Science has a long institution of collaboration; ideas flourish and multiply when they're shared, and that creates progress.
Until the 1990s, the only way to readily share that information was via published journals; an expensive, slow way of sharing information that requires the movement of little pieces of paper from place to place.
There's no reason it should be like this any more. And I argue that this is actually hurting our ability - as a species - to progress.
The Internet (and specifically Google at the time of writing) is the biggest source of information on the planet. Potentially, everything could be out there, readily accessible by everyone. It has way surpassed my wildest dreams in that.
But the information isn't itself useful (and this is where Google comes in). What's really useful is the mining of that information. The ability to enter keywords, and find related articles. The stuff that lets you take data points and connect A to B.
We are on the cusp of a revolution in science. For the first time in the history of humanity, you don't actually personally have to do experiments to test a theory. The sheer weight of numbers of other people out there, doing the research, and publishing their results removes the burden of performing those experiments themselves from the individual scientist. We're democratizing science, and making it accessible to the intelligent individual in a way that previously was only possible for the theoretical sciences. You no longer need tenure, or to be working in a research facility, to actually draw conclusions from research.
And data-mining allows that to come to the fore. In the near future, I can even imagine a world where Google itself could be mechanized. Computers themselves could draw conclusions from all of the research data, and come up with useful correlations. It's the Kurzweil singularity; at some point the system feeds off itself, and will spiral off to infinity.
But what's stopping that now?
The problem we have right now is that for most papers, only abstracts are available online. The actual detailed information is stored in academic publications, such as Phys. Review Letters A, and the Journal of Neuroimmunology - but the barrier to entry is too high for the skilled individual. It just plain costs money to read those papers. Scientific progress isn't supposed to work that way. It's supposed to be for the benefit of all of us.
That's not to say that there aren't considerable advantages to the peer review and publishing of papers. In fact, it's still an essential part of the scientific method. And it could continue, but not in the way it stands right now.
We should change the way academic papers are published. We need to democratize this system. And partly, this has already happened. Pubmed, run by the US National Library of Medicine, and the National Institutes of Health is a good example of how to collect papers online. arXiv is another example - it's Cornell University Library's database of preprints of papers in Physics, Mathematics, Computer Science, Quantitative Biology and Statistics.
How should it work?
- All papers get published online, either in collector sites such as those mentioned above, or on the individual author's sites - preferably both, so they can be archived for the future.
- The journals take the papers, they are reviewed and refereed, and the papers which pass muster are published by them. This allows a bound archive of the best of the best work; the stuff that we know is real. It also provides an instantly accessible catalogue of verified high quality work, which those journals could charge for. What they're providing here is convenience, and a level of trust - which in an increasingly growing, polluted internet information space is becoming more and more important. (I've noticed recently that it's much more difficult to search for something on Google now, than 2 years ago... without serious AI advances, that problem is only going to get worse).
- The authors and collector sites mark the papers as "peer reviewed", and provide references to where they were published, after they are published. This means that people can still access the useful information, and still have a hope of finding out which papers are valid - or not.
Sure, the scientific publications will make less money this way. But frankly, I don't have much sympathy for them; we're way past that business model's useful lifetime (as much as I, an ex-freelance journalist, regularly bemoan that). We could entirely bypass this system by providing something like Digg for scientific publications. At least this way, they're still involved in the game.
Come on people. Let's get some science done here, and use the singularity to our advantage.
Simon Cooke is an occasional video game developer, ex-freelance journalist, screenwriter, film-maker, musician, and software engineer in Seattle, WA.
The views posted on this blog are his and his alone, and have no relation to anything he's working on, his employer, or anything else and are not an official statement of any kind by them (and barely even one by him most of the time).