The Dawn of the Hyperpaper

Copyright: Astronomical Society of the Pacific

Peter B. Boyce

American Astronomical Society


Computer networking is changing the way that astronomers communicate, and is doing so faster and more radically than most had dared to predict. But one thing has remained certain: The peer-reviewed journal, or its electronic equivalent, is as needed as ever.



According to the accepted lore, one needs a crystal ball to peer into the future and see what's in store. But the view in a crystal ball is dim, and different people see different scenes in the same ball. So why even try to look into the future? The answer to this, at least, is clear: The world of astronomy is changing rapidly, and we need to figure out where we're going so that we can plan how to get there. The discipline has new instruments that produce information faster than ever before. There are more astronomers than at any other time in history, and they are writing more scientific papers. Computer networks are connecting the world together, transforming how astronomers observe, compute, and collaborate.

Astronomical publishing is changing, too. But how? What will astronomers be doing in 10 years, or even five, to communicate their research results to their colleagues? It's an important question for several reasons, starting with the sheer size of our publishing enterprise. The Astrophysical Journal is expected to run to nearly 25,000 pages this year, or 120 typeset pages produced every working day. Add to that the 10,000 pages of the Astronomical Journal and the Publications of the ASP, plus the more specialized journals such as Icarus, and it's clear that publishing is a large component of astronomy in North America.

For 16 years it has been my job to anticipate the future and to prepare the American Astronomical Society to serve the changing needs of the astronomical community. The AAS, like the ASP, was founded to facilitate communication among its members. Publishing is three-quarters of the AAS budget; a major mistake would have a huge impact upon the finances of the society, not to mention the vigor of astronomical research. Over the past three years, I have been gazing into that cloudy crystal ball a lot.

Three years ago the AAS set out to plan the future (or see the October 1992 AAS Newsletter) and, since then, our electronic publishing plan has evolved just as we expected, with one exception: The pace has been faster. In 1992, when we developed the plan, the Mosaic graphical browser for the World Wide Web had just been introduced, and the crystal ball was as clear as it had ever been. Mosaic's ability to transmit a document that mixed graphics and text was what we had been waiting for. Mosaic made it possible to transmit scientific papers complete with complex mathematics, line drawings, and pictures. And it was free. Developed by high-energy physicists as a tool for aiding collaboration over the computer networks, Mosaic adhered to an open standard that anyone could use. This availability has propelled the Internet far beyond the computer cognoscenti.



The Web Rules

Four years ago, our electronic-publishing team could see the World Wide Web being used within five years to distribute scientific publications electronically. We could see astronomers busy at terminals, reading articles, searching an electronic database for papers on a specific topic, linking directly to referenced articles, looking at the original observational data, and generally using the web's capabilities for seamlessly accessing information distributed around the world. Many other people had a more limited vision of the future. They saw astronomers using their terminals to read articles that looked exactly like the ones in printed journals. The least imaginative visions typically came from people already engaged in producing the paper journals. It was hard for them to see beyond the paper format and the functions they were used to. Authors seemed concerned only with electronically preparing and submitting papers. Readers concentrated on how to search for papers on a given topic. Those who lacked Internet experience had trouble envisioning how a world tied together by the network would function. In particular, it was surprising how often the CD-rom was mentioned as the ultimate electronic medium. By contrast, our team c oncluded that the CD-rom is too limited to be useful; it does not give enough access to enough material to allow effective searches.

As we progressed with our project, it was hard to reconcile the narrow views with the exciting possibilities. We had assembled a team with diverse backgrounds and views, gathered information from many sources, and kept an open mind about the communities involved. In the end, however, nothing could beat reality. Taking small, experimental steps and incorporating feedback from people actually using the system was critical in developing radical technology. Last fall, the Electronic Astrophysical Journal Letters became a reality.

The world now has one astronomical journal on-line and, by all reports, it works well. Within a year, the electronic version of the full Ap.J., all 2,000 pages per month, will be on-line. By January 1997, the transition will be complete: Readers will have to pay for access. [Features] Subscribers not only will have access to the journal; they will also be able to develop personalized, intelligent search profiles to alert them of articles on specific subjects. Bibliographic references from the articles will link automatically to the Astrophysics Data System so that the abstracts and, for AAS journals, the full text will come up on readers' screens. Forward-reference links will bring up, at the click of the mouse, articles that cite the one being read; these links will be added to the original article as relevant articles are published. To see how this works, move to the botom of this page and click on "References to this article."



Continuum Emission

This kind of interconnectivity is the basis of the journal of the future. This is yet not appreciated by most astronomers. Most have not used a true electronic journal with effective links to the world's literature and data. Those who have are coming to understand how the immediacy of the electronic world is changing how we do science. Data and literature are becoming available equally to small and large institutions. We will soon be able to do good science from anywhere that is connected to the Internet. Electronic journals are erasing the divisions among observations, interpretation, intermediate tabular data, and finished literature. Instead of having discrete components, the process is becoming a continuum.

There will be additional developments over the next two years: the inclusion of motion-picture clips in electronic journals, making it possible to display computer simulations; 3-D presentations under the control of the reader; and active mathematical formulas whose underlying algorithms can be run by the reader.

Some scientists profess to see a future in which authors can write as much as they want and post their work to electronic preprint databases, such as the Los Alamos preprint server (http://xxx.lanl.gov). On the server, their papers receive public comments and their worth can be judged [see "Peering into Peer Review," November/December 1995, p. 32]. While this system is useful for informal, rapid communication of new results, I do not believe it will supplant the peer-reviewed scholarly journal. Consider the myriad ways in which journals are used. A student goes to the library to look for the definitive paper on a certain subject, finding it among the many that chronicle the history of modern astronomy. A historian uses the library to find, say, a paper from the 1920s describing observations that launched a whole area of astronomy. Journals, in other words, are the record of a body of knowledge. This is one of their functions that hasn't changed significantly over the last 70 years. Or consider how deans and granting agencies weigh candidates. Often they look at the articles published in high-quality, peer-reviewed journals, neglecting the conference proceedings and other unrefereed publications. You can say nearly anything at a conference I've done it myself but the referees and editors will hold you to a higher standard when you publish in the Ap.J., the A.J., or PASP. Papers in refereed journals are one measure of the quality and productivity of a research scientist.

These are among the roles filled now by the paper journals and by preprints, and those roles won't change or disappear just because the medium changes. It's a sure bet that whatever form publishing takes at the turn of the millennium, it will have to fulfill the need for rapid, accurate communication; for archiving the body of scientific knowledge; and for providing some means of evaluating an author's standing in the community.



PETER B. BOYCE is a senior associate at the American Astronomical Society in Washington, D.C. He was Executive Officerr of the AAS from 1979 to 1995. His email address is pboyce@aas.org.


[Return to Top]


Features:

The paper version of this article in Mercury included a figure which helped to illustrate the the navigational links built into the articles. This electronic version can show the concepts live.

Two specially useful navigational features are :

1. The ability to link from the figues directly into the text where the figure is discussed. Some users say that you can start with the pictures and jump into the discussion from there more easily than you can scan the pages of the paper version.

2. Links from references to the abstracts (and sometimes the full text) of the referenced articles.

[Return to text]




This article originally appeared in Mercury, 25, 28, 1996.
Mercury is the bi-monthly membership magazine of the Astronomical Society of the Pacific.
[Return to Top]