I. MISSION FOR ELECTRONIC PUBLISHING

The mission statement for AGU's electronic publishing program is:

To use advances in communications, networking, and other relevant computer technology (1) to enhance the value of and access to AGU's publications for individual scientists today and in the future and (2) to assure AGU can continue to serve researchers' need for information either directly or through libraries and other organizations acting as intermediaries.

II. ENVIRONMENT FOR ELECTRONIC PUBLISHING

AGU's Overall Program Mission and Goals

The basic premises on which AGU's publications program rest will not change as a result of the opportunities and challenges presented by advanced technology. The overall mission for publications in all media will continue to be "To publish any scientific research of sufficiently high merit in the geophysical sciences."

The foundation stones for AGU's publications program will continue to be as follows:

AGU publications exist to disseminate timely information to the geophysics community and to researchers and students in related scientific fields. To achieve their objectives, AGU journals, books, and other publication products must be placed in the hands of individual geophysical scientists for their personal use.

AGU publications must also be on file in libraries where they can be accessed today by engineers and scientists outside the community of geophysics and related sciences and where they will be available for future generations.

AGU publications are to be designed and operated in a way to attract the highest-quality papers.

AGU products and services delivered in electronic format are expected to adhere to the quality, timeliness, and financial goals of the overall publications program.

The Electronic Publishing Environment in 1997

Over the next 5 years the power and availability of hardware and software that can be used in creating and disseminating publications will increase.

By 1997, we might expect the following conditions to prevail.

Connections and Networking
The Internet will reign supreme throughout the world. Everyone will have connections, even the small institutions. For those few people who are not at institutions with Internet connections, access will be available via phone lines.

Hardware
The growing power and falling prices of computer hardware will make it possible for most scientists to have the equivalent of our present workstations available on their desks. Everyone will have access to the equivalent of a 386 machine with VGA graphics. All institutions will have a PostScript printer available.

TeX with Macro Package
At least one standard macro package will be in use for the submission of manuscripts. We can assume the first standard [for AIP and its member societies] will be some extension of ReVTeX. Other standards will be able to gain acceptance only if they have a large user base or their developers make a translation package. It is likely that at least one full screen editor will be accepted as a standard, too.

Graphics Standard
The scientific community will have standardized the means for transmission of graphics. There will probably be a few accepted standards. Embedded Postscript files will certainly be one of these standards. [This technique is in widespread use now for transmission of data to and from the Space Telescope Science Institute.]

Data Storage Systems
At least in astronomy and possibly in other fields, there will be a well-established set of distributed data bases supported by NASA (and probably NSF) in which public domain observational data will be stored. The data will be accessible over the Internet.

Data Related to Journal Articles
Data "banks" will be able to store information from published articles, including data tables, algorithms, lists of objects, and other subsidiary data.

Indexing, Retrieval and Browsing Tools
There will be an entirely new suite of software available to help perform the search and retrieval functions. It will be possible to select and retrieve journal articles with the help of intelligent algorithms that will include full text searches and go far beyond present day indexing and keyword schemes.

The preceding predictions are based on a document prepared by Peter Boyce of the American Astronomical Society and Judy Holoviak, AGU, for the joint task force of staff members of AIP and its member societies. Members of the task force generally agreed with this picture of the tools we will have to work with by 1997, although some thought the time frame might differ a bit for different societies. The predictions are most applicable to industrialized nations. Because of the worldwide nature of geophysics, AGU might expect a greater degree of duality in terms of those with these computer tools and those without.

Competitive Factors

Because of the seeming ease of dissemination in electronic form, there will be many who want to distribute what we now know as journal articles. Some of the factors we will need to deal with include:

self-publishing, which bypasses the review system and which may not be adequately archived for the future.

the many different kinds of document delivery services, which are springing up like mushrooms. (Several subscriptions agencies and libraries are now in the business. All want the right to store articles electronically for future delivery and provide royalty income to the publisher.)

commercial publishers with deep pockets, who are now actively engaging in experiments with electronic delivery on different campuses.

government agencies, who are anxious to use their computer might to serve science but may trample the source of good journal publishing in the process.

libraries, who want to be the direct link between author and user.

on-demand printing suppliers.

nontraditional publishers, who have the hardware and software capabilities and have heard that journal publishing is lucrative.

