Contents
1. Introduction
To enable a secure and easy-to-use workflow for any user working with data in Research Drive, a Project Folder Owner can set up a folder structure that allows access to individual users or one or more groups of users which can have different functions and permissions.
As a Project Folder Owner for a project, you:
- Create a logical folder structure
- Manage access to project data: Grant access permissions on folders to users and groups
- Manage user groups: Creating user groups and adding/removing users
When creating such a structure, we recommend keeping the FAIR principles in mind:
FAIR recommendations for folder structuring
1. Create a clear folder structure:
Organize your project folder with a clear hierarchical folder structure. Start with top-level folders such as "raw_data," "processed_data," "homework_assignments," and "scripts." Within these, create subfolders based on experiments, data types, or time periods (like dates, months or years). This improves findability and makes navigation intuitive.
2. Use descriptive file names:
Use consistent and descriptive naming conventions for your files and folders in Research Drive. Include key details like the content type, date, and version (e.g., experiment1_results_2023-10-01_v1.csv). This makes it easier to identify files at a glance, enhancing accessibility. Check out the page on File and Folder name recommendations here.
3. Validate your data:
Create automated or manual checklists that validate data integrity before uploading files to Research Drive. This can include checks for file completeness, format consistency, and adherence to predefined data standards. Regularly validate your data to ensure they remain accurate and reliable over time.
4. Include Comprehensive Documentation:
Create a README file at the root of your Project Folder in Research Drive that outlines the structure, file descriptions, and methodologies. Additionally, consider including metadata files for datasets that describe variables, units, and data collection methods. This supports reusability and helps others understand your data.
Bonus tip: Creating subfolders in your Project Folder can be particularly useful when you collaborate with other Research Drive users on a project, but not every user should have the same access rights to that specific directory. For this reason, you can create a group of users, for example "students", "teachers" and "researchers" and hand out permissions to each group accordingly.
2. What is a project folder?
A project folder is basically an account, but with the following characteristics:
- The project folder has a quota
- The project folder is shared with the Project Folder Owner
The reason for this is simple; people move in and out, but projects generally persist for a longer period of time. By making a functional account per project, all data in the project folder becomes and remains property of the project account.
The project folder can be shared as needed with any Project Folder Owner or member.
Additionally, billing becomes easier because you can do this per project folder, for example when an internal billing system bills per usage on a project basis.
Projects
Each project can have one Project Folder Owner, where multiple projects can have the same Project Folder Owner. In fact, a likely scenario would be for the group admin to be the Project Folder Owner for all the projects of your organization.
In other situations, the Primary Investigator will be the Project Folder Owner for the project.
3. How to structure your project?
Below you will find an example of a good Project Folder structure according to the FAIR principles as mentioned in the introduction.
3.1 Start with a new Project Folder
A new project folder can be created by a Research Drive user with the role Contract Admin or Dashboard Admin via the Research Drive Dashboard. Once the folder is created, the admin can assign you as the Project Folder owner.
When you log in as a Project Folder Owner, you should see your new project folder. Here the admin used a descriptive name (weather_predictions_project) for your research project.
As you can see, the folder is currently empty, 0 KB in size. It is also not shared with anyone but you.
3.2 Establish the folder structure in the project folder
The Project Folder Owner can initially create a folder structure based on the project's needs and sharing requirements. To do so, create a clear folder structure and use descriptive file and folder names.
Make sure you include a READme file that includes an outline of the project's structure, file descriptions, and methodologies. Consider including metadata files for datasets that describe variables, units, and data collection methods. You can use the integrated markdown editor to create your READme file within Research Drive.
Let's explore the data structure further by zooming in how you could store your raw data.
At this level, you can for example separate your data by years in which the data was collected.
Zooming in on the year 2023, you can separate your data based on the months in which the data was collected.
Each month folder contains CSV files of the predicted versus the actual weather for each day. Give your file names a descriptive name such as
20240909_predicted_vs_actual_weather.csv
Using a informative file name such as the ones in the image above helps users to understand what the contents of the file are. In addition, the files can be sorted based on the dates they were collected.
3.3 Share the project folder with custom groups
In order to make management of permissions and users easier, Research Drive allows a Project Folder Owner to create groups that contain multiple users. These group can then be selected when sharing folders.
A Project Folder Owner can create as many groups as is needed. A user can be added to as many groups as needed.
Let's continue with the previous example project folder weather_predictions. In the example above, we created a clear project structure and in this example we will set different access permissions to each of these folders for different groups of people (Teams):
In the next step, we will create 3 different groups, each containing different members. These groups are named researchers, students and externalcompany. The key point is, that researchers need write access to raw data, results and exam_answers. Students need read permissions on raw data, but write permissions on results. For example, students should be able to write the results of their homework assignments, but not read the exam answers. Employees of the external company that build a new software tool based on the results of the models need read access to the results folder.
To create a custom Team, navigate to the top menu and click on "Contacts":
Here we can create Teams and add users to them. In this example, we created the groups researchers, students and externalcompany. We added the user accounts to the applicable groups.
Now that we have groups, let's go back to our project folder. By clicking the circular icon next to the folder name, you can add groups and set permissions. Here we gave the researchers write (can edit) access to the raw data, and the students View-only access:
If you wish to learn more about how to set different sharing permissions, checkout this page.
Do not hand out sharing rights to other users/groups
If you hand out sharing rights to another user or Team with access to the contents in your project folder (or a sub-directory of your project folder), this person/Team can now share the contents of your folder with anyone. Importantly, once the second party shares the folder/file with another person, the project folder owner is not able to change reading, editing or sharing rights of the third-party user.
We therefore recommend to NOT hand out sharing rights, unless absolutely necessary.
If your project requires you to hand out sharing rights, it is possible to keep track of shares of your folder/file in the side window of the shared folder.
You can also check the reporting app:
If you do not see the Reporting app icon in the top menu, you have to request access to the reporting app by contacting the dashboard admin of your institute. Documentation on how to interpret the data as presented in the reporting app can be found here.