I endorse the archiving of raw data proposed by John R. Skoyles (1992).
1.3 I was surprised to learn how naive I was. It happened at a conference of embryonic electronic journals. One proposal was for an extremely rigorously refereed electronic journal in a clinical field. I don't recall the exact question, but the reply was something like, "Of course we won't transmit the raw data." Pressed, the publisher (or another participant) pointed out to the naifs in the room that clinicians spend years amassing data and don't want to hand their life's work to someone else who could then "just do an analysis" and get publication credit without doing all the work. The explanation helped me understand those motives for secrecy, but I was still disappointed.
1.4 It seemed to me then, as it does now, that the better scientists would not want to keep their data secret. Secrecy smacks of patents, of profits, of trying to own ideas. I have always imagined that "real" science involved sharing what one had learned or couldn't figure out. I still believe that one vector in the "origin" of modern science was the novel ability to share the (comparable) data one had gathered.
1.5 So I am delighted to hear Skoyles's proposal for data archiving. I hope that it can be implemented, that it will be greeted warmly, and that people who don't want their data archived will find it increasingly difficult to get their interpretations published.
Skoyles, John R. (1992) Public Electronic Archiving and Retrieval of Raw Scientific Data. PSYCOLOQUY 3(29) data-archive.1