Note: This paper provides a sampling of technology-based activities in STI as of March 1994. It is not a complete survey of projects and initiatives.
There are over seventy peer-reviewed electronic journals, according to the latest edition of the ARL OSAP Directory of Electronic Journals, Newsletters and Academic Discussion Lists (in press). A minority are in scientific or technical fields. Psycoloquy, which began in 1990, was the first refereed journal with rigorous peer-reviewing standards. The Bulletin of the American Mathematical Society was one of the first journals on the Internet to display mathematical formulae.
The MIT Press and the MIT Libraries have announced the publication of the Chicago Journal of Theoretical Computer Science, which represents a new relationship between publishers and libraries to meet "the scholar's desire for quicker peer review and dissemination of research results, the library's need to develop systems and structures to deal with electronic journals...[and] the publisher's need to develop an economic and a user model for electronic dissemination of scholarly journals."
Preprint distribution has traditionally served the important function in the sciences of rapid distribution of scientific papers. Preprints are often the main way that some information reaches its primary audience. Electronic preprint services have begun to replace the traditional system of distribution through the mail. The electronic services facilitate quick access by a wide audience.
A well-known electronic preprint service in the sciences is HEP-TH. Begun in 1991, HEP-TH is a electronic preprint archive and distribution service in the field of theoretical physics. Researchers submit electronic versions of preprints which are then stored indefinitely. Users can scan a list of preprints available, read the abstract, and request an electronic copy. In early 1993, the system had over 2000 users. Preprint bulletin boards in over a dozen other subdisciplines of physics have been established. There is also preprint activity in a number of other disciplines, including mathematics, computer science, history and philosophy.
In 1989, the National Science Foundation funded the American Mathematical Society (AMS) to develop an electronic service for mathematicians and a prototype electronic journal. The service, e-math, provides access to publications, software, preprints, and discussion lists. The Bulletin of the AMS , on the Internet since 1992, displays non-textual material such as mathematical equations and formulae.
NASA, the American Astronomical Society and the University of Chicago Press collaborated on a multi-year project, STELAR. Several astronomy journals were converted to page images and made available in prototype form. The AAS and the University of Chicago Press are now developing a platform to make the AAS journals available electronically.
The American Physical Society has three electronic publishing projects underway. They are (1) making past volumes of the Physical Review available over a Wide Area Information Server in collaboration with the Los Alamos National Laboratory; (2) scanning Physical Review Letters and Physical Review E and storing them on optical disks, in cooperation with the Naval Research Laboratory Library; and (3) soliciting proposals to create an on-line version of Physical Review Letters, to be published along with the printed version.
The Association for Computing Machinery (ACM) contracted in 1993 with OCLC to develop an electronic publishing system. The in-house electronic publishing system will integrate ACM publishing functions into an automated system that will encompass writing, editing, composition, production, archiving and eventually distribution of over 40,000 pages per year. The system will be completed in 12 to 18 months.
The American Chemical Society plans to expand its Chemical Journals Online(CJO) Project. CJO is currently a text-only database which includes 57 journals. ACS plans to add figures, tables, photos and formulas to the database.
TULIP (The University Licensing Program) is a project developed by Elsevier in conjunction with nine research libraries to test systems for networked delivery and use of journals. Bit-mapped journal pages from 43 Elsevier and Pergamon materials science journals are delivered via the Internet and made available over local campus networks. Full implementation of the project at all nine sites is expected during 1994-1995. The University of Michigan was the first site up and running. University implementation varies at each site, with some approaches focusing on search capabilities, while others stress the browsing function. TULIP is conceptualized as an experiment, a learning experience that will inform future electronic distribution ventures.
The University of California and AT&T Bell Laboratories are collaborating with several publishers on the Red Sage project to explore issues surrounding scientific communication in a networked environment. Its objective is to make journals in molecular biology and radiology available over the university's campus network, in a way that makes the computerized version as similar as possible to the print format. The "RightPages" software developed by Bell Labs displays both text and image files. The project will examine issues beyond the technical factors, including the economics of electronic journals.
The Chemistry Online Retrieval Experiment Project (CORE) is a collaborative effort by the American Chemical Society, Bellcore, the Chemical Abstracts Services, the Cornell Mann Library, and OCLC. It will create an electronic database of chemistry literature and a system to deliver it to desktop computers. CORE will provide ten-years of full-text of 20 American Chemical Society Journals, as well as selected monographs from Springer -Verlag. Multiple interfaces to the database, including hypertext links, are being developed to permit a variety of ways to display, search and navigate the material.
This five year project supported by the NSF is investigating the creation and application of an emergent "collaboratory" model to support distributed collaborative science in ways that reduce barriers of both time and distance. Participants are the University of Michigan, SRI International, the University of Maryland, Lockheed Palo Alto Research Center, and the Danish Meteorological Institute. Scientists, students, and staff, by way of Internet, can now conduct real-time experimental campaigns using a suite of observational instruments located in Greenland. Data streams, graphical representations, commentary, and annotations of the group are captured and available for later replay, refinement, editing, and inclusion in an integrated-media electronic journal.