AGU will need to spend time monitoring these new sources of competition and taking appropriate actions to assure the Union can continue to fulfill its mission.

Legal, Legislative, and Regulatory Considerations

Changes in legislation affecting publications, concerns about copyright, and various governmental regulations have always been of concern to AGU. The new media simply add some new considerations and possible complexities, such as those mentioned below:

Copyright
The role of copyright is under considerable discussion with respect to electronic publishing. Some individuals and organizations see copyright going away; some see it staying in place. Some "experts" see the need for publishers to obtain rights for "performance" if works are multimedia. Copyright will not be a simple issue.

AGU has been a strong advocate for copyright as an important tool in dissemination. We do not see our position changing, but we do see the need to stay current with the debates and any changes to the law.

We also expect to be concerned with issues related to integrity of information that is delivered via the network. What role can/should copyright play?

NREN (National Research and Education Network)
Scientific societies need to represent the interests of their members and the concerns of society programs as planning, funding, and implementation of NREN occur.

Leasing Arrangements
AGU will not create most of the software that will be required for electronic products we produce. We will need to license any software we wish to provide with our electronic publications.

We will also need to develop new types of agreements with "purchasers."

There will be fewer simple buy/sell arrangements.

Some Financial Considerations

Many proponents of electronic publication in the library world expect tremendous cost savings and therefore major decreases in their subscription costs.

These arguments are actually rather naive and unrealistic. Most publishing costs will not go away and new ones will be added. The following financial considerations will be shaping the environment for AGU:

1. In journal publication about 80% of the cost of publication arises from editorial selection, copyediting, and preparing material for printing. These costs are will remain in electronic publishing.
2. New costs will be added for improved indexing and other tools needed for retrieval.
3. While we are in a mode of providing both print and electronic versions of the same publication, there will be minimal savings on the print side (possibly less paper and postage) but increased costs for the dual electronic production.
4. We must plan and pay for systems that will keep the digital files in good condition. (We all know horror stories about magnetic tapes that have deteriorated beyond use.) We must also consider the expense of shifting materials to new platforms.
5. It is likely that some payments will shift from up-front subscriptions to pay-for-use. This is a radical change in the economic picture. Libraries expecting to see savings from such shift are overlooking the fact that they will have to pay something just to have material available.
6. As mentioned under competitive factors, the document delivery services would expect us to live on royalties. Some countries have fixed systems on royalty payments that make such a future look bleak indeed. Payments cannot be designated to any given work and therefore do not flow back to the publisher who invested in its production.
7. The financial risks increase as we move into new territory. The up-front costs of CD-ROM products, for example, are very high and developing test products can be costly. It may be difficult to get good marketing information without trying some experiments. We will need to budget for some experimentation.
8. Shifting to new production methods may involve significant capital expenditures.
9. Currently, 60% or more of the pages AGU publishes are provided as author-produced copy. AGU incurs significantly lower costs in producing these pages; however, we have no digital file of these papers. AGU is likely to incur some added costs when working from an author's electronic files. We need to examine the impact of switching from the current low-tech author-produced copy to electronically submitted material.

III. STRATEGIES AGU EXPECTS TO EMPLOY

The following strategies are intended to be complementary, and none is exclusive. Some strategies will have greater relevance than others, depending on the part of the community being served.
To position AGU solidly at the forefront of electronic publishing. This means that we must take an aggressive posture and may need to be a pioneer in areas of key importance to our constituents.
To develop and implement experiments in electronic publishing in which emphasis is placed on products and services for which there are no print antecedents. At the same time AGU must remain cognizant of the fact that some of our constituents will not be able to participate in services or use products that are solely digital.
To develop novel means for electronic delivery of current and future print products, that could ultimately effect a transition from print format to electronic format as the means of delivery. AGU must, however, recognize that for the foreseeable future the Union will need to provide both print and digital formats of current print products.
To move toward having increasing percentages of the material for AGU's publications be submitted in standard digital formats in order for AGU to have the greatest flexibility in organizing, producing, and disseminating publication products and services. This will mean AGU must choose, conform to, and promulgate appropriate standards. It may also mean that AGU will have to take the lead with other interested societies in developing standards where none exist or are likely to be developed in the near term (such as in the digital representation of maps).
To place priority on providing value-enhanced publication services, such as access to the data underlying figures, manipulable mathematical expressions, links to external data bases, and three-dimensional or moving representations of data.
To develop techniques, products, and services that will help individuals find and easily retrieve what they need from the published literature.
To make participation in AGU publications easier for authors by providing templates and other such tools for preparing materials to meet AGU standards. This may mean adopting tools that are readily available from other sources or developing them either alone or in cooperation with other interested societies.
To work closely with other related societies in developing the directions of electronic publication.
To assure that appropriate steps are taken to safeguard and extend the useful lifetime of material distributed in electronic form. This may require providing technological and financial means for protecting masters and for moving information and software to new platforms or distribution media.
To improve the preservation of and access to AGU's printed publications so that as the paper disintegrates the literature that should be retained will continue to be available. As acidic paper becomes too brittle for use, AGU needs to decide how we will fulfill our goals of making our publications available and to what extent digital formats can and should be employed.

