Copyright and Licensing

1.0 Reading :

Read the FAIR Cookbook Data Licenses page to explore modern standards in data management. As you read, focus on how different open licenses address three core rules of data sharing:

• Attribution: How to properly credit the original data owner.
• Permissions: What you are allowed to modify, transform, or redistribute.
• License Inheritance (ShareAlike): When derivative datasets must carry the exact same terms as the original.

2.0 Scenario Discussion:

Have the participants Evaluate the following 3 Scenarios:

The scenarios will come back at the end of the course in a more in-depth follow-up discussion:

Instructions for Participants:

Below are three real-world research scenarios. For each scenario, analyze the researcher’s goals, evaluate the legal constraints, and recommend the best specific license (e.g., CC0, CC BY 4.0, CC BY-NC, ODbL, PDDL). Be prepared to justify your choice to the class in the follow-up discussion at the end.

Scenario 1: The Machine Learning Corpus

• Context: Dr. Chen compiled a raw dataset of 100,000 anonymized climate sensor readings gathered from global weather stations.
• Goal: She wants researchers and AI models around the world to freely aggregate, clean, format, and train models on her dataset without getting bogged down by legal citation requirements across millions of data points.
• Question: What license (or legal tool) should she apply, and why?
• Answer Key for Instructor: CC0 (Public Domain Dedication) or ODC PDDL. Reasoning: Eliminates legal friction and avoids “attribution stacking” in automated AI pipelines.

Scenario 2: The Curated Ecological Map

• Context: A postgraduate researcher created a manually verified, highly curated database mapping endangered plant species across European forests.
• Goal: He wants anyone to reuse, modify, and build upon his database structure, but he requires two conditions:
  1. Anyone who uses or adapts the database must cite his original published paper.
  2. Any modified version of his database created by others must also be released openly under the exact same open terms.
• Question: What license should he select?
• Answer Key for Instructor: ODbL (Open Database License) or CC BY-SA 4.0. Reasoning: Meets both attribution and “Share-Alike” requirements while specifically targeting database structure.

Scenario 3: The Cross-Border Public Health Survey

• Context: A research team collected survey responses on public health habits. A commercial pharmaceutical company wants to integrate this data into a digital health app, while an academic team wants to merge it with demographic data.
• Goal: The principal investigator wants to maximize public benefit and commercial translation, but is considering adding a CC BY-NC-ND tag to “protect” the data.
• Question: Why is CC BY-NC-ND a bad choice here? What should you recommend as a Data Steward?
• Answer Key for Instructor: Recommend CC BY 4.0 or CC0. Reasoning: CC-ND prevents merging/reformatting the survey data with demographic data. CC-NC prevents the pharma company from building a digital health tool around it.

3.0 Lecture

Target Audience:

Researchers, Data Stewards, Postgraduate Students

Instructor Goal:

Walk participants through data licensing choices without drowning them in legal jargon. Focus on practical research impacts.

Section 1: Setting the Hook — Public Accessibility vs. Legal Reusability

What to say / emphasize to your class:

“Before we dive into the licenses themselves, we need to clear up a massive myth in open research. Many researchers think that if they upload a dataset to Zenodo, Figshare, or an institutional repository, it’s automatically ‘open’ for anyone to use. **That is legally incorrect.**In most legal jurisdictions, if you post a dataset online without an explicit license attached, the default legal status is ‘All Rights Reserved.’ That means even if your intention was to share it openly, another scientist cannot legally copy, modify, combine, or build upon your dataset without asking your written permission first.A Data License is simply a standardized contract or waiver that tells the world exactly what they can and cannot do with your dataset—saving everyone time and legal headaches.”

Section 2: Creative Commons (CC) Framework

Instructor Note:

Start with the permission building blocks before showing full licenses. It keeps students from feeling overwhelmed by acronyms.

2.1 Breaking Down the Permission Marks

Explain that CC licenses act like building blocks using two-letter tags:

• BY (Attribution): “You can use this, but you must give me credit.”
• NC (Non-Commercial): “You can use this, but only for non-commercial purposes.”
• ND (No-Derivatives): “You can copy and share this, but you cannot alter or remix it.”
• SA (Share-Alike): “You can remix this, but you must share your new version under the exact same license.”

Class Example to Use:“If you see CC BY-ND, what does that mean? It means a user must credit you (BY), and they can share exact copies, but they cannot adapt or modify your dataset (ND).”

2.2 Deep Dive: Recommended Licenses for Data

Emphasize these two primary options for research outputs:

• CC0 (Public Domain Dedication):
  • Instructor emphasis: Call this the gold standard for raw data and metadata.
  • Why? Explain attribution stacking. “If a researcher combines 50 different datasets that all require attribution (CC BY), anyone using that new mega-dataset technically has to cite 50 different sources in a specific legal format. CC0 waives all rights completely, eliminating this legal hurdle for big data projects.”
• CC BY 4.0:
  • Instructor emphasis: Excellent for curated datasets, research papers, and figures where standard academic credit is expected.

2.3 Warning Your Students About NC and ND

Instructor Note:

Spend extra time here! Researchers often instinctively choose NC or ND without realizing how destructive they are to open science.

Why warn against CC-NC (Non-Commercial)?“Researchers choose NC because they don’t want ‘companies profiting off their work.’ But in practice, ‘non-commercial’ is legally fuzzy. It blocks public-private research partnerships, hinders AI/Machine Learning models trained by mixed teams, and stops non-profit startups or commercial spin-offs from translating research into real-world applications.”Why warn against CC-ND (No-Derivatives)?“ND is a death sentence for data integration pipelines. Cleaning data, changing file formats, translating column headers, or merging two tables technically creates a ‘derivative work.’ An ND license strictly forbids this, making the data useless for automated data pipelines.”

Section 3: Open Data Commons (ODC)

Instructor Transition:

“Now, why do we need Open Data Commons if we already have Creative Commons?”

What to explain to your class:

“Creative Commons was built for traditional copyrighted works like text, images, and music. However, databases have a unique legal structure—especially in places like Europe, which has a specific ‘Database Right’ separate from copyright.Open Data Commons (ODC) licenses were written specifically for databases, separating the structure of the database from the data points inside it.”

Point your participants to the Open Data Commons documentation and cover their 3 core licenses:

1. PDDL (Public Domain Dedication and License): The ODC equivalent of CC0. Complete waiver of rights.
2. ODC-By (Attribution License): Equivalent to CC BY. Reusers can do anything, provided they cite the database.
3. ODbL (Open Database License): An attribution + “Share-Alike” license. Reusers can remix the database, but any new database they create using your data must also be released under ODbL.

Section 4: National, Governmental, and Institutional Contexts

Closing advice for the instructor to share:

“Wrap up by telling your participants to always check local and funder mandates before picking a license. Many public funding bodies (like Horizon Europe or NIH) or institutional repositories have default licensing requirements.For example, some government agencies require an Open Government Licence (OGL), and many universities now mandate CC0 for all research metadata.”

4. Wrap-Up & Debrief (5 Minutes)

Reconvene the class and have groups share their answers for Scenario 3. Emphasize that as future researchers and Data Stewards, their role is to remove legal barriers to data reuse while protecting academic integrity through proper citation.