The Human Genome Initiative was the first full-scale effort to apply the principles of "Knowledge Management," the involvement of librarians and information managers at all stages of a project in order to collect, organize, manage and provide access to the massive archives of data being generated. The Human Genome Initiative, which comprises the Human and Plant Genome Projects, is an ambitious fifteen-year effort that seeks to decipher the complete genetic code of the human genome. Funded by several U.S. government agencies, the initiative involves the disciplines of chemistry, biology, physics, mathematics, engineering, computer science and information science. It is an international collaboratory project with an estimated budget of $200 million a year.
In the fall of 1993, the National Science Foundation announced a joint initiative with the Advanced Research Projects Agency to fund programs of research, development and testing of elements of a digital library on a significant scale in a distributed environment. The NSF considers information accessed via the Internet to be the ingredients of a digital library, and programs funded under the initiative will explore not only aspects of connectivity but how to achieve digitization of large amounts of information in an economically feasible manner, and how to store, search process and retrieve information in a user friendly way. Awards of up to $1.2 million per year for up to four years will be made in the areas of: electronic data capture, categorization and organization; development of intelligent software for browsing, searching, filtering and related functions; and utilization of networked databases throughout the nation and the world. Nearly 80 proposals were submitted to NSF, awards will be announced in the summer of 1994.
The US Global Change Research Program is part of the worldwide Global Change Initiative, which seeks to observe, understand and predict global change. A fundamental part of the Initiative is data and information gathering in order to conduct research, develop a model of global change, and make assessments. The initiative is noteworthy because it is formulated on the principle of science-driven data management, carried out by researchers and other users working together with data managers from start to finish. The USGCRP is a multi-agency project and its information and data aspects are handled by the US Global Change Data and Information System. The system is designed to enable a diverse user community to learn what data and information are available, have key material available in easily accessible forms, and be assured of its quality and continued availability.
Several studies of the economics of scholarly communication have been issued in the last five years. Economic Consulting Services' Study of Trends in Average Prices and Costs of Certain Serials Over Time, a consultant report prepared in 1989 as a part of the ARL Serials Prices Project, includes discipline-specific price trends.
In its 1990 report to the National Science Foundation, Communications in Support of Science and Engineering, , the Council on Library Resources published a study on the correlation between library resources and scientific productivity. The study, "Library Resources and Research Productivity in Science and Engineering," was conducted by Nancy Van House. It found that there is a correlation between scientific and engineering library resources and scholarly productivity, as measured by publications and faculty honors. The study raised questions about the effect of declining library purchasing power on scientific productivity. In a continuing effort to explore the economics of scholarly communication, the Council has recently funded an investigation into the economics of networked information to be conducted by Paul Peters of CNI and Richard West of California State University.
A special issue of Serials Review, (Volume 18 #1-2), "Economic Models for Networked Information", included several papers on STI. Richard Katz's article, "Academic Information Management at the Crossroads," is a particularly useful discussion of the need for a "thorough study of both the macro- and microeconomics of creating, publishing, distributing and managing publication quality academic information."
University Libraries and Scholarly Communication, published in 1992 by ARL for the Mellon Foundation, includes an analysis of book and serials prices by field. In the spring of 1994, the Mellon Foundation issued a discussion paper, "Scholarly Communication, Academic Libraries, and Technology," which outlines the Foundation's priorities for project funding. The Foundation intends to begin by funding projects in the area of journal publication, such as the economics of publishing and scholarly communication, and alternative approaches to the publishing/delivery/storage/retrieval aspects of journals.
See also Economics of Scholarly Communication: A Working Bibliography, included as Appendix D to the Task Force report.
Prepared by Diane Harvey, ARL, for the AAU Task Force, March 1994.
INFORMATION GENERATION AND CREATION
Includes data collection and analysis/synthesis.
AUTHORING
Writing, revising and improving.
INFORMAL PEER COMMUNICATION
Access by peers, distribution of preprints etc.
EDITORIAL AND VALIDATION
Editing processes, peer review (quality control), market assessment by publishers and editors (identification of unmet needs, what else exists, what needs to be published).
OWNERSHIP, PRIVACY AND SECURITY
Copyright issues, policy issues i.e. confidentiality, guaranteeing the authenticity and authority of text.
DISTRIBUTION
Making copies after the first copy available on a wholesale basis .
ACQUISITION AND ACCESS
Includes personal and institutional purchase; includes access and ownership. Includes the decision to acquire, or decision not to purchase and to rely on access.
STORAGE
Holding and making available (the "place" dimension).
PRESERVATION AND ARCHIVING
Includes decision to archive and preservation (conservation) functions (the "time" dimension). Includes assuring the security of the item in storage (e.g. in stacks or archived on network).
INFORMATION MANAGEMENT
The processes of identifying, describing and structuring the item in order to facilitate "discovery." Includes processes such as bibliographic control.
LOCATION AND DELIVERY
Identification of sources of information and obtaining the information. Includes reference and training.
