Blog by Robert Michael Lundin, Cardiff University - RDA Europe Plenary 7 Early Career Programme Winner

As a 4^th year medical student in the UK, the path that lead me to the Research Data Alliance (RDA) plenary was different to many of the other early career program participants. I have no background in IT, bioinformatics or programming, but as a medical researcher I care passionately about how research data is analysed, shared and published.

 As I sat down on the plane to Tokyo I had just found out that I was taking over as chair of the National Student Association of Medical Research (NSAMR). This also meant that I suddenly inherited a rather complicated problem. Over the past year the organisation has been attempting to launch a journal for medical students, but it had proven difficult to determine the most suitable format, schedule and technical solutions.

While a research journal targeting medical students might not seem particularly important, there are many reasons why one is desperately needed. As the focus on teaching students about evidence-based practice increases, medical students are also expected to be involved in more research. This has resulted in most medical schools running 8-week student select research components, intercalated degrees where medical students take one year out of their medical studies to do research and a number of other opportunities. With the increase in research opportunities, medical students in the UK no longer receive extra points from simply presenting at conferences. In order to achieve extra curricular points for applications they now have to be published in a PubMed-indexed journal before their final year. The need for an outlet of these thousands of projects created annually is therefore increasing every year.

This was in the back of my mind as I sat down in the lecture theatre to listen to the Open Symposium on Data-Driven Science organised by the Japan Science and Technology Agency (JST). For someone at the very earliest stages of their research career, open science and open access has not even reached the extra-reading sections of our medical curriculums. From the opening talks, however, it became clear that in order to make open science a reality we have to take steps to put the publicly funded research back where it belongs in the public domain. To do so it is simply not enough to make research data available through repositories, but we also have to link the resulting papers to the data to increase the utilisation. Furthermore, a move towards open science will only be possible once researchers can be convinced that research data in the public domain can beneficial for not just society, but also their own work.

Throughout my week at the plenary I was very impressed to hear about the developments by the Publishing Data Services working group. By collaborating directly with publishers and researchers, the output sets the precedence for what the journals of the future will look like. It is particularly the outputs from the working groups covering Persistent Identifiers (PIDs), Data Citation, Metadata Standards and Data Bibliometrics that will be able to drive this process as we push for better standards and openness. However, while publishers are open and actively involved in change, it will take a long time to implement all the recommendations in order to sufficiently test and refine solutions for the thousands of articles published each year, not even considering article archives.

It was not until the final day when I was watching the Data Services present their output it became clear to me that the RDA’s need for testing and implementing outputs and recommendations and my own challenge in finding a structure and purpose for the medical student research journal overlapped perfectly. The creation of a brand new journal with initially low volumes of articles and no backlog of articles to update and change allows implementation of all RDA recommendations. RDA partners and collaborators can then see the effects of outputs in action before deciding which aspects to implement first. Additionally, with a completely open access journal we can show new researchers the benefits of open science before the even graduate. A successful implementation of a medical student research journal can be replicated as needed to cover all areas of undergraduate research fields to provide new opportunities for early career researchers while simultaneously promoting open science and testing RDA outputs.

It is particularly the multi-hub model develop by the Data Services group that allows for interlinking articles and data through OpenAIRE. A demonstration is available for the Data Literature Interlinking (DLI) Service and the API that has been created is the first step in linking data and articles throughout the scientific community. Another important development from the working groups is the THOR Project which allows linking of PIDs through DOI, DataCite, CrossRef and more. For a journal to fully take a step into the future of Open Science it will need to fully incorporate these principles and implement the developed APIs.

The following rendering and description is what I envision a medical student research journal would look like applying RDA recommendations and additional comments and discussions from the sessions during the plenary. The following section focuses on the end-user experience of the journal. For more information on the technical aspects of the recommendations, please see the working groups mentioned above.

Starting with the foundation of an Open Access, online-only journal, the overall design is based on comments on journal article entries becoming increasingly cluttered as more features are added. By removing all non-essential details from immediate view it is possible more comfortably read an article quickly. Article details reveals DOI, version of article, current status and dates of publishing. 

The main text section only displays the section you want to read. This significantly reduces the information load and makes it easier to refer to the tables and graphs below without jumping between sections of the page. You can alternatively download the entire article as a PDF or change the viewing options to display all the text. As an online-only journal it is easy to allow high-resolution, full colour graphics for articles to improve the visual appeal. The article references have been shortened to display just the most important pieces of information; title of paper and author. Clicking on the reference would display the full reference in an information box.

Details on the authors has been moved to the right sidebar where clicking on the author name will reveal affiliated institution such as ORCID, twitter and more. With increased number of authors, the font size would decrease to improve readability. Similarly, a list of collaborators for the project would be revealed by clicking on the item. It is important for researchers to have easy access to the reference in a suitable format to improve the quality of citations. The reference box would display the reference where the button can be used to change format required by the institution or journal.

An important aspect of improving the online reading experience stems from the increased importance of article metrics. While having citations is great, social integration means we also want to know how many views, tweets and shares an article has in order to more appropriately gauge the impact. As many researchers prefer to download a PDF which does not provide any of these metrics, an important goal of a more user-friendly article page is for the reader to prefer to online experience. Displaying additional metrics and making sharing easier means researchers can take greater pride in their work and more easily important the impact of their work.

To increase discussion on articles and increase publication transparency it would be possible to display a moderated open review (registered members of the public) and comments from the senior reviewer. It would potentially be possible to allow the author to decide whether open and peer reviewed comments should be displayed. The right sidebar also makes it possible to display additional important information such as funding, information on datasets with a hyperlink direct to the repository and the software used to analyse it. Additional features could be added to display recommend articles based on the article type to promote further reading on a topic and other articles by the same author(s).

In summary I believe I student-led research journal could be an excellent way for the RDA to test and implement new recommendations and technical solutions, such as APIs. The Open Access, online-only journal could be the first step for implementing all new relevant outputs by providing a testing platform with a relatively low volume of articles for evaluation. At the same time the journal would reach young researchers at an early stage in their careers in order to promote Open Science. If successful, the concept could be expanded to additional fields which would allow testing at a greater scale in order to help RDA obtain feedback from real-use scenarios and RDA partners carefully consider which recommendations and features to implement to suit their structure and needs.