearer method of resolving what might otherwise be ambiguities as to permitted uses and limitations. As we work in an increasingly international environment, it is difficult to rely on national statutes for clarity.

Publishers also want efficiency in the administration of contracts as much as libraries do. Therefore, publishers will often try to minimize non- substantive negotiations, particularly on boilerplate items, and put the energy into fundamental points. That does not mean, however, that publishers should be inflexible in meeting legitimate needs.

SPECIFIC ICOLC COMMENTS
- Looking at the ICOLC statement, Elsevier Science have the following comments on the preferred practices requested by the statement's authors.

PUBLISHERS SHOULD BE WILLING TO DEAL WITH CONSORTIA.
- We agree. We are negotiating licenses with consortia for both our local (Elsevier Electronic Subscriptions) and remote (ScienceDirect^TM) electronic services. Both libraries and publishers can benefit from consortia licensing. While the benefits of consortia arrangements are most evident with locally-held files (where the infrastructure cost is shared by several institutions), it is also possible to construct win-win agreements for licensing of remote access.

STATE ALL TERMS IN THE CONTRACT AND HAVE ALL TERMS NEGOTIABLE.
- Certainly all terms should be stated and, in general, all terms should be open for discussion. Often "boilerplate" is far more easily adjusted to suit a library's requirements than one might think. The key is for both sides to make clear what is desired and why. Often it is possible to arrive at an acceptable compromise. And, as noted above, keep in mind -- in deciding which issues to try to change -- the need both sides have for efficiency.

CONSORTIA SHOULD BE ABLE TO SHARE PRICING AND OTHER TERMS WITH EACH OTHER.
- We recognize that in most U.S. states contract terms for state institutions are a matter of public record. Therefore, it is understood that public contract terms will be shared. However, we feel librarians must also understand that publishers are interested in experimenting and in tailoring their offers to the specific opportunity presented by a given consortium. Each consortium is different, so it will be normal to have some differences in agreements. Electronic products are often customizable and it is difficult to compare the terms and conditions of two consortia. In the longer term, even fewer organizations will be receiving the same mix of content and functionality. Publishers may change their strategies or pricing propositions as they learn more or get market feedback. Consortia should take advantage of this flexibility to craft the best arrangements rather than pushing a publisher to make identical offers.

LIBRARIES SHOULD NOT BE ASKED TO PAY FOR UNDELIVERED FEATURES.
- We agree. This is an example of what makes comparing the prices paid by different consortia difficult. The deals may be negotiated at different stages in the service's development process and may reflect a different set of features or file availability.

NO HIGH PREMIUMS FOR BETA VERSIONS.
- Again, the principle is sound. One may question what is considered "high" and what is a "beta version", however.

LIBRARIES SHOULD NOT PAY ALL R & D COSTS - THESE TO BE SHARED WITH SHAREHOLDERS.
- Elsevier Science has set its prices for electronic products to reflect market and competitive realities, not recovery of costs. The prices we are charging for the electronic contents and infrastructure are below cost. We hope that there will come a time when our costs drop sufficiently to be at or below our electronic prices.

LIBRARIES WOULD LIKE TO BUY ELECTRONIC WITHOUT THE PAPER AND AT A COST LOWER THAN THE COST OF A PRINT SUBSCRIPTION.
- Both EES and ScienceDirect permit buying of an electronic-only subscription. There are several alternatives for purchasing and, under some of these alternatives, a decision not to buy the paper results in the electronic version being slightly lower in price than the paper subscription. We would note, however, that over time the electronic version will be less and less a mirror of the paper, as it will have non- printable items and other features of greater value. Therefore, it should be expected that there will be more and more distinction between the paper and electronic editions and their pricing.

PUBLISHERS SHOULD HAVE MULTIPLE, FLEXIBLE PRICING MODELS.
- Elsevier has probably lead the way here, as we have both remote and local options and, within each, more than one model for licensing. In addition, we are testing new models with the University of Michigan in its PEAK project (http://www.lib.umich.edu/libhome/peak).

BUNDLING PRINT AND ELECTRONIC SHOULD NOT BE THE SOLE OPTION AND PUBLISHERS SHOULD NOT TIE TO A BASE YEAR OR NO CANCELLATION.
- We offer a number of alternatives. Some of our individual journals are available on their individual Web sites free to individuals at institutions subscribing to the paper. One of the reasons for this is to minimize overhead costs and permit outreach to authors and readers of these journals. Our broad institutional products (EES and ScienceDirect) do not bundle paper and electronic - as noted above, you can buy electronic only. However, we do also try to look at the total amount of money being spent on Elsevier journals at that institution, and we can offer terms (for example, caps on annual increases) if there is a willingness to make a longer commitment. Libraries always have the alternative to choose another, less binding option (with the understanding that the terms will generally not be as favorable).

ELECTRONIC FILES SHOULD BE AVAILABLE AT OR BEFORE PAPER.
- We agree and we are working to reach that target. Our current pricing reflects current delivery practice.

LIBRARIES WANT PERPETUAL ACCESS TO CONTENT, TRANSFERABLE TO DIFFERENT DELIVERY OPTIONS.
- This is a very complicated area and one that will take a great deal more discussion. Ironically, this has generally not been an issue in the twenty years of online database services (e.g., Dialog). Services have come and gone, as have databases, without the requirement of perpetual access. As we move now into more archival information being (potentially) solely available in electronic form, the concerns are growing. Elsevier Science is committed to working with libraries to find solutions for long-term access and archive issues.

LIBRARIES SHOULD BE ABLE TO ARCHIVE (E.G., MAKE BACK-UP COPIES)
- We do not now have wording in our licenses that covers this and we would want to work through specific examples with library partners.