RECOGNITION
Institutional rewards and recognition.
DIFFUSION
Access to those outside author's primary community.
UTILIZATION OF INFORMATION
By user
EASE OF USE
How easily and effectively does the system make information accessible to known users and potential users?
TIMELINESS
How long does it take for the information to become available?
RESPONSIVENESS
How quickly can needed information be identified and accessed?
ACCURACY
How error-free is the information at each stage in its life cycle (through mechanical or system transmission)?
AUTHENTICITY
How much does the information get distorted or changed as it moves through the system (through human processes)?
PREDICTABILITY
How reliable and consistent is the system in maintaining levels of quality and availability?
ADAPTABILITY
How flexible is the system in providing new approaches to information or providing access for unanticipated users?
RELEVANCE
How well does the system provide mechanisms such as filtering and assessment of information?
ELIGIBILITY
Who has access to information in the system?
COST
What are the system and unit costs, and to whom?
RECOVERY
How well is the system able to avert or recover from error ( caused by mismanagement or lack of resources to make the system work)?
INNOVATION
How well does the system perform research and development to provide system innovation?
EXTENSIBILITY
How well does the system integrate between media? Between disciplines? What is the system's ability to build and extend itself without a total restructuring?
NOTE: These categories are not mutually exclusive; they are a mix of individual and institutional players with significant overlap.
CREATORS
COPYRIGHT OWNERS
SCIENTIFIC SOCIETIES
ABSTRACTING & INDEXING SERVICES
PUBLISHERS
VENDORS, JOBBERS AND OTHER WHOLESALERS
UNIVERSITIES
UNIVERSITY LIBRARIES
UNIVERSITY COMPUTER CENTERS
GOVERNMENT AGENCIES (FEDERAL AND STATE)
INDUSTRY
NON-PROFIT ORGANIZATIONS
USERS (SCIENTISTS, ENGINEERS, OTHER PROFESSIONALS, ETC.)
In order to evaluate different options for managing STI, the Task Force developed an analytical framework (Appendix B) that describes the system of scientific and scholarly communication. The analytical framework identifies system functions, performance attributes, and system participants.
Using the components of the communication system identified in the framework, three models of information creation, dissemination, and use -- Classical, Modernized, and Emergent -- were formulated and assessed. The following is a description of each model, along with the Task Force's assessment of how each model performs the functions of scientific and scholarly communication.
The term "model" as used in these formulations is not used in a predictive or prescriptive sense. The models are, rather, intended as descriptive representations of various ways that scholarly and scientific communication operate now and could operate in the future.
The print-based traditional scientific scholarly communication model is exemplified by the scholarly journal. The print journal remains, for most fields, a primary model of scientific and technical communication. It is estimated that there are over 20,000 scientific and technical journals worldwide. Current institutional support for STI has been predicated on this model.
The scientific journal, as the primary method of print-based communication, supports the authoring function well in the areas of eligibility, accuracy, and integrity. It fares less well in adaptability from the author's perspective, since the format for journal articles is fairly rigid. In the traditional journal system, authoring functions are performed by individual scientists or groups of scientists, with costs absorbed by the author's institution.
As a means of informal peer communication, the print journal is particularly poor in serving the performance attributes of, for example, timeliness, responsiveness, and adaptability. Distribution of preprints by authors and presentation of conference papers are means of overcoming this disadvantage.
Editorial and validation processes as performed by journal publishers and peer reviewers, however, work in a fair to good manner, especially in providing accuracy, integrity and relevance. Timeliness is the attribute performed least well by this function. The peer review system is an established strength of the print-based system. Editorial and validation costs are borne by publishers and by the institutions which employ the editors and reviewers.
Functions related to ownership, privacy and security, such as copyright, are well served by the traditional journal. The fixed nature of the print journal, which provides such positive attributes as integrity, accuracy and predictability has a downside, which is a lack of adaptability in providing new approaches to information. As confidential materials are not published in most scientific journals, the functions of privacy and security of STI are not invoked.
Performance of distribution functions varies from good (providing ease of use and integrity) to poor (problems with timeliness, responsiveness and adaptability). The time lag in journal publication is particularly problematic, as is the inability of print journals to provide flexible approaches to information. Distribution is performed by publishers, vendors and subscription agents.
Acquisition and access, from the personal and institutional perspective, function well in assuring predictability, integrity and relevance. They perform less well from the perspective of ease of use, timeliness, responsiveness and eligibility. Both distribution and acquisition/access discriminate in favor of ability to pay and thus negatively impact eligibility and responsiveness; although institutional access (for example, by libraries or academic departments) increases equity of access.
Preservation and archiving of print materials varies depending on the type of material. The printed journal is well managed in terms of preservation and archiving compared to other kinds of scientific information. The attributes most important to preservation-- integrity, accuracy and predictability-- are well served. Preservation and archiving are enabled by institutional investment in library buildings, collections, and services. The methods of preservation and archiving vary and may involve reformatting from paper to film or magnetic media. The method chosen, especially if it includes reformatting, will impact (positively or negatively) on attributes such as ease of use and timeliness.
