**Title**: FAIR for Data and Texts Not in the Open: Overcoming Legal, Technological, and Economic Barriers **Instructors**: [Ye Li][2], Chemistry and Chemical Engineering Librarian, MIT Libraries; [Laura Hanscom][3], Head of Scholarly Communications and Collections Strategy, MIT Libraries; [Katie Zimmerman][4], Director of Copyright Strategy, MIT Libraries. ### [Course Schedule Link][5] **Description**: The rise of applied data science, digital humanities, machine learning, and artificial intelligence has resulted in an increased need for computational access and reuse of research data and publications. Researchers have begun to build FAIR (findable, accessible, interoperable, and reusable) and open data practices for data they are generating; however, much research requires access to existing structured and well-curated texts and data from proprietary sources that don’t currently meet the FAIR standard. To accomplish this, many researchers are partnering with libraries, which frequently have long-term subscription access to such resources, to gain computational access and rights to reuse for text and data mining (TDM) and machine learning purposes. @[OSF](mptq2) Negotiating for such access and rights poses technical, economic, and legal challenges. In some cases, researchers have to negotiate access and reuse at the individual research group level. In this course, we will interactively explore these issues through case studies from real-life examples and share resources and tips that will help researchers, librarians, and vendors to “move the needle” toward FAIR data. The joint effort of researchers, librarians, and vendors will be required to sustainably ensure that resources move towards FAIR standards, and that researchers can share their own research output FAIRly. **Class activities include**: - Small-group critique of license terms for computational access and reuse of publications and databases - Mock negotiation between researchers/librarians and vendors. - Hands-on practice accessing a database through publicly available API services (e.g., Crossref, PubChem) and comparison with other access models. - Group discussion of cutting-edge questions on computational access and reuse. **Level**: Beginner to Intermediate **Intended audience**: The course is aimed at science and engineering researchers, digital humanities scholars, librarians, and vendors of subscription products interested in computational access; administrators interested in understanding the risks and challenges, including intellectual property issues and commercial interests, of providing access to these resources; and technical and support staff who may be called upon to set up access to these resources and provide storage and security. **Requirements**: Participants should have an interest in and basic awareness of issues related to computational access and reuse of texts and data not in the open. **Course Learning Objectives** At the end of the course, a participant will be able to: - Define the scope of computational access and reuse of data and texts. - Describe computational use cases of data and texts in various disciplines - Articulate the tensions between vendors interests and researchers priorities, particularly when it comes to intellectual property and potential commercial interests in research output. - Describe the various access models available for text and data mining and machine learning research, and describe the benefits and limitations of each (e.g., implications for storage and data security). - Facilitate the conversation between vendors and researchers to make the data and texts more findable and interoperable. - Identify the parts of a license that correspond to the rights that can enable or disable a researcher to acquire FAIR data and share their research output FAIRly when using subscription-based data and texts as main source. - Apply tips and techniques learned in the course to negotiate and suggest language to balance a vendor’s interests with the needs of a research community. - Design pilot services for partnership between librarians and researchers to acquire resources for computational access and reuse. - Describe evolving issues in this space and their long-term implications **Course Topics** This course will be presented over two days for 3 hours each afternoon and will cover these topics: - Computational access - Computational reuse - Text and data mining - Machine learning - FAIR - Library licensing - Negotiation [1]: https://osf.io/mptq2/D9pK/view?usp=sharing [2]: https://drive.google.com/file/d/1lzjs1PQ96VC2WeSRAkB50q4aWi3pDKZo/view?usp=sharing [3]: https://drive.google.com/file/d/18uk9dMS5zoyKeTedHuv4wqA8ON93rIlk/view?usp=sharing [4]: https://drive.google.com/file/d/1CwUxG0X0p3lPtHdBAX7EaCFdF_ZLE9I0/view?usp=sharing [5]: https://osf.io/tbkj7/wiki/Course%20Schedule/