Have you ever downloaded a research dataset, opened the files and realised you had no idea what they contained or how to use them? This is where a README file comes in. A simple text file can provide essential context, making research data easier to understand and reuse. To help researchers get started, the Czech Data Steward Community has developed a practical README template.
A README is a simple file that accompanies a dataset and provides essential information about its contents and use. It explains what the individual files contain, how the data were collected and processed, what the variables and abbreviations mean, and under what conditions the data can be reused.
Some of this information may already be included in a scientific paper or separate documentation. However, these resources are not always stored alongside the data and may become difficult to access over time. A README offers a straightforward way to keep essential information together with the dataset.
Including a README when depositing research data in a repository is now considered good practice. It saves time not only for other researchers but also for the original authors, who may need to revisit their data months or years later.
“In an era where we create more data than we can handle, us researchers and data stewards can easily lose track not only of datasets from others, but also of the ones we create. READMEs have been around for years, mostly used for software, however. It seemed therefore essential to give the opportunity to all of us to have a nice template that can serve datasets from different disciplines,”
explains Georgia Koutentaki
The content of a README may vary depending on the research discipline, type of data, or repository requirements. However, it should provide enough information for others to understand the dataset without having to search for additional documentation.
Key elements include:
Additional useful information may include the dataset's structure, its development history, the date the README was last updated, and the version of the dataset it describes.
Particular attention should be paid to licensing and reuse conditions. Users should be able to quickly determine what they are permitted to do with the data and whether any restrictions apply.
When datasets are shared as part of a double-blind peer review process, care should also be taken to avoid including information that could reveal the authors' identities.
To make README creation easier, the Collective of Data Stewards from the Czech Data Stewards Community has developed a general-purpose README template for research datasets. It provides a clear structure and helps researchers identify the information needed to document their data effectively.
The template is freely available on GitHub and can be adapted to different research needs. Versions are available in TXT, Markdown, and JSON formats.
The project also aims to develop discipline-specific templates and further align them with the FAIR principles to support the findability, accessibility, interoperability, and reusability of research data.
Anyone interested in contributing to the development of README templates can contact Georgia Koutentaki at georgia.koutentaki@cvut.cz.
In the seminar recording, you can also find other sources for READMEs, such as The Turing Way. Hopefully, our examples of filled-in READMEs will serve nicely for all those who need a guide. We have the template in 3 formats: .txt, .md, and .json, so people with different preferences can use the template easily and connect it to machine-actionable services.
“The template was created with the development of the Czech NDI in mind, to provide a common way to introduce information to the repositories that exist and are being developed. However, it can also be used outside of the Czech RDM ecosystem. In the seminar recording, you can also find other sources for READMEs, such as The Turing Way. Hopefully, our examples of filled-in READMEs will serve nicely all those who need a guide. We have the template in 3 formats: .txt, .md, and .json, so people of different preferences can use the template easily and connect it to machine-actionable services,”
adds Georgia Koutentaki
Practical examples, useful resources and insights into the project's development were presented by Georgia Koutentaki during the README (please) – introduction training session at the Summer School for Data Stewards 2026. Ready to create your own README?
Explore the available resources:
The National Library of Technology is launching the National Repository to help research organizations and other institutions comply with the new legal requirement to open-archive scientific publications.
The new Onedata service is designed to make it easier for e-INFRA CZ users to work with large volumes of research data. It provides access to data stored in different locations within a single shared environment and simplifies both their transfer to computing resources and their further use throughout the research process.