Storage of print-based materials has traditionally been the purview of the library. Institutional investment in building and maintaining library facilities is the major aspect of storage cost. Since storage arrangements are made on a distributed basis, performance of such attributes as ease of use, timeliness and adaptability can vary. Predictability and integrity are usually good, depending on security measures taken by an institution.
From an information management perspective,which encompasses identifying, describing and structuring the information base, the print-based system performs in a fair to good manner. Costs to support these functions are borne by the institution, directly in the case of investment in institutional structures, and indirectly in the case of (for example) subscriptions to indexes. The current institutional support for libraries has developed in response to print-based information resources. Universities, colleges, governments at all levels, and corporations have invested in library buildings, collections and services to manage STI. Commercial investment in development of secondary information sources such as indexes and bibliographic tools, have contributed to the library's ability to manage print-based resources. The information management function performs best in responsiveness, predictability, accuracy, integrity and relevance, while functioning less well in the areas of ease of use and timeliness.
Location and delivery functions perform in a good manner in the areas of accuracy and integrity. Eligibility is performed in a fair to good manner, while other attributes such as timeliness, and adaptability are not well served by the print journal system .Investment by libraries in systems which reflect materials holdings and agreements which enable cooperation are the foundation for location and delivery of the print journal. Delivery through conventional interlibrary loan is being supplemented by new forms of document delivery, which may improve location and delivery functions.
The print-based scholarly journal serves the function of recognition well. Institutional policies on tenure and promotion are based in part on quality judgements that look at the number of articles published and in what journals.
The print journal provides a primary means of information diffusion, making articles accessible to audiences beyond the author's primary audience. It performs this function in a way that is easy to use and accurate while maintaining integrity and eligibility. It performs less well in the areas of responsiveness, adaptability and cost. Relevance varies depending on performance of the information management function.
Print-based information serves the utilization function well in the areas of ease of use, predictability, integrity and eligibility. The stable format of print journals and their availability through libraries or individual subscription foster utilization. However, print-based materials perform less well in timeliness, responsiveness, adaptability and relevance. Utilization depends on investment in acquisition, access and information management by individuals and institutions; and on services provided for interlibrary loan/document delivery from remote collections. Delivery of print information in electronic format may enable improved utilization.
The model implicit in the Task Force charge describes a distributed (as opposed to a centralized) system of STI collection and dissemination which would coordinate an unspecified number of regional libraries to provide national-scale information services. The modernized model described here is neutral on the issue of centralized versus decentralized collections and services. Such a judgement is best addressed through an assessment of performance attributes including cost.
The modernized model focuses first on what can be termed the "post-press" processes, that is, the functions after distribution of an article or a journal. The modernized model is to a great extent the contemporary model, in that most aspects are available today.
In the modernized model, which can also be called the "document or article delivery", "on-demand" or "just in time" model, information is still largely published and resides primarily in paper form, but there is an array of expanding options for discovery and delivery. At the same time, the modernized model can also accommodate emerging electronic journals that are authored, published, and stored in electronic form.
The modernized model can have three manifestations. In the first, referred to as Modernized A, system inputs (from author to publisher) can be in paper or electronic form, access to information can be through paper (such as print indexes or abstracts) or electronic (online databases) means, but information output is still primarily delivered to the user in paper form (by mail or fax from a commercial document delivery service, for example). In the second variation (Modernized B), information inputs and access remain the same as in Modernized A, but information output can be delivered to the user either in paper or electronic form because of a higher degree of institutional infrastructure readiness. Modernized C, which presumes a major change in university policies on intellectual property (e.g. non-profit retention of some or all rights to the articles created by university faculty and staff), as well as a high degree of infrastructure readiness in terms of network capability, enables information inputs, access and delivery to be primarily in electronic form.
Examples of the modernized model can range from commercial document delivery services such as CARL Uncover, to innovative projects such as Elsevier's TULIP project or the Red Sage cooperative project between the University of California at San Francisco, Springer Verlag and AT&T Bell Laboratories. All these modernized access and delivery services impact the STI management functions after initial publication, and some are planning to broaden their impacts to the functions before publication.
Informal peer communication is facilitated in this system since preprints are available electronically. Ease of use, timeliness and responsiveness are enhanced, while eligibility may be limited. Costs are network based and borne by the institution mounting the preprint system and by the user's institution. In developing the electronic preprint system, authors have modernized in advance of publishers.
Because authors and editors can now communicate electronically, performance of editorial and validation functions are enhanced. Timeliness and adaptability are performed particularly well by an electronic system of communication between author and publisher.
As this system is still largely based on print products, ownership, privacy and security issues raised in the traditional print journal system still apply. Commercial document delivery services provide a mechanism for automatic collecting of copyright fee payment, but may make it administratively inconvenient or impossible to exercise fair use rights. In Modernized B, where the information delivered to user may be in electronic form, questions of copyright and of text authenticity may be raised. In Modernized C, changes in university policy in regard to ownership of intellectual property could result in wider availability of material as well as significant cost savings.
