Scientific papers have normally been exposed to peer review before publication as a guarantee of quality and originality. However, the delay entailed can slow down the rate at which a specific area is able to develop. In addition, the more specialized publications have to pass on their costs to a smaller target audience. This can lift their price to a point where they are too expensive for a generalized library, even in the US, to acquire and makes them relatively inaccessible in many parts of the world.
To address some these concerns, it was not uncommon for pre-prints to be made available prior to completion of the peer review process. Turning this stage into an electronic form, which could be easily e-mailed, was only one small step further and to put these into an archive available to all was the final stage. Starting in the early 1990s, the physics community (led by Paul Ginsberg at the Los Alamos laboratory), has led the way in introducing electronic publishing on the web supported by the U.S. National Science Foundation (try their URL at xxx.lanl.gov). This has been so successful (25,000 new papers annually) that proposals are under active discussion in the US to extend it to other disciplines. Nature (the UK journal) provides a running commentary on the rapid developments taking place in this area which is clearly the forerunner of massive global changes in technical and other publishing value chains.
Nature Vol. 397, 21st January 1999 pp. 195- 199
Nature Vol. 395, 10th September 1998
Varian and Shapiro Information Rules: A Strategic Guide to the Network Economy McGraw Hill 1998