August 24, 2016
Academic researchers tend to be extremely overworked, even when they focus “only” on publishing their articles in the best possible journals. Very often I hear from them that they are too busy to write blogs. Yet, in my opinion, making research open on every stage should not be seen as additional work.
When analysing the output of De Gruyter Open Author Survey I was continuously writing partial updates here on the blog. At the very beginning I published a complete dataset and some R code before publication of the report from the research. In consequence, I think I can say that the research was conducted in an open manner. And of course, I wish all research to be at least as open as this one.
It is quite obvious that society can benefit from open communication of all stages of research, especially, when we are talking about ground breaking academic research dealing with the crucial problems of today. Conducting research in an open manner helps in reducing the number of errors that eventually get published. An open notebook of a researcher may also expose dead ends to his/her colleagues, before they get themselves into them. Reporting findings in near real time is an ideal help for the research community to find the right way quicker and more effectively.
Of course, De Gruyter Open Author Survey was not an academic project. Academic researchers usually aim to publish their results in a recognized academic journal and promoting their output in other ways is not very important for them, even when the real readership of famous scientific journals is limited. Contrary to this, writing, a number of good blog posts were one of my main goals when doing this research.
I think that it is worth working in an open notebook manner, because it helps other researchers to make the world a better place. But there are several other reasons.
1) It is true that keeping the data and the code clean and well described is time consuming. It is necessary when you want to make them open, and will consume some additional working hours, because a person who has not participated in your research at all needs much better descriptions than your coworkers. However, when you come back to your code or dataset after some time, it is a wonderful feeling to realize that everything is clear.
2) I lost my computer accidentally while I was working on the output of this survey. Of course, the whole project was synchronized with my own Cloud anyway. However, after the computer was broken, I realized that I needed to have a look at my old Processing code. I had used it some time ago to create some maps. And when I needed it recently to make another map, showing global income distribution. I found the necessary code in an old blog post. If it was not there, I would have been in trouble. There is no safer place in the world than the Internet, so when I have something important on my computer, I am always impatient to put it on-line.
3) Writing a report from the research was much easier for me as I had all of my main points already explained in blog posts. Of course, work on a report was an excellent opportunity to recount and rethink everything once again, but it took less time since my previous notes were organized in the form of blog posts. I am always taking notes while working on something, but blog posts are much easier to understand and re-use than my average note. When writing a blog post I always try to imagine that I speak to somebody who has a little knowledge on my subject. And therefore these posts are in a useful form for me, when I come back to them.
4) Publishing partial outputs, code and data brings more visibility to your work. It really works according to my experience.
To conclude: no, open science is not a waste of time. It is good investment of research work that pays off and brings some additional satisfaction. And I hope that more and more researchers will take advantage of working in this way.
Image by LaurMG under Creative Commons Attribution-Share Alike 3.0 Unported license (cropped).