In the modernized system, when intermediary agents such as commercial document delivery services are used, articles are the primary unit of distribution in contrast to the traditional print-based system where journals were the unit of distribution. "Publication" may occur without the generation of paper copy, as in electronic journals. Electronic distribution could enhance timeliness, ease of use, and responsiveness of distribution mechanisms. In Modernized C, institutions may become distributors, and may assume editorial and validation functions.
The major impact of this model occurs in functions after distribution, such as acquisition and access. There are more options for fulfilling the functions of acquisition and access beyond acquiring/loaning/borrowing a paper copy. A greater variety of channels exist for acquisition, either in paper or electronic format. Acquisition and access costs may shift from traditional library acquisition budgets to budget lines that support commercial document delivery or network-based services. Acquisition and access functions and budgetary support in this model may or may not be centralized in the library -- an individual or academic department may also support them.
Storage of information could be significantly affected. For materials reliably available via other sources, storage would no longer be required by local institutions. Document delivery, either through the library or directly to the user, would put "non-returnable" information into the hands of the user so that the library would in many cases not retain a copy of the item. For these reasons, storage space required for individual institutions may not continue to expand at present rates. Digitization of items already in the library, and an increase in electronic outputs could further shift storage costs from shelving to computer storage capability.
Preservation and archiving is also significantly affected in the modernized model. These functions also take on increased importance if the storage sites for information delivery are reduced. The emergence of electronic formats requires development and adoption of guidelines for ensuring the integrity, accuracy and predictability of electronically stored text and graphics. Institutional archiving and preservation costs in facilities, reformatting processes and services will be affected by two trends: the speed by which the initial STI output is distributed in electronic instead of paper format; and the extent to which remote document delivery substitutes for locally held collections.
Information management processes which aid identification are crucial to a system which depends on the ability of a user to "discover" an item that may not be available locally. An "on demand" system requires exemplary information description and identification. Information management costs may decrease in modernized systems. With increased online availability of full text, sophisticated search and retrieval mechanisms can replace traditional descriptive cataloging. Moving from bibliographic description of individual items through cataloging, to a process of assembling and linking records and databases and connecting them to location and delivery information, may reduce the labor intensity of the bibliographic control process as it becomes more automatic.
Location and delivery are key functions in the modernized model. The success of document delivery rests on knowing what is available, where it is available, and how to obtain it. The location and delivery function encompasses reference and consulting to users, and training to maximize use of the system, as well as the communication of an information request and the delivery of needed items. Eligibility for all campus users is a key attribute for delivery of remote information to substitute for local ownership. Traditional interlibrary lending for copies of articles or other "nonreturnable" items,is not employed in this model. Instead, article delivery in the modernized model is through a variety of paper and electronic means: mail or fax delivery of a paper item ordered through an online database, full text CD ROM products, or through the network. This model shifts from the "just in case" system of library acquisitions to the "just in time" article delivery mode; it follows that costs will shift from library storage to location and delivery functions such as commercial document delivery fees or network costs.
Wider options for utilization of information would be provided by new varieties of electronic outputs such as searchable electronic text. There would be new costs for utilization because of investment in desktop capabilities and in equipment such as faxes, printers and scanners.
This model accounts for the use of computing and communications technologies to share instrumentation, primary data, and software tools as well as to share information per se. It also accounts for the use of these technologies for genuinely innovative purposes, such as for interactive, multi-media information environments, rather than for the modernization of page formatted information from paper to network storage and access media. It views each scientific community as a knowledge management system, and it calls upon scientists, information technologists, and librarians to act as partners in all components and activities of that system. Information environments generated by this model are referred to as "collaboratories." Early examples of such environments can be found in the human genome and global climate change communities.
Information generation and creation is a major feature of this model, particularly so in comparison to the other two models. The digital technologies that operate the instruments of scientific research and discovery are more and more the same as those that support the channels of scientific communication and publication. And, information generation and creation in science no longer has to be the solitary, private affair that it once was. Perspectives and skills historically associated with other information management functions are now needed in the process of information generation and creation.
The major change evidenced by the authoring function in this model is the increase in the number of authors associated with a given work. This increase reflects both the "big science" character of contemporary scientific research and the expanded number of collegial relationships scientists are able to establish and maintain using electronic mail and other networked communication facilities and techniques.
In this model the difference between information peer communication and authoring is difficult to sort out in the abstract. Informal peer communication between scientists along the edge of a given research front serves many of the purposes of "authoring" for those same scientists in other models. In addition, many formal, authored communications originate in informal peer communication processes that do not anticipate their eventual end-state. Existing principles and codes of scientific behavior regarding formal communication must be updated and translated to yield principles and codes of scientific behavior regarding informal peer communication and its relationship to formal, authored communication.
This model incorporates and extends the editorial and validation function of the other models to assess and control the quality of primary research data as well as of findings reported via formal channels of scientific communication and publication. It also supplements formal editorial and validation mechanisms, such as editors and reviewers, with informal ones, such as conversations in networked discussion forums about pre-prints, works-in-process, recently published works, and so forth.