IV. TYPES OF PRODUCTS

The directions and strategies outlined in this document cover the formal publications of the Union. Although some of these same approaches may be applicable for informal exchanges, they are not the main concern of this paper.

AGU Electronic Publication Products will include the following:

• Refereed journal and book articles and accompanying materials approved for publication
• Monographic books
• Software
• Maps, data sets, and organized bibliographic, textual, numeric, and graphic data bases, which are intended to be stand-alone products.
• News publications that have been developed and produced in accord with AGU standards

Some of these products will appear in print formats as well. The electronic versions of such print versions will be intended to provide additional capabilities for the user. They will not simply be a visual representation of the printed page on a computer screen.

V. OPERATIONAL TARGETS

The Publications Committee has identified several specific projects for the next 2-4 years. Others will likely be added. Some may move forward or backward in time depending on identification of member needs, technological advances, and financial considerations.

The list of projects and services that help define where AGU is going in electronic publishing is outlined below.

1. JGR 4 Years Hence
• a fully electronic and searchable "executive summary" (in keeping with the new paradigm being considered for journal articles)
• body of the paper as an electronic image (at the minimum) but not necessarily or consistently digitized and searchable
• tables electronically accessible and usable in numeric form
• for some graphs the underlying data available electronically and usable in numeric form
2. Meeting Abstracts 4 Years Hence
• available electronically in advance of the meeting
• submission from author in electronic format
• information about date and time of presentation provided in an electronic format that can be manipulated by the user for producing customized programs
3. New, Fully Electronic Journal
• may be developed in conjunction with other societies (such discussions are currently underway to see whether a multi-society journal proposal can be successfully developed and implemented)
• no print antecedent or companion
• will test many of the concepts of value-enhanced publications: access to external data bases; manipulable math; three-dimensional and moving graphics
• will test author and reader acceptance
• will help point way for possible transitions of existing print journals.
4. Maps and Similar Data in Digital Format

One such product is in development now and two digital maps of the Global Geoscience Transect (GGT) project are planned to be released in 1993.

Digital GGT
• software for creating or using the data files to be available free if FTP or at small fee if diskette
• printed handbook and sample data files to be used in classroom—priced similarly to other AGU books
• data files to be used with software—priced for individuals and institutions
• on-demand printed versions to be sold by AGU
• data files to be used with other computer models
Other Map Products
• none in planning stage yet
• agreed that future maps published by AGU should be available exclusively or alternately in electronic format
5. Templates for Author-Prepared Copy
• LATEX macros available by 1993 Spring Meeting
• WordPerfect and Word templates planned
6. Abstracting and Indexing Services
• EASI (Earth and Space Index)
• available April 1993 (for 1991 & 1992) as diskette
• formatted data base covering books, journals, translations, and Eos
• commercially available search software provided with index; product also allows some manipulation of data to create different reference styles
• by January 1994 include material back to 1988 and through 1993; monthly online updates planned for 1994 and forward
• subscription-based downloadable product contemplated as an alternate to diskette
• may expand beyond AGU publications
Abstract Service
• proposal under consideration
• may have print companion
• selected rather than comprehensive service; emphasis on quality and currency
7. Rapid Delivery of Eos
• time-sensitive material in Eos delivered in electronic form
• plans to be developed in 1993
8. Personalized Information Services
• members to specify a "retrieval profile"
• system based on AGU's index set for providing information matching the profile
• information available on or through Kosmos matching that profile to be delivered to user
• ultimately link to a document delivery service for AGU articles