The accepted system for exchanging scientific information is publishing in a prestigious scientific journal. That journal system has been around since 1665 – it is slow and rewards work unevenly. It’s only survived because it provides a fairly reliable means (peer review) of knowing which scientific papers are worth believing. What we need is a system to bring that peer review online and into real time. The faster scientists can share with and critique each other, the faster scientific progress proceeds. Many have tried to bring social networking to scientists thinking that such a forum could upgrade peer review for the 21st century, but these networks have failed. Why? The old journal system promises money and glory, new systems seem dangerous and untested. Fear that the open exchange of data and online peer review will keep scientists from receiving credit and rewards for their work is keeping the scientific community in the last century. For now.
Competitive vs. Collaborative
Let’s imagine you’re a TV executive working on the next big prime time comedy. You have a lot of money, time, and respect tied up in making that sitcom a success. Are you going to want to share your idea with the executives at the other TV networks? Of course not, they might throw something together and scoop your idea. (There’s a Bewitched vs. I Dream of Jeannie reference here somewhere).
Science is a lot more like prime time TV thank you think.
Scientists scoop other scientists all the time. ALL THE TIME. Occasionally it’s done with malice, often it’s a coincidence, but mostly it happens when one scientist learns something about someone else’s work and POW! – suddenly they have an insight into the field. Ideas are contagious, and the scientific community is constantly coughing on one another.
If you’re a scientist you probably really want to share ideas. To collaborate with colleagues. After all, the best ideas are often shared, debated, and honed through conversation. But you don’t want to share too many details, you might get scooped. The safest way (for a scientist’s career) to share ideas is through peer reviewed journals because they provide a definitive means to say “I was here first.” So that’s what scientists use.
In many fields, there is a winner-take-all reward system. Get published in the best journal the quickest and you can get more funding, more citations, more prestige. Get published second and you can be out of luck. Encouraging scientists to publish quickly may seem like a way to insure quick scientific progress. The incentives to work hard are definitely there. But this system marginalizes the work of many scientists to bring focus on the work of a few.
You see, the reward mechanism in the scientific journal system is self perpetuating. Those who are published most often get rewarded with the means (funding, citations, connections) to get published more often. Journal success is skewed to those who have already succeeded. Past success may be a good indicator of future success, but it’s no guarantee. We tend to lose the contributions of those scientists who didn’t make the first cut, and that’s the majority of scientists out there.
Trying to step out of this system (not publishing in major journals) is paramount to career suicide because journal publications are the major way other scientists have to learn about your work.
The other problem with the journal system is that journals are an incredibly slow way to communicate. It can take years for work to be submitted, reviewed, revised, resubmitted, approved, published, and then responded to. Inside a lab, where collaboration reigns, scientists can share data in real time and get things done quickly. Between labs, where competition reigns, scientists share data through the slow journal system in order to protect their work. It’s like having the fastest super computers in the world communicating through dial up modems. It doesn’t make sense.
Social Networking for Scientists, A Valiant Effort, an Epic Fail
Social Networking sites for Scientists (SNSS) could allow scientists to move the peer review process online and into real time. It would accelerate the exchange of data and speed up scientific progress. And it would still allow professionals to share scientific data outside the disruptive (and typically under informed) scrutiny of the public. A SNSS could really improve upon the old journal system. But such attempts have failed repeatedly. Here’s how:
Some SNSS are too narrow. Nature Network is very successful; it should be, it’s backed by one of the largest names in scientific literature: the Nature Publishing Group. NN has a fairly wide base, it contains many different papers, blogs, and groups. People exchange scientific information there…but if you’re not in the life sciences you probably don’t use it. That’s a broad generalization, but it largely holds true.
Other SNSS focus on sharing published papers, not data. LabMeeting works as a place for scientists to exchange documents. It even has built in software from Scribd to help in that process. However, scientists can circumvent this site (and mostly do) by sharing these documents directly (via email, ftp sites, etc). There’s no incentive for sharing raw information publicly in LabMeeting.
The same goes for Web of Science (part of Reuters’ Web of Knowledge). Look at its content though and you see it’s evaluating and aggregating journal papers, not providing raw data or even original peer review. Web of Science is a great resource, but it’s just a place to plug into the old journal system, not innovate it.
Most SNSS simply don’t have a fully developed community. Research Gate is a newcomer that’s gaining steam (250k users, 35 million articles) but connecting to its user base is really hit or miss. (I couldn’t find anyone I have ever heard lecture or written a story about). If you know people on Research Gate, chances are that it is a decent place to network (there are blogs, groups, long comment strings, and forums). If you don’t know anyone on RG…well you might luck out and be in one of the fields that’s well represented.
A leader in the SNSS phenomenon has yet to emerge. SNSS deaths are easy to find (SciLink?) but there are always new ones coming up to take their place (Scholarz.net is now in beta!). There is no Facebook of science, there is no Twitter for scientific papers, there is no LinkedIn for sharing data, there is no YouTube for sharing sexy science videos. Well, actually, YouTube has plenty of science videos on it. It also has Keyboard Cat. It’s a mixed bag.
The problem is that SNSS have focused on networking, or served as a new place to find old-style journal papers. They haven’t provided an online arena where scientists can feel comfortable providing data for real time peer review. More importantly, SNSS haven’t reached a critical mass that would give them the clout to reward the scientists who use them. Why do so many scientists want to get published in Nature? Well (as circular as the logic is) because so many other scientists want to get published in Nature. A SNSS will need prestige, will need respect, will need interest from the scientific community in order for it to succeed. That success will naturally allow work shared in a SNSS to serve as a basis for grants, prizes, and scientific reputation.
There are two possible ways this could play out as I see it. One, things continue as they have been. Scientists talk with each other at conferences, they share data with collaborators, but they continue to be very cautious about making research public. Don’t want to get scooped, or even worse, publish findings before they’re certain of their work. In this scenario, the old journal system will continue as it has for the past 350 years.
Scenario two: major reform to the scientific system. Peer review has to move from analog to digital and drag funding along with it. Yeah…probably not likely. Still, with a major over haul in the way scientific work is rewarded scientists will be encouraged to share their work publicly and in real time. Until you don’t have to worry about being scooped, until you don’t have to worry about being criticized for publishing too soon, you won’t want to share data openly. Once those obstacles are over come, then there will likely be one site that gathers a large following and rises above the rest. A SNSS that every scientist uses.
I think open source forums and communities are the models to look to here. Willow Garage is successful and shares in an open robotics social group. OpenWetWare serves as a reference tool and (to some degree) social site for those in synthetic biology. When competition surrenders to collaboration, the flow of ideas can pour forth unabated. I think that accelerates scientific discovery. I hope the same sort of real-time exchange can be extended into the global scientific community as a whole. We need a place where data, ideas, critiques, debates, and collaborative work can flow. We need to take the peer-review process into the 21st century. If the scientific community made the effort, a social networking site could accomplish that goal. I hope we’ll find a way to make it happen. Soon.