This model addresses a number of unique ownership, privacy, and security issues and, in the main, these issues arise from (1) the fact that the model embraces primary research data as well as findings reported via formal channels of scientific communication and publication, and (2) the fact that it supports scientific communities that are highly collaborative. This model does not break new ground with respect to the other two models as regards copyright management in electronic versus print media. It does frame intellectual property issues in completely electronic environments by noting that what constitutes intellectual property in these environments is a difficult question to answer, as are questions of how to apportion and protect ownership to and of that property. This model also raises new issues regarding the privacy of research subjects and sites because the identities of these subjects and the locations of these sites can be inferred much more readily and widely from primary research data than they can from published findings. Finally, security is a heightened concern in this model because the negative effects of corrupted or altered primary research data are much more damaging to science as a whole than is the corruption or alteration of the findings of a single publication.
This model attends more to an integration function than it does to a distribution function as a pressing need of the scientific communities that it supports. It assumes a world in which scientific knowledge bases and tool sets are quite naturally and highly decentralized, even fragmented, according to the centers of excellence and the pools of resources that define their associated scientific communities. This contrasts with the rather more centralized, if distributed, world views implied by the other two models. This model also assumes a world in which scientists choose among a variety of non-profit and for-profit mechanisms and channels that provide them with a wide range of audiences and benefits when they decide how to publish (etymologically, to "make public") their knowledge bases and tool sets.
The acquisition and access function provided by this model (1) is very much more "just in time" (acquisition and delivery of something at the time it is actually needed) than it is "just in case" (acquisition and delivery of something in advance of it actually being needed), (2) is very much more "virtual" (use of something housed by someone else) than it is "physical" (use of something housed by oneself), and (3) the person effecting such acquisitions or accesses is very much more likely to be someone associated with a scientific team, project, or department than it is to be someone associated with university-wide service such as a library, which is not to say that that someone is likely not a librarian. Even so, acquisition and access activities under this model are conducted within a university-wide framework that comprises (1) the university's information technological infrastructure, (2) the university's basic information resources and tools, (3) the training and support services offered by the university, and (4) the information resource and service cooperative agreements and commercial contracts maintained by the university.
The unique preservation and archiving function of this model arises from (1) how difficult it is to distinguish between formal (and, therefore, presumed appropriate for preservation and archiving) and informal (and, therefore, presumed not appropriate for preservation and archiving) communications among scientists, and (2) how difficult it is to differentiate between personal, professional, and institutional interests in and responsibilities for preservation and archiving. This model addresses preservation and archiving as a subfunction of most if not all of the other functions, and it is particularly focused on anticipating preservation and archiving in the information generation and creation and informal peer communication functions. It sees preservation and archiving as primarily an institutional and professional rather than a personal function. And, it conceives of preservation and archiving as a function that serves scholars who are interested in science as well as the scientists who do science.
The storage function of this model is first and foremost the responsibility of the individual scientist or of the individual scientific team, project, or department who/that generates or contributes a given knowledge base or tool set to the scientific community. Individual scientists and individual scientific teams, projects, and departments may appeal to institutions, professions, or commercial enterprises for assistance with fulfilling this responsibility. Furthermore, as the interests and fortunes of individual scientists, teams, projects, and departments change, the storage function under this model may be assumed by such institutions, professions, or commercial enterprises.
The information management function of this model is quite different than that of the other models. The differences here arise, in the main, from differences already considered above in the discussion of the information generation and creation, distribution, acquisition and access, and storage functions. Particular attention is paid by this model to information management for the benefit of those who are not on the forward edge of a given scientific research front: cross-disciplinary specialists, students, future generations, and the like. Another special interest in information management under this model is the potential of computer and communication technologies to enable a transition from a world in which scientists look for information to a world in which information looks for scientists: topical subscription services, personalized journals, intelligent agents, and the like. This model addresses information management as a subfunction of most if not all of the other functions, and it is particularly focused on anticipating information management in the information generation and creation and distribution.
This model subsumes location and delivery under distribution,acquisition and access, preservation and archiving, storage, and information management. It assumes a highly decentalized and fluid world of scientific and technical information in which the same knowledge base or tool set is available from an ever changing set of suppliers, each of which is able to deliver (access to) that knowledge base or tool set in a variety of value-added packages and according to a variety of technical and financial schemes.
Scientific recognition and reward principles and protocols are rekeyed to account for the distinctive behaviors, processes, structures, and outcomes of this model. In particular, this model requires that science assess, recognize, and reward excellence in the publication of primary research data, tool sets, and other innovative elements of scientific knowledge bases as well as findings and theories such as those that appear in the traditional, primarily print, scientific literature. The model also requires that scientific awards reflect the fact that highly collaborative science complicates the process of attributing credit for a given scientific break-through to one individual or to a relatively small number of individuals.
