Science publishing on the fast lane, plus optionally in journals

About two weeks ago, PLoS, Google Knol and NCBI announced a potentially groundbreaking collaboration: PLoS Currents — a new platform within Knol and mirrored at NCBI — allows for rapid submission of research results to the eyes of the public prior to, or possibly instead of, formal publication.

The idea is not new — arxiv.org has been operating a preprint repository for almost two decades in the TeX-based sciences, and Nature Precedings for about two years in the remaining scientific fields. What is new here is the combination of preprints with an encyclopedia — Knol — and its embedding in the framework of a larger repository — Rapid Research Notes, operated by NCBI, the computational arm of the NIH — which is open for other publishers to join if they so wish. This way, a systematic record-keeping of information that has traditionally been transmitted only via meetings and conferences is now on the horizon.

The initiative is timely — Knol was launched last summer but the laudations around its first anniversary could easily be mistaken for obituaries (mainly because it failed to rival Wikipedia the way many had expected), while experiments with the coupling of Wikipedia contributions and formal publication have now been going on for more than half a year at the journal RNA Biology. It is also timely because indications accrue that the current scientific publishing landscape might change dramatically soon.

People at PLoS do not seem to embrace the role of merely observing these developments — they prefer to help them take shape. From this perspective, it is not surprising that PLoS would venture into wiki-like waters, though Knol is certainly not the only option for such activities.

Specifically, data do not currently play a prominent role at Knol (though integration with Google Wave might dramatically change this), but they do so at OpenWetWare, a wiki platform where researchers keep their electronic lab notebooks online and in public. Wouldn’t it be natural to write up “Rapid Research News” in the environment which hosts the data, and to link there from any scientific or popular article on the subject? Both of these platforms lack, however, integration between the individual research contributions and with existing knowledge: The introductory sections of the initial PLoS Currents collection on Influenza A H1N1 basically all link to different (and usually not updatable) references instead of one introductory article on the topic that could easily be drafted, hosted and updated in the same environment, as other wikis  — especially the Wikipedias — have shown.

In that respect, it is perhaps of note that scholarly wiki environments already exist which grow their articles in context. Scholarpedia, for instance (a wiki with anonymously peer-reviewed articles and an ISSN), is expanding from a computational neuroscience core, while WikiGenes links genes with their functions, and Citizendium places special emphasis on context by using semantically Related Articles for site navigation (similar to the “scope” of traditional journals, or to the Frontiers Distillation System).

This way of structuring knowledge may also have implications for research funding: Suppose a topic is covered on a platform like this such that all the relevant current knowledge on it is contained therein but no subtopics (in Citizendium’s parlance) exist which would merit their own article — wouldn’t this provide for a mechanism to identify areas for future research that is much simpler (and more transparent) than what we have today? Instead of researchers spending their time iteratively writing and reviewing grant applications, they would concentrate on keeping a shared body of knowledge (or knowledge exchange system) up to date in their fields of expertise, while funders would look for “missing subtopics” or “missing recent updates” they could tag with the amount of funds they are prepared to invest there. Distribution of these funds to the appropriate researchers could then be achieved by posting bids to and discussing them on public platforms like FundScience (which hosts this blog), using prior contributions to the whole knowledge exchange system (e.g. data, interpretation, theory, incorporation into scientific context) as a measure of merit instead of the infamous Journal Impact Factor that dominates such decisions today.

Where are the journals in that system? At least not in the very prominent position they are occupying now. Given that PLoS ONE is on track to become the world’s largest scientific journal in about a year or so, it will be especially interesting to see what impact its combination with wiki-style interactive research environments like PLoS Currents might have on science communication as a whole.