Research is highly dependent on digitalization. We use computers to store, analyze and communicate research. Research visibility is highly dependent on the internet and search engines. Publishing is also mostly made online today, again being dependent on digital publication or data repositories.
Research data is communicated in a specific form called a digital asset which is actually a digital form if information is read and communicated to the audience and other authors using machines and computers.
How we publish and access research is dependent on the digital asset and computer communication. To improve these aspects, we can apply the FAIR principles. FAIR stands for Findability, Accessibility, Interoperability, and Reuse of digital data.
1. Findability
The accessibility of research and the associated contributed data findable is essential. Reaching audiences and communicating the research is one of the main reasons for publishing any research data.
Findability is the first characteristic of the FAIR principles. It describes how well the research can be found by potential readers. Research is about communicating your results to your readers and stakeholders. Today, we live in the world of digitalization. The best ways to find research are computers and search engines like Google, Yahoo, Bing or any other we use every day.
If you were wondering why doi numbers are important for research, findability is your answer. The doi number serves as a unique identifier and enables human-computer interactions (search engines) and easier finding of the results.
Additionally, doi number makes sure an author's research has its unique identity and protects author rights.
Another common identifier is ORCID. ORCID serves as a unique identifier for research authors. Combined with doi, ORCID numbers provide a fantastic framework for making research identifiable in a digital world. In addition, finding research articles is very accurate and confusion around similar Researchers and their Research is avoided in a world of hundreds of millions of research articles online today.
Meta-data is essential to making your research findable. Using the context around the data, such as protocols, methods used, instruments, reagents used, taxon names in the study, location, time or any other contextual detail can provide the keywords to make your research more readable and findable by search engines.
Another important principle of FAIR is to associate the main data and the meta-data. Unique identifiers from the data should be clearly associated with the information in the meta-data.
Indexing is essential to ensure your data is found by different data repositories .
In addition to FAIR principles, making can be done by provide links to your research on social and professional media to make your research findable. These aspects can drastically boost the findability of your research.
2. Accessibility
Make your research as accessible as possible. Data retrieval should be simple for potential users. Data should be freely accessed and open sourced. Free means “free of cost” and open sourced means “the codes and scripts bound to the data are publicly available.” These two principles are vital for accessible data for any researcher. According to FAIR principles, meta-data should be available and open sourced, but the trend is to have the full data as accessible as possible.
Most data repositories have a very clear way of communicating with anyone over the internet. They use http - specific internet protocols to make it easier to send and receive information over the internet.
Protect your data
Even though accessible data is generally a good thing, sensitive data should be protected and the right profile of researchers and authors should have access.
Data can be FAIR even if it's not 100% open to everyone. FAIR principles suggest that the most essential principle is to have clear definition and protocol on who can access data and how. Authentication is one of the principles applied in data repositories so the right profiles can access the data under the right conditions. Author should apply a clearly defined set of rules on how data can be accessible.
3. Interoperability
Interoperability enables use of published research and data in different ways by different platforms, contributing to the ways research can be used more effectively.
Interoperability means data should be stored in a way researchers can access it with ease with various platforms. The way data is described and stored should enable its standardized reuse and integration with other data sources, different programs, programming languages, and platforms. This means researchers can reuse data for different purposes with an accurate and predictable way of communicating the data.
Vocabulary for marking data
Use standardized units, unique identifiers, and other marks for data so it’s easily readable and operable.
Significant parts of the data analysis occurs within different programming languages. Store data in an applicable format like a JSON or .csv file. These formats are interoperable across the majority of platforms and programming languages.
4. Reuse
Finally, reusing data means the data is used to the maximal extent and not archived after its first use.
One of the best ways to make a scientific contribution is by making the data available for others to reuse. One specific dataset can always be observed from different perspectives and used in different research areas. If it's used only once by the primary author, the majority of the data potential is often never used.
Reusable data enables the scientific community to increase the amount of data available online and accessible to everyone.
Today, some of the most significant drivers of research are online open data repositories, such as Harvard Dataverse, NCBI, EMBL and many others.
Data contributors
Authors that contribute their data (data contributors) for reuse make huge contributions to science. They facilitate science development not just through transparency; they provide data to many other researchers. Data contributors are essential in today's science. Data contributors are also cited just like other research authors; data contribution is a form of authorship of research data.
License of reuse
It's not just reusing data, but making sure license of reuse is present in the database or repository where data is stored. The license should be accurate and intuitive to the researcher which provides a perspective on the terms under which data can be used.
If you are looking for permissive licenses frequently used in Research, MIT - One of the most permissive, free of charge, but also enabling many modifications and uses of the Research data. Frequently used if the limitations data use is to be set at the minimum and Apache 2.0 - License is also permissive similarly to the MIT, where in both cases citations and making the license visible when reusing data is always a good approach.
Identify the license terms under which you can reuse data from other authors or data repositories. If the license is unknown, contact the author for the terms.
Final thoughts
FAIR principles ensure data is findable and useful for both data contributors and researchers looking for data for their own research. Clearly defined accessibility will make data more transparent and secure, while interoperability should be present to avoid different platforms not being able to use the research data.
Reusing data is one of the most important drivers of today's research. By making the data reusable and clearly defining the conditions the data can be reused, data contributors are key stakeholders in research along with the publishers of the research.
Researchers frequently publish both the research and the corresponding data to make the research more transparent and enable other authors to reuse the data.