PUBLISHERS SHOULD OFFER OPEN LOCAL SYSTEMS, PORTABLE TO MAJOR PLATFORMS AND COMPLIANT WITH CERTAIN STANDARDS.
- Our EES service (which grew out of the TULIP experiment) has always been a champion of open architecture. We are not currently providing the data in MARC format, but the headers are SGML and can be routinely translated into local library systems.

PUBLISHERS SHOULD OFFER A VARIETY OF FORMATS: REAL PDF, SGML, HTML, ETC.
- Our ScienceDirect service offers some journals in real PDF and HTML (generated from SGML on the fly) and will ultimately make all journals available that way. Our EES local solution also plans to offer real PDF, HTML and SGML. However, we would note that there are very significant costs associated with this for the publisher. Libraries should expect that (1) not all publishers will be able to do this and (2) for many publishers there will be a need to charge a higher price to cover these options.

PUBLISHERS' OFFERINGS SHOULD BE ABLE TO BE INTEGRATED WITH OTHER SYSTEMS.
- Again, this was a principle on which TULIP and EES were built. We completely support this and are very willing to work with libraries to make it happen.

GIVEN THAT E-INFORMATION PROVIDES NEW CAPABILITIES AND VALUE-ADDED FEATURES, PUBLISHERS SHOULD NOT PLACE ANY UNDUE RESTRICTIONS OR BURDENS ON USE.
- This is a very broad statement, as what a librarian may call "undue restriction or burden" may be felt as essential for the publisher. Many content providers are highly concerned that the digital capabilities make infringement easier and therefore places them at risk. Our basic philosophy is that each institution should enter into a license on the part of the users who belong to that institution. The Elsevier Science licenses are very liberal as to what those authorized users can do (including course pack use) and this has not been a problem area in license negotiation.

PUBLISHERS SHOULD PERMIT FAIR USE UNDER THE LAWS OF THE COUNTRY OF ORIGIN AND PERMIT PAPER AND ELECTRONIC ACADEMIC INTERLIBRARY LOAN.
- Contrary to the statements made in the introductory section of the ICOLC statement, we do not think fair use is "under attack". Many fair use laws have ambiguities built into them and publishers - just as librarians - are in good faith trying to resolve those ambiguities in the electronic context. Licenses are an appropriate way of doing this, as license wording is generally clearer than statutory language.
     As to the specific ICOLC statement, we admit to being uncertain about the notion of using the laws of the "country of origin". It seems to us that the applicable fair use laws should be those of the country of the licensee, but we are open for discussion on this. (Would German librarians agree they are subject to U.S. fair use laws because the ScienceDirect database is housed in Ohio?)
    With respect to ILL, this has been a long and difficult discussion. In the U.S. we have asked that librarians be more open to acknowledging that the electronic world is not the same as the paper environment and therefore there need to be new guidelines (CONTU having been specifically negotiated to cover photocopying in the pre-document delivery world). That has not happened.
    We have recently announced a change in our ILL policy to permit the use of licensed electronic files as a source for printing copies to be sent to other non-commercial libraries. The new policy is intended to provide libraries with greater flexibility and efficiency in meeting their national lending obligations.

LIBRARIES SHOULD NOT BE LIABLE FOR CONTENT MISUSE IF THE LIBRARY DID NOT "INTENTIONALLY ASSIST IN OR ENCOURAGE SUCH BREACH TO CONTINUE AFTER HAVING RECEIVED NOTICE."
- There seems to be agreement that the online service provider (in this case, the university or the library as the contracting agency) must be aware of an abuse before any repercussions would be considered. The question is: what happens once the abuse is known? If the library steps in but nothing changes, are there no consequences? It is not our view that the library should have criminal or financial liability. However, depending on the circumstances, continuing abuse could lead to termination of the license.

ELECTRONIC FILES SHOULD BE OPEN TO WALK-IN USERS.
- Our licenses permit on-site access for members of the public who are permitted to enter and use the library.

USAGE DATA SHOULD BE SHAREABLE WITHIN THE CONSORTIUM.
- We agree, provided the data sharing is indeed limited to consortium members.

THE PROVIDER SHOULD GATHER AND SHARE USAGE DATA WITH THE CONSORTIUM.
- Again, in principle we agree. Libraries and publishers need to negotiate realistic expectations about what data can be readily gathered by the content provider (particularly where there are third-party aggregators involved).

ANONYMITY AND CONFIDENTIALITY OF INDIVIDUAL USERS SHOULD BE MAINTAINED.
- We understand the implications of the privacy issue. It should be understood that there is a compromise between monitoring for misuse and invasion of privacy; in the former one looks to the pattern of use, not the specific files searched, search terms or end user identity.

PUBLISHERS SHOULD ADHERE TO LIBRARY PRINCIPLES ON PRIVACY.
- Probably no problem - but these are not familiar to most publishers.

BE FLEXIBLE AS TO AUTHENTICATION.
- We offer IP, ID and password, and a combination (where an individual user within an IP environment can elect to set up an ID and password in order to do the customizing that individual accounts permit).

IN SUMMARY
- There are a lot of comments here, but it is our contribution to the dialog that the authors of the ICOLC statement want to have. Elsevier Science is in agreement with, and practices, most of the points raised in the ICOLC statement. But we do have slightly differing views on a few areas and look forward to discussion, either with ICOLC or with individuals negotiating consortium and single institution licenses.

Contact:
Karen Hunter
Senior Vice President
Elsevier Science
tel. 212-633-3787
fax 212-633-3764
k.hunter@elsevier.com