Electronic Publishing Programs: Issues to Consider Issues in Science and Technology Librarianship Fall 1996 DOI:10.5062/F4HH6H29 URLs in this document have been updated. Links enclosed in {curly brackets} have been changed. If a replacement link was located, the new URL was added and the link is active; if a new site could not be identified, the broken link was removed. Electronic Publishing Programs: Issues to Consider Elizabeth W. Brown Library and Information Center Georgia Institute of Technology Andrea L. Duda Davidson Library University of California, Santa Barbara At first glance, the proliferation of Web journal publishing, particularly from the commercial sector recently, may seem like manna from heaven -- librarians can now offer users searchable full-text of important scientific journals, as well as enhancements found only in the electronic versions. When it comes to electronic publications, however, librarians must consider a host of details. Some electronic publications have limitations in their licensing agreements that may make them less than ideal. Some licenses state, for instance, that libraries may not make copies (either electronic or print) for users who are not part of the library's main user group. This makes interlibrary loan of these materials impossible. Other licenses state that access must be limited to the library's primary user group. How does one enforce this when a member of the community comes into the library and wants to use an e-journal or database? Requiring a library password for access puts these journals into a category with other restricted databases; yet in most libraries, anyone who can enter the library can look at print journals. It is not unreasonable for librarians to expect to provide the same access whether the resource is in print or on the web. Librarians can either accept licenses as presented and live with the consequences or reject specific objectionable terms, with the inevitable resulting delays. The University of Texas Office of General Counsel has prepared a software and database license agreement checklist which is available at . The checklist allows librarians to check off clauses showing their requirements (e.g., whether the item will be accessed from a single machine only, from one building, from the entire campus, or from off campus) and asks about clauses that may limit the library's rights (e.g., a clause that gives the vendor the right to terminate without providing the library with a similar right). The analysis will then respond with things to look for in the license or clauses that should be added. Access using passwords or IP addresses can have an effect on ease of access to web-based journals. In some cases, publishers haven't allowed for multiple IP addresses on a single campus. Requirements such as registering each class C address have caused headaches for the librarians responsible for compiling this information. Institutional passwords can also cause headaches -- unless the publisher provides scripts so that entering passwords is transparent, librarians have to find a way to make this information available to their users. In addition, individual subscriptions aside, librarians may not fully comprehend how terms vary from publisher to publisher until they subscribe to more than one electronic journal product. For instance, one subscription may limit use by IP address to a section of campus with a maximum of 25 simultaneous users, the second subscription may limit access to users in the library building, and the third may not use IP address but instead a campus ID which has to be programmed somehow by the library. Librarians then have the challenge of clearly presenting these new services and their use restrictions to the users. Interestingly, librarians going through some of the more troublesome setups may realize that the terms seem more suited to individual subscribers rather than institutional subscribers. Indeed, offering e-journal products to library customers may very well pose as big a challenge to the publishers as to the libraries purchasing them! In some cases, publishers are opting not to offer their electronic products to institutional subscribers yet (e.g., AAAS with Science Online full-text, as mentioned in the previous article). At this point it may be difficult for libraries to make commitments to provide access to e-journals since the prices are in flux. Many publishers are making e-journals free in 1996 or 1997 with subscription prices for future access to be announced later. Thus, if libraries take advantage of free online versions in 1996 and 1997, they must plan for 1998 and 1999 budget allocations or face yanking away access that may have been available for a year or more. This extra budget allocation comes at a time, however, when many libraries are canceling subscriptions. Librarians are faced with a tradeoff between providing enhanced access to a title already available in print, or subscribing to a different title of interest. Another worry is committing to subscribe to a service with an extremely reasonable price now, with the possibility that the prices can go up substantially later. Librarians have experienced this phenomenon with other new product formats in the past, and at least one new electronic journal has had a significant price increase from its first year to its second. The issue of archiving e-journals has been an active topic on the ARL-EJOURNAL list. (Subscribe by sending a message to listproc@cni.org with the text "subscribe arl-ejournal ".) Who is going to be responsible for archiving electronic journals? Do publishers have enough incentive to provide this service or will archiving be left up to libraries? What formats will be used for storage? The discussion has featured few quick and easy answers but much thoughtful deliberation from all sectors. Some organizations are going ahead with promising initiatives, however. OCLC has undertaken an archiving project called