Open access to scientific publications enables everyone, everywhere to access the accumulated knowledge of humankind and keep up-to-date with the latest scientific discoveries. Free and global access to knowledge can only benefit the common good. Informed citizens can rightfully access the outcome of science funded by their taxes. Scientists can quickly learn the results of their peers, using it to improve their own research and avoid redundant work. Breaking the (pay-)walls of science helps promote cooperation over competition.
The traditional (subscription-based) business model of scientific journals is based on an outdated publication process that relies entirely on a few publishing companies to disseminate scientific results. However, we now live in a technological era where the average person can use a smartphone to instantly publish their thoughts (through images, tweets, blog posts) and reach a target audience of thousands in a matter of seconds.
On the scientific realm, online collaboration platforms like Authorea
allow researchers to write a manuscript and collaborate in real-time. Pre-print servers like arXiv
allow authors to immediately disseminate their results while waiting for peer-review. This protects them from being scooped during the (sometimes long-lasting) peer-review process and gives other scientists faster access to results that can improve their own work. Analytic platforms like Altmetric
give authors an overview of the impact of their work by tracking down the number of views, downloads, tweets, access demographics and so on.
However, open access should not only address the issue of article access. Reproducibility is one of the pillars of science. In this regard, all collected data upon which hypothesis are tested should be available for replication. This issue becomes more relevant as we move towards a world of Big Data, where terabytes and petabytes are becoming common words within the scientific community. Research results are often the outcome of processing raw data collected by large-scale experiments through intricate workflows that crunch the numbers until a polished output is attained. This means that a lot of the assumptions, formulas, and fine-tuned parameters are hidden inside thousands of lines of code, like needles in a haystack.
It is fundamental that data and source code are published together with the respective articles under open source
licenses. Many researchers are now turning to platforms like GitHub
to publish their software and interactive tools like Jupyter
notebooks to share their workflows in a reproducible manner. A lot of data repositories are also becoming increasingly available for authors to share their data (see for instance the EMBL-EBI data submission portal
). It is important that not only authors but also reviewers and editors are aware of these resources and vouch for their utilization at the moment of publication.
Amidst the torments of publication pressure (so-called publish or perish), the quest for high impact factors and fancy press releases (desired by authors and editors alike), one should not forget the importance of negative results. These are often forgotten vital pieces of information, which are just as worthy of open publication, as they can help to improve experimental protocols, avoid redundant experiments and, consequently, the waste of tax-payers money.
Finally, one cannot forget that open access is not a magic solution to all problems. It is a business model that moves the costs of publication from readers (through institutional subscriptions) to authors. This puts an extra burden on the authors on top of their other research costs. Furthermore, scientific publishing is a for-profit business that heavily relies on the goodwill of peer-reviewers that voluntarily do the revision work for free. During peer-review, researchers take time away from their own projects to evaluate and help improve the work of others in their field. The same others with whom they will compete for academic positions and research grants. It is hard to imagine a more altruistic job than peer-review. Publons
is a recent project that aims to give credit to reviewers through a social network of peer-review track records. This is a step in the right direction but hardly sufficient.
One way to mitigate both problems (publication costs and reviewer recognition) is for publishers to start compensating the peer-review work with reduced publication fees. Furthermore, invoices should be more transparent and discriminate the individual costs of the publication process. This would help authors know how each journal adds value to their article along the publication pipeline (editorial work, graphic design, press releases, etc), and make an informed decision on which publisher will give them the best value for their money.