The Need for Open Science

I love science, but it is a bit of a mess. It is not very organized, it is not very fast, it has some bias, it isn’t very available to the public, and the people doing it don’t always talk to one another. These are just some of the problems that we need to acknowledge and try and solve. Luckily, the Internet is here to help change everything.

In more detail, here are some of the problems that science currently faces:

Science is Messy
Science is organized in a very simple way – each time an author writes a scientific paper, s/he refers to the journals that s/he found useful, via a citation. As Michael Nielsen put it in his book Reinventing Discovery, “citation is perhaps the most powerful technique for building an information commons that could be created with seventeenth-century technology”. In fact that is when it was created, and it hasn’t changed very much since. It is built for organization on paper publications, and hasn’t really been updated for the Internet. The literature is also littered with citations to findings in other papers that apparently don’t exist. Often, there is no adoption of standard methods, with standard units, in a common language that make results comparable across studies.

Science is Partly Hidden
Many research projects don’t result in findings that the researchers or journal editors think are exciting; it is just part of the scientific process. The problem arises when those findings aren’t published. Maybe the author has a more exciting manuscript that he would rather work on, or maybe the journals don’t want to publish such a ‘boring’ paper. Either way, this biases and polarizes the overall findings in the literature – we end up with papers that only strongly prove or disprove what they set out to look at. One problem that arises from this is known as the file drawer problem.

An equally troubling problem occurs for the manuscripts that actually are published. The manuscripts are published, but not the data. A century ago, publishing the data would have seemed absurd. But now, with the Internet, the question really becomes: why aren’t we? As meta-analyses demonstrate, there is way more information locked away in most studies than the original authors pull out, and more importantly, by pooling data from large numbers of studies, it is more possible to see the bigger picture this way. By not publishing this data somewhere on the Internet, in some sort of organized fashion, we are essentially throwing out a lot of hard-earned work.

Scientists aren’t Sharing Very Well
In the age of social media, the people have largely embraced the ability to talk to one another and share ideas and information freely. Sadly, this doesn’t yet include most scientists. Many scientists are afraid of talking with one another too much for fear that someone will steal their ideas and publish them first. This is why they sometimes hoard their data, and keep entire studies a secret at least until they are published.

Aside from sharing with one another, they are also not yet very good at making articles accessible to the public. Sure, if you are a student, then you have access to a lot of the scientific literature, and if not, you can still get a lot by going through the library (if you live near one). But the fact that someone can’t easily access publicly funded research because articles are locked behind expensive pay-walls is more than frustrating.

Reinventing Discovery
Image from the cover of the book by Michael Nielsen

An Open Science Revolution is at Hand
At a fundamental level, I really think that these are all issues of poor science communication, which is why I find them particularly fascinating. But, I’m not here to leave here on a negative note – that just isn’t my style. I set up this website to discuss how we can go forward and improve the openness of science, so just stay tuned, because I’m very optimistic about this. A quiet revolution is fundamentally changing the way humans make discoveries, through:
– An explosion of open-access journals like PLOS
– Citizen science projects like Galaxy Zoo
– Data sharing websites like figshare
– Journals that are pushing to publish negative results, like the Journal of Negative Results in Biomedicine
– Researcher-focussed social media sites like ResearchGate
– Online lab notebooks like Carl Boettiger’s
– Dozens of other initiatives that you can check out at Digithead’s Lab Notebook or the Open Science Wiki

If you simply can’t wait for me to read more about this topic, I am happy to recommend Michael Nielsen’s book Reinventing Discovery. As a manifesto for a new kind of discovery, called Networked Science, it really lights the imagination on what is possible when Open Science is embraced.

I’ve barely scratched the surface with what I think are some of the important issues, but I want to know what you think I was silly to have left out – leave it in the comments!

7 Comments

    1. Thanks for passing that along Graham! What an amazing resource – I’ll update this post with a link 🙂 Also, I can only imagine how humbling such acknowledgement would have been. Very cool.

  1. Great mission! I’m writing my Master’s thesis on this topic along with the coming changes in higher education (I see your last blog post was also on MOOCs). I don’t care if its published in a journal or not, so I’m going to make it freely available online when its done for anyone to read.

    1. Thanks Brian, and great to hear about what you’re up to! Very bold about your focus on online access over journal publishing, I think you’re right, it is the way forward. Send a link this way once you’ve got it up – I’d love to read your insider’s-scoop!

      (Also, sorry for the comment moderation/delay – these are the first comments here and I had to change the default settings on WordPress)

      1. What are your thoughts, if any, on this article on the privatizing of American science?

        http://www.nytimes.com/2014/03/16/science/billionaires-with-big-ideas-are-privatizing-american-science.html

        – I think to really examine Open Science and improve the system, one needs to take a holistic approach and look at everything involved. For example, from a top-down approach, higher education is going to have to change due to massive student loan debt, MOOCs offering competition, etc. How will that change institutions in the future? Institutions employ the majority of scientists that the open science movement effects. What impacts will that have on their employment, the funding they receive, the attraction of science research as a future profession, etc?

        Those are all macro forces that could have disruptive implications. For example, maybe tenure or the way it is achieved (i.e. journal publication) is no longer a major contributing factor, and instead, other metrics will reach the forefront and these changes spur a more shared, collaborative existence in science due to changing incentives.

        I think the issues go far beyond just the general sharing of data. That, at least in my opinion, is a sub category of other overarching issues that will one day inevitably be disrupted. In what way, when and what the exact consequences these disruptions will have on the underlying research process is tough to accurately predict because there are a plethora of variables involved.

        1. Great article Brian, and you raise a good point about the big picture changes.

          In terms of privatized research and projects funded through philanthropy, I think its pretty cool to see the results that is coming out of it. It is fast, and it takes risks, but has some biases/drawbacks (eg. focusses on ‘sexy’ topics, and generally avoids basic research). To quote the article, “By contrast [to public research], the new science philanthropy is personal, antibureaucratic, inspirational.” But, “[p]hilanthropy is no substitute for government funding.” American science seems to be “losing its competitive edge,” and it is going to take a complex suite of actions to keep them in the big leagues.

          In terms of big-picture, holistic approaches, I agree with your thoughts. It is a hugely complicated topic that requires an equally multi-faceted strategy for improvement. I’ve actually been writing a blog post about this for sometime in the next week. As a prelude, I think we’re watching the beginnings of a real revolution in how academics and research are carried out, and that the future includes even more group specialization and greater online presence. Stay tuned!

Comments are closed.