Electronic Collections Online. They offer to archive the files of publishers' journals at no cost to the publisher. Libraries will subscribe to these journals through the publisher, as they have traditionally. Libraries will then be able to subscribe to OCLC's ECO archival program for access to the electronic copy of the journal. OCLC promises to maintain these files into the future and maintain a record of each library's subscription. (Gherman 1996) The amount of disk space required for archiving journals is staggering. In a message to ARL-EJOURNAL, Ken Metzner reported the following: Academic Press has 173 journals online in the IDEAL service. This amounts to about 20 Gigabytes per year. It represents perhaps 4% of the STM literature. So, one year of STM as a whole would be half a Terabyte. Indexes would add more. New multimedia elements added to research articles could easily double it. Is that a lot? Depends on your perspective. Certainly can't be done for free in your spare time. (Metzner 1996) While individual libraries may not be in a position to store digital versions of journals, they may be able to turn to an organization like JSTOR (Journal Storage). JSTOR began as an effort to ease the increasing problems faced by libraries seeking to provide adequate stack space for the long runs of backfiles of scholarly journals. The basic idea was to convert the back issues of paper journals into electronic formats that would allow savings in space (and in capital costs associated with that space) while simultaneously improving access to the journal content. It was also hoped that the project might offer a solution to preservation problems associated with storing paper volumes. To demonstrate the concept, the Mellon Foundation sponsored a pilot project to provide electronic access to the backfiles of ten journals in two core fields, economics and history. Five library test sites were selected initially. Every issue of the ten participating journals published prior to 1990 -- approximately 750,000 total pages -- has now been converted from paper into an electronic database that resides at the University of Michigan and is mirrored at Princeton University. Using technology developed at Michigan, high-resolution (600 dpi) bit-mapped images of each page are linked to a text file generated with optical character recognition (OCR) software which, along with newly constructed Table-of-Contents indexes, permits complete search and retrieval of the journal material. (JSTOR 1996) This topic is discussed further in "Preserving Digital Information: Final Report and Recommendations" from the Committee on Preservation and Access and Research Libraries Group. It is available at Where the electronic journals have print analogs, indexing is generally not a problem. However, for electronic-only journals, no indexing may be available. Chemical Abstracts Service has announced that few electronic publications meet their criteria for inclusion in Chemical Abstracts. Since CAS began abstracting and indexing electronic documents in April 1995, only 197 papers from 14 online journals and 45 papers from two conferences have been added. Before adding electronic documents to Chemical Abstracts, documents must meet the following criteria: report new information of chemical or chemical engineering interest, present information in a form similar to that of a scientific paper (i.e., reporting novel findings with some discussion of the experimental details), identify authorship, be publicly available, have some relative permanence or continued accessibility, and be original publications, not issued previously in another form or medium. (Shiveley 1996) Obviously, these and other issues give librarians much to think about in planning for "electronic expansion" of their library's collections. Each library will have to assess all its users' needs and work hard to balance priorities during this "transitional period." Indeed, Dean Miriam Drake notes elsewhere in this issue the conflicting generational points of view on campuses and the resulting challenges for library planners. In the meantime, librarians need to continue to express their concerns to publishers about the issues raised in this article. In some cases, publishers have not fully fleshed out their electronic publishing goals (hence the lack of pricing information!). Many publishers have actively sought (and continue to seek) input from their library customers in order to provide information in the most usable way; for instance, the American Astronomical Society cites a survey of librarians conducted in spring of 1996 as a deciding factor in setting the subscription policy for its new electronic edition of ApJ. Librarians and publishers still have work ahead to understand each other's needs and limitations. Over the next two years, some technical issues may be resolved and more standardization may actually begin to emerge in how the products are implemented. On the other hand, resolving the issues of intellectual record--archiving and indexing--may well represent one of the biggest challenges to the library profession this century. What other issues do librarians and publishers need to address? How have you solved problems arising from providing access to e-journals in your library? Are publishers or librarians being unrealistic in their expectations? We invite both publishers and librarians to send e-mail to the editor for possible inclusion in a future issue. References: Gherman, Paul M. [gherman@library.vanderbilt.edu]. "Re: archive sites and more." In ARL-EJOURNAL. [arl-ejournal@cni.org]. November 20, 1996. "JSTOR: The Need." [{http://www.jstor.org/about/need.html}]. 1996. Metzner, Ken. [kmetzner@acad.com]. "Re[2]: Kick off discussion: Archiving." In ARL-EJOURNAL. [arl-ejournal@cni.org]. November 20, 1996. Shiveley, Eric. [eshiveley@cas.org]. "CAS experience with Internet publications." In CHMINF-L. [{CHMINF-L@IUBVM.UCS.INDIANA.EDU}]. November 21, 1996.