Question: do you think revision is worthwhile ?

Daniel Mietchen answered on 15 Jun 2010:

Many things can be revised, and in science, it is often a good idea to do so from time to time. However, my interpretation of “worthwhile” would include the way in which the revisions are performed and recorded, and sending around documents as email attachments is less efficient than having a central copy of the document that can be edited by all the relevant people (wikis are one example; software developers know and use a variety of such tools).

In science, the two central kinds of documents are manuscripts submitted for publication of research already performed, and research proposals submitted in the hope of obtaining funding for further research. These documents often receive incredible amounts of revisions, yet almost none of this process is visible to the world, which I think is inefficient.

The system has started to change, though, as some journals have opened up the process. For instance, some of my manuscripts are freely available online as a discussion paper (how we submitted it), along with the video supplement (that can be more conveniently viewed here), the interactive discussion (in which external reviewers commented on the quality, scope and presentation of our research) and the final revised paper. There are even some journals which have adopted a wiki model for revisions – Scholarpedia, for instance. And the more this model spreads, the more interesting it becomes to pay attention to the revisions, as compared to the article itself. The history of the article Magnetic Resonance Imaging, for example, reveals that Nobel Laureate Paul Lauterbur accepted the invitation to write this article one day before he died. One of his colleagues completed the article, an anonymous reviewer commented, and everyone can make suggestions on how to further improve (or simply update) it.