Tuesday, 22 January 2013

Fixing Scientific Publishing

I have already written in the last post how scientific publishing works. A lot of people has been arguing for open access journals, and there is some good in that. Of course, open access has its own problems, like the fact that you have to pay huge fees and still doesn't have freedom to choose your own copyright rules. 

My favourite site is, of course, arxiv.org. However, in order to make sites like it really work there is a circle that must be broken. It's the funding one. Call it the Funding Feedback Loop if you like. I will use the initials FFL just for convenience here. This is one of the worst things that exist in science today and one of the responsible (if not THE responsible) for the tyranny of the scientific publishers. It works like this.

Science needs money to keep running (as everything else) and, as is common in life, those who really have money to invest in science are not scientists in general. Therefore, they need to decide how to invest their money and where. For simplicity and laziness reasons, the people they hired to make this decision came up with "objective" criteria, which means finding some numbers which are easy to get to create a ranking system. Those people are not really interesting in evaluating the research itself, because this is a complex task which involves thinking and decision taking, and they are not really interested in that. Instead, they do a quick search and find a bunch of indices which seems reasonable.

For some reasons, good ones in the past, some scientific publications became highly prestigious, like Nature and Science. There are others, of course, but they can be ranked by something which is called "impact index", something objective and calculable. No need to think. This, associated with number of citations and things like that, are the dream of the decision taking groups. They just need to gather those numbers and their job is done.

Of course, classifying science according to those numbers is like classifying countries according to GDP. There is a correlation, but ask the poor people of Brazil if the fact that its GDP is high does really matter to them. The problem is that, after people start to rely on the numbers to take the decision, they forget that those numbers are biased, incomplete, full of noise and, in many cases, prone to a lot of corruption and manipulation.

In science, what happened is that people stopped concentrating on scientific knowledge and started to aim in what journal they wanted their paper to be published. The funding agencies only give money to those who publish in Nature, so people struggle to publish in Nature and Nature gets more and more power. The same for other prestigious publisher.

Unless funding agencies stop requiring scientists to publish in certain journals to get funding, there is no way to liberate scientific knowledge simply because scientists need money to survive! Are you going to pay the scientists if the funding agencies don't? Don't try to fool me, I'm pretty sure you won't. But then, how can we break the FFL?

Part of the solution is for funding agencies to stop being lazy and doing a serious job on decision making! But that's complicate! Yes! It is! But if you do easy things the result is invariably rubbish! Everyone knows that, or SHOULD know that! Stop relying on easy reading ranks and do a serious analysis of what is really important in scientific research. That would increase the costs? Yes, it would. There's no free lunch. If you want to do something with quality, you need to invest time and money.

Therefore, to break the monopoly of the powerful publishers, the FFL needs to be broken. In order to do that, the decision taking about funding needs to rely on a more detailed analysis of what to fund. One that relies on the quality of the research independently on where it is published. Invest on that and you start to take power from the publishers. Of course they will react to that, and they will react aggressively! That's because there is a lot of money involved and nobody wants to let it go! They obviously have powerful lobbies. There's obvious a lot of corruption in the system, the kind of legal corruption of course. But the only way to minimise it is to doing the right thing.

So, if you are a serious funding agency, one which is not chained by lobbies or agreements and really wants to invest your money in serious science. If you are a rich person that loves science and want your money invested in a way that really creates free knowledge for future generations, here is my advice:

Take part of the money available for the funding and invest in a serious group of people to analyse the scientific value of the proposals instead of their "objective" measures. Pay them to do a damn good job! Hire scientists who like science to do that analysis job, not administrators. Spending part of the money on that might guarantee a better use of it.

There is only one thing: be careful with corruption on that process! As I said before, there is not free lunch. If you want things to be of quality, you need to be always alert! It's like democracy. If you stop paying attention to it, it gets corrupt.

No comments:

Post a Comment