Science 2.0

I read an interesting article “Science 2.0 -- Is Open Access Science the Future?” in Scientific American a while ago. This article essentially asked whether or not the easy collaboration that the Internet allows will be a good thing or a bad thing for science. Interestingly, this article was actually based on comments that the on-line article, “Science 2.0: Great New Tool, or Great Risk?” received on the Scientific American web site, so it’s actually an example of the very type of collaboration that it talks about.

Supporters of open and collaborative science claim that it’s inherently more productive. That's definitely good. Critics of it say that if scientists aren’t careful then open and collaborative work can lead to ideas being stolen. Less-than-ethical people might take credit for someone else’s work. They might even patent it. That's definitely bad.

I personally wouldn’t want every step of my research published on-line, say in a blog or on Twitter. This is because when I’ve had jobs where I did research I found that for each good idea I had I also had several bad ideas, and if I openly talked about all of the bad ideas I’d probably look fairly foolish. On the other hand, I can also think of one particular case where someone might have pointed out a mistake that I was making and saved me six months of frustration. Overall, though, I don’t think that I’d like Science 2.0.

It seems to me that there are really two outputs from research: new knowledge and the people who understand what it took to gain the new knowledge. And it seems to me that although you might still get the first of these two with Science 2.0, the second really wouldn’t be the same. I believe that I learned more from my failures than from the successes, and that if you took away the pain of the failures, you’d also take away most of what you really learn from doing research. You might reach the end more quickly with Science 2.0, but the people doing the research wouldn’t get the same benefits from getting there as they would otherwise. In the long run, this will probably erode the ability of people to actually do significant research.

It will probably take a while to see whether open and collaborative research works or not, but you might wonder how the same approach might work for information security. That’s one area where openness is strongly encouraged. The principle that security technologies should be fairly open has been known for over 150 years and is essentially codified in Kerckhoffs’ principle.

Maybe we’re really already there. Although you won’t hear about the details of the research on Twitter, you can already find electronic publication copies of lots of research in cryptography on the IACR’s e-print server. These papers may not be the exact version that gets published, but they’re very close, and they definitely contain all of the new ideas, even if a few of the details of how they’re presented may eventually change.

Come to think of it, you can also get electronic copies of papers in other fields also. The Cornell University Library’s arXiv e-print server now has close to 600,000 papers from physics, computer science and mathematics. There are probably other similar sources out there that I don’t know about.

I don’t really see people doing many blog posts or tweets about the details of their research while I do see lots of papers becoming available on pre-print servers, so I suspect that’s where Science 2.0 is really headed.