I was recently commissioned to write a short piece about the adoption of social media and networking tools by scientists for The Euroscientist. I briefly covered various aspects of the evolution of the web and the notion of a Facebook for Science. Of course, there are lots of networks out there and both Twitter and FriendFeed are being used to boost the interactivity of scientific conferences, but until we get to at least web 2.1 and the next step towards the semantic web (3.0), it doesn’t seem that the majority of scientists are actually taking to these tools.
I asked Brian Krueger of Labspaces for his thoughts on the topic and he was inspired to produce a fully fledged blog post.
“I think for scientists to be fully engaged with the online world, they need to see something in it for them – something to change the way they currently do things,” he told me. “In other words, the online world has to offer them something they need and want. I think some sites and services like Mendeley [dubbed a last.fm for research papers] and perhaps my own site, ResearchBlogging.org are helping them see that. But much of what is currently being done online doesn’t directly help scientists do what they need to do, so they’re not engaging with it.”
Munger suggests that peer review might be the next area of revolution. “The peer review system as it stands is antiquated, slow, and ungainly,” he explains. “I think a system of post-publication review might work best. Scientists would publish their work as soon as they felt it was ready for broader consumption. Then they would submit to ‘publishers’ who would ensure that the work was reviewed, placing it in the appropriate ‘journal’ based on the research’s importance and quality.”
Munger suggests that this approach, rather than submitting to and getting reviewed by several journals in the hope of being placed in the most prestigious spot, would mean that each article would only be reviewed once, thus saving time and resources.
“Reviewers wouldn’t be asked to comment on so many manuscripts, and authors wouldn’t have to be constantly resubmitting,” he says. “PLoS is taking a leadership role in this post-publication review movement.”
In order for such a system to work it would have to be transparent and open so that readers could know why a particular article was selected for a particular journal. “This in turn will lead to even more online engagement as reviewers respond to each other, and readers and authors respond to the reviews,” adds Munger.
Open notebook chemist Jean-Claude Bradley (no relation) of Drexel University believes my couching this post in such a way is a case of writing from the perspective of the glass being half empty vs half full. “You can argue any situation from the standpoint that it is a failure because adoption is below some arbitrary number,” he told me. “We could argue that science itself is failing because most of the population are not scientists, etc.”
JC Bradley is very optimistic about the direction that Open Science is taking, indeed he believes more and more people are getting involved, even if the percentages have not increased beyond some arbitrary value. “I think that more people will become more open despite not necessarily agreeing with the philosophy as their hand is forced,” he adds and offers an example in a blogpost entitled Secrecy in Astronomy and the Open Science Ratchet.
He does concede that there are still obstacles to widespread adoption. “Right now, the bottleneck is just finding people who are willing to share, giving them simple tools and helping them,” he says. “I don’t think technology is the limiting factor. Having semantics magically appear without any effort whatsoever isn’t going to happen soon – even if such systems did exist someone still has to verify the information and correct it. But with a very tiny amount of effort it is possible to abstract semantic information from organic chemistry Open Notebooks – even if the people recording the notebook don’t do the abstracting – we’re doing that for the Todd group for example and putting their info into Reaction Attempts database. JC Bradley adds that, “Every field will have to solve the problem of easily representing their data in semantically meaningful ways but the key factor is finding people willing to share.