Preprints provide an opportunity for astronomers to share their results quickly with the astronomical community, in the six to nine month period between acceptance of a paper for publication and appearance of the paper in the journal. Preprint distribution schemes vary widely, from sending preprints only upon request, to sending preprints to all astronomical libraries plus a list of 50 or 100 astronomers who might remotely have some interest in the subject of one's paper. The preprint is a publishing phenomenon which is likely to persist as long as there is a many month delay between acceptance and publication of a journal paper, although in future some or all preprints may be distributed electronically rather than on paper.
A subcommittee of the AAS Publications Board, chaired by myself, was charged with examining the format of the AAS/WGAS Latex Macros package which produces preprints. I prepared a draft document which was reviewed by the preprint subcommittee at the June 1992 AAS meeting. This revised version incorporates the suggestions made then.
At the May 1992 Publications Board meeting, it was felt that the current preprint format generated by the AAS/WGAS Latex Macros, using the aaspp.sty style file, was unacceptable and that changes were desired. Certain essential characteristics of a preprint were agreed upon. First, the preprint should not mimic a page of one particular astronomical journal, such as the Astrophysical Journal. This slights the other journals such as the Astronomical Journal and the Publications of the Astronomical Society of the Pacific, and furthermore blurs the distinction between the preprint and the final published version of the paper in the journal. Second, the preprint should be easily read. Other features also seem desirable, such as compactness to save trees and mailing charges, the ability to include figures and figure captions gracefully, and some flexibility so that more than one style of preprint could be accommodated for those with very divergent tastes. Additionally, there is a clear need for something which will estimate how many pages a given text file will produce in the Astrophysical Journal Letters, due to the stringent page limit of that journal. This estimate of the number of ApJ Letters pages might be possible to include as part of the output of a preprint formatter, although other approaches may be more desirable, such as a separate page length estimator program.