Michael Nielsen has written a thoughtful essay over on his blog asking why scientists have been so slow to pick up on new web 2.0 technologies (found via Bora’s blog). It’s good to see that many of his conclusions echo my own (here too), that the big problems are a lack of time and incentive. He offers some potential solutions, and reasons why people should be using these new tools. A few responses, as always attempting to cut through the evangelism, cross-posted over in his comments thread:

—article continues—

On Wikipedia–Neilsen comments that the reason more scientists aren’t writing Wikipedia articles is, “that contributing to Wikipedia isn’t really science.” In some ways he’s right–being part of an anonymously authored group-posting isn’t something you’re likely to get career credit for. But there are other reasons as well, as anyone who has tried to contribute to Wikipedia can tell you, number one being the near impenetrable web of rules surrounding participation, and the overly zealous guardians of those rules. From my own experience, when trying to add a few facts or correct a few errors in a Wiki entry, every single attempt I made was immediately deleted, citing some obscure and inscrutable rule. Rather than try to challenge the gatekeepers, I simply gave up on trying to help out. As pointed out elsewhere, experts are not welcome on Wikipedia (noted here too), and scientists are likely to be experts in the field in which they’re writing. So, not only is there no incentive, they’re also likely to run into hostile opposition to their participation.

On Extreme Openness–one of the problems with this concept is that it will inevitably lead to even more information overload than we currently face. Right now, we’ve at least got a layer of editorial oversight, vetting of articles before they’re published. If you instead just dump everything into a bin for the user to sort out, that’s a huge timesink. I know, I know, “the wisdom of the crowds” and all that, but still, someone has to read all of those articles. How much of your week are you willing to commit to reading the bottom 5% of the dregs of that bin? I’m not talking about the bottom 5% of Nature articles, I’m talking about the bottom 5% of articles so weak that they are unpublishable by even the lamest of journals. Someone has to read those papers and tag them as the dregs. Is it you? Is this a good way to spend your valuable time? Also, he notes, we need much better tools to sort through all this as well, and extreme openness certainly can’t work without those better tools. As recently noted, a system like this leads to conformity and consensus (at least so far) rather than a widespread range of papers and opinions. And that’s a problem.

On the OA movement–yes, they’ve certainly made great strides and are having a strong influence on the culture of science. That said, as recent reports point out, the economics of the OA movement have so far failed to show that they can be sustainable across a wide variety of publications. We may just be in early times of a long term movement, but I’d be hesitant to throw out the system we currently have for an unproven one.

On quantifying one’s contribution–that’s the million dollar question here, how do we incentivize participation? Right now, all that matters is what influences funding committees, hiring committees and tenure committees. Can you really get these groups to give weight to a Slashdot style karma rating? If so, isn’t such a system open to gaming? A well-known blogger with lots of online friends who all link to his blog and papers would end up with a higher rating than someone without the social networking skills that does better, more groundbreaking science. Politics already plays way too big of a role in scientific success these days, and I fear a system based on one’s social networking skills would only exacerbate the problems (although you’d shift the power to a different group–perhaps that’s why those already using science social networks are so enthusiastic about them, as mass uptake would make them the real power brokers of science).

Citations from arXiv–he mentions the ranking of paper quality in arXiv by the number of citations received as a success story–but isn’t this the same metric that’s currently under fire from so many people? How is this any better than Thomson/ISI’s Impact Factors?

On collaboration–first the mention of the level of discourse on open source programming forums–wouldn’t one argue that open source programmers are under different economic and career pressure than academic or industrial scientists? The very nature of an open source program is quite different from scientific results in a competitive environment with limited funding and limited job space. If you create a program that serves my needs, I can use it. If you get scientific results that are the same as those I’m working on, I can’t use those to further my career. Hence, I’m under pressure not to participate with helpful advice.

On FriendFeed–this, like Twitter, is a tool that many are extremely enthusiastic about, but that I can’t see being adopted by many scientists. Simply put, who has the time? It’s one thing to ask for the time required to participate, to write blog posts, tag papers for del.icio.us, to make YouTube videos and to constantly alert people as to which Starbucks you’re sitting in and which type of latte you’re drinking. It’s quite another to ask people to follow hordes of others as they do the same thing. How many hours a day am I supposed to spend seeing what others are blogging about, tagging and drinking at Starbucks? I can barely keep up with my e-mail, now I’m supposed to read micro-blogs?

On the Economics of collaboration–there’s a big difference between buying a pair of shoes and devoting your extremely valuable time and thousands, if not hundreds of thousands of dollars in reagents and equipment to a collaboration. Trust is established not by chatting with a new friend, but by following the publication record of your potential collaborator. What have they accomplished? Can they really do what they claim they can do? If so, they will have published it, or at least have a respectable record established. I don’t think you’ll ever see people meeting strangers in chat rooms and working together, at least not on a large scale. The stakes are too high. Would you be willing to roll the dice on a year of one of your graduate student’s careers on some stranger you met online?