BE Board:Dinner Discussion/Alternate publishing models

Currently, much of the discussion around scientific publication centers around the open access debate. While this is a critical problem in scientific publication, there are several others drawbacks to the publication process that also need to addressed. In this dinner discussion, we will go over some of those shortcomings and discuss approaches to solving them.

Shortcomings of the current publishing model

1. Single, slow channel of publication
   There is only one main level of information dissemination: peer-reviewed, highly polished summaries of work in journals. These papers generally take months or years to write, submit, review and publish.
   • Published results are likely 1-2 years old due to the long publication process.
   • The high barriers to publication prevent people from reporting uninteresting or failed experiments. Thus, there is needless duplication of effort across labs.
2. Few documented evaluations of articles
   Current publishing approaches do not encourage sytematic, open evaluation of published works.
   • It would be helpful to be able to read not only an article itself but also reviews and discussion of the article by other scientists.
3. Failure to take advantage of the online medium
   Journals articles are edited and formatted for a print medium while being distributed primarily online. They fail to take advantage of the features offered by the web.
   • Articles often lack the detailed methods information necessary to reproduce the results of the work.
   • There is a near absence of hyperlinks in journal articles.
   • Data is often not made available in a machine-understandable format that facilitates others to use the dataset independently.
   • Periodic journal issue as a means of aggregating information is outdated. Papers should be collected based on relevance to a specific topic rather than date of publication within a broad topic area.

Proposed improvments

1. Mechanisms for short-timescale publication
   • There are an abundance of technologies that permit people to easily and quickly distribute information over the web like wiki's, RSS feeds and blogs. The research community needs to take advantage of and value these alternative means of information dissemination.
2. Post-publication discussion and evaluation
   • A mechanism by which scientists could post summaries/reviews/critiques/comments about a given paper. By enabling users to rank these posts, worthwhile commentary should rise to the top of the pool. Such a system can be developed as a separate layer from the existing publishing system. The Faculty of 1000 is an example of an attempt at a site of this form but restricts access.
3. Novel publication formats
   • While research articles are a useful mechanism for communicating your work, the article format needs to use the advantages of online content. Articles could be treelike in setup such that readers could follow links in order to read more and more detail about a given piece of work. Thus, an article could be written on multiple levels: for everyone between the general scientific audience to the novice researcher who wants to recapitulate the experiment.

Barriers to change

The reward system in science, and to a lesser extent engineering, only values traditional publications. In order for the publishing system to change, institutions must value not only publications but other types of contributions during the hiring, promotion and tenure process. What can we do to push such change forward?

References

1. Open Networks and Open Society: The Relationship between Freedom, Law, and Technology. Talk by Hal Abelson and John Wilbanks.
2. ScienceCommons
3. Science 2.0/Brainstorming
4. Publishing Group
5. Nature Peer Review 'Debate'
6. comments from the ArXiv.org folks