This model subsumes diffusion under information generation and creation, distribution, acquisition and access, preservation and archiving, storage, and, especially, information management. A risk of this model is that the information environment of a particular scientific specialty becomes so optimized to the objectives, methods, and findings of that specialty that the environment becomes inaccessible to scientists and other interested parties who do not share that specialty. Partnerships of scientists, information technologists, and librarians are used by this model, particularly in the information generation and creation and information management functions, to assess and control this risk.
This model produces major, positive changes in the utilization of information, changes that incorporate and extend the positive changes produced by the other two models. These new changes flow, in the main, from the "context of work" that this model creates by providing highly integrated, customized, and responsive access to scientists, instruments, primary research data, tools for analysis and visualization of research data, theories and findings, commentary and opinions, expository and curricular materials, and other people, resources, and services of value to science and scientists. This model also creates this context of work wherever the scientist might be whenever he or she might be there, provided, to state an obvious but important caveat, that the scientist in question has access to the requisite computing and communications technologies at that time and in that place. The dramatic improvements in the utilization of information enabled by this model translate directly into dramatic improvements in the quality, pace, and productivity of scientific research.
Baumol, William J. and Matityahu Marcua. Economics of Academic Libraries. Washington, DC, American Council on Education, 1973.
Brinberg, Herbert. "Information Economics: Valuing Information." Information Management Review 4 no. 3:59-63 (Winter 1989).
Byrd, Gary D. "An Economic `Commons' Tragedy for Research Libraries: Scholarly Journal Publishing and Price Trends." College and Research Libraries 51:184-195 (1990).
Carroll, Bonnie. "The Value of Information." Drexel Library Quarterly Summer 1985 p. 39-60
Cummings, Anthony et.al. University Libraries and Scholarly Communication: A Study Prepared for The Andrew W. Mellon Foundation. Washington, DC: Association of Research Libraries, 1992.
Cummings, Martin M. The Economics of Research Libraries. Washington DC: Council on Library Resources, 1986.
Description of Scientific and Technical Information in the United States: Current Status and Trends. (Draft Report).Jose-Marie Griffiths, Principal Investigator. Knoxville, TN: Center for Information Studies, University of Tennessee, July 1991.
"Economic Models for Networked Information," special issue of Serials Review 18:no.1-2 (1992).
Griffiths, Jose-Marie. "Existing Cost-Benefit Picture" in Exploration of a National Engineering Information Service. Ithaca, NY: Cornell Information Technologies and Media Services Printing, December 1992.
Information Economy in the U.S.: Its Effect on Libraries and Library Networks. Network Planning Paper No. 10. Washington DC: Library of Congress, 1985.
Kantor, P. "Levels of output related to cost of operation of scientific and technical libraries." Library Research 3: 141-54 (1981).
Kantor, P. "The NENGIS: Economic Aspects: Costs, Benefits, Sources" in Exploration of a National Engineering Information Service. Ithaca, NY: Cornell Information Technologies and Media Services Printing, December 1992.
Kantor, P. " Three studies of the economics of academic libraries". Advances in Library Administration and Organization 3 (1986): 221-286.
Katz, Richard N. "Academic Information Management at the Crossroads: Time Again to Review the Economics." Serials Review 18 no. 1-2: 41-44 (1992).
Katz, Richard N. "Financing New Information Access Paradigms, or Why Academic Information Managers Need Cost Models." Cause/Effect 15 no. 2:6-12 (Summer 1992).
Key Papers in the Economics of Information. White Plains, NY: Knowledge Industry Pubs., 1983.
King, Donald W. Statistical Indicators of Scientific and Technical Communication, 1960-1980; Volume II: A Research Report. Rockville, MD: King Research Inc., May 1976.
King, Donald W. Systems Analysis of Scientific and Technical Communications in the United States, Annex 4: The Cost Model. Rockville MD: King Research Inc., March 1978.
Lewis, David W. "Economics of the Scholarly Journal." College and Research Libraries 50 : 674-688 (1989).
Machlup, Fritz. Information Through the Printed Word: The Dissemination of Scholarly, Scientific and Intellectual Knowledge. New York: Praeger, 1978.
Machlup, Fritz. Knowledge: Its Creation, Distribution, and Economic Significance. Princeton NJ: Princeton University Press, 1980.
Machlup, Fritz. The Production and Distribution of Knowledge in the United States. Princeton, NJ: Princeton University Press, 1962.
Rep, Aatto. "The Value of Information: Approaches in Economics, Accounting and Management Science. " ASIS Journal 40 no. 2:68-85 (1989).
Rubin, Michael R. The Knowledge Industry in the United States, 1960-1980. Princeton NJ: Princeton University Press, 1986.
Rush, James. "In the Name of Access: The Economics of Information Dissemination." ASIS Bulletin :13-15 (December 1992/January 1993).
Spigai, Fran. "Information Pricing." Annual Review of Information Science and Technology :39-74 (1991).
Stroetmann, Karl A. "Management and Economic Issues of Library and Information Services: A Review of the German Literature." Journal of Information Science 17:161-173 (1991).
Taylor, Robert S., ed. Economics of Information Dissemination: A Symposium. Syracuse NY: School of Library Science, Syracuse University, 1973.
Taylor, Robert S. Value-Added Processes in Information Systems. Norwood, NJ: Ablex, 1986.
Tressel, George W. A Critical Review of Research Related to the Economics of the Scientific and Technical Information Industry. ED 111429, 1975.
Van House, Nancy. "Library Resources and Research Productivity in Science and Engineering: Report of a Pilot Study" in Communications in Support of Science and Engineering. Washington, DC: Council on Library Resources, August 1990.
West, Richard P. and Richard Katz. "Implementing the Vision: A Framework and Agenda for Investing in Academic Computing." EDUCOM Review :32-37 (Winter 1990).
ARL: a Bimonthly Newsletter of Research Library Issues and Actions:
161, March 2, 1992, "Among the Leading Geochemistry Journals," p. 13.
162, May 1, 1992, "Among the Leading Mass Spectrometry Journals," p. 11.
163, July 10, 1992, "Among the Leading Women's Studies Journals," p. 7.
S. C. Abrahams and R. A. Matula, "Crystallographic Publishing in Retrospect and Prospect," Acta Crystallographica a44 (1988): 401+
American Chemical Society, "ACS Journals: Price Comparisons," Washington, DC, May 1988 and other years (unpublished)
American Mathematical Society, "Survey of American Research Journals," Notices of the American Mathematical Society, 30 (1983): 715; 33 (1989): 287-291; 36 (1989): 1193 - 1198; 40 (1993): 1339.
American Physical Society, Unpublished Overheads presented to the ARL Scholarly Communications Committee by Harry Lustig, Treasurer, APS, at the ARL Membership Meeting, May 1992.
Jane Baldwin and W. M. Baldwin, III, "Let the Buyer Be Aware: High Price Does Not Necessarily Mean High Quality," Advances in Serials Management 3 (1989): 111-129.
Henry H. Barschall, "The Cost-Effectiveness of Physics Journals," Physics Today, July 1988: 56-59.
Henry H. Barschall, "The Cost of Physics Journals," Physics Today 39 (December 1986): 34+.
Henry H. Barschall and J. R. Arrington, "The Cost of Physics Journals: A Survey," American Physical Society. Bulletin, 33 (1988): 1437-1447.
John O. Christensen, "Cost of Chemistry Journals to One Academic Library, 1980-1990," Serials Review 18, no. 3 (Fall 1992): 19-34.
John O. Christensen, "Do We Know What We Are Paying For: A Comparison of Journal Subscription Costs," Serials Review 19, no. 2 (1993): 39-61.
John E. Costa, Arthur G. Sylvester, "The Crisis in Scientific Publication," GSA Today (January 1993): 13-15. (The first in a series of three pieces in 1993; See Molina below).
Economic Consulting Services, "A Study of Trends in Average Prices and Costs of Certain Serials Over Time," Washington, DC, Association of Research Libraries, March 1989.
"European Mathematical Council Survey of European Mathematical Periodicals," Notices of the American Mathematical Society 33 (1986): 905.
Lynn Fortney, etc., "Index Medicus Price Study," EBSCO Industries. Three reports in this series have been published as separate large-format booklets.
Kenneth E. Marks, Steven P. Nielsen, H. Craig Petersen, and Peter E. Wagner, "Longitudinal Study of Scientific Journal Prices in a Research Library," College & Research Libraries (March 1991): 125-138.
Martin Mulldorff, "A Survey of Statistical Journals," IMS Bulletin 21, no. 4 (July/August 1992): 399-407.
B. Meyers and J. L. Fleming, "Price Analysis and the Serials Situation: Trying to Solve an Age-Old Problem," Journal of Academic Librarianship 71 (1991): 86+.
Bruce F. Molina, Forum Column: "Issue: Geoscience Literature Pricing," GSA Today, March 1993: 58-60. (Part of a series; see Costa above)
Bruce F. Molina, Forum Column: "Issue: Geoscience Literature Pricing (Part 2)," GSA Today, April 1993: 93-94, 107. (Part of a series; see Costa above)
S. R. Moline, "The Influence of Subject, Publisher Type, and Quantity Published on Journal Prices," Journal of Academic Librarianship 15 (1989): 12+.
Roger Noll and W. Edward Steinmuller, "An Economic Analysis of Scientific Journal Prices: Preliminary Results," Serials Review (Spring/Summer 1992): 32-37.
Optical Society of America, "Price Comparison Study of Optics Periodical Literature 1984-1986," Prepared by Myers Consulting Services, 1988.
Periodiques de Chemie dans les Bibliotheques Universitaires," Bulletin Bibliotheques de France 32 (1987): 246+.
Paul H. Ribbe, "Assessment of Prestige and Price of Professional Publications," American Minerologist, 73 (1988): 449-469.
Paul H. Ribbe, "The Value of Serials: Relevance to Producers and Consumers of Scientific Journals," Proceedings of a Symposium of the Geological Information Society, Dallas, November 1990.
B. R. Spies, "The Effectiveness of Journals in Exploration Geophysics," Geophysics 45 (1991): 844+.