Day 3, July 11 at 10:00am
Presenters
James Whalley, Senior Instructional Design Technologist, Cornell University
Marina Tokman, Assistant Director of Learning Technologies, Cornell University
Summary
This session explored innovation through the use of the Canvas API. We learned of diverse Canvas API applications at Cornell University, from SDS accommodations to mass assignment date differentiation to automated due date reminders. Also discussed were how these cases inspired CAPICO@Cornell (Canvas API Community) and presenters’ visions for expanding API utilization. Furthermore, presenters shared methods for determining Canvas API use at participants’ institutions.
Details
There are many API use cases at institutions that help to improve the teaching and learning experience in Canvas. At Cornell, those uses include:
- File uploads (instructor)
- Content use queries (admin)
- Enrollments (admin)
- Course dates (instructor/admin)
- Syncing to-do items to Todoist (student)
- Assigning peer reviews to undergrad TAs (instructor)
As well, Senior Lecturer Kyle Harms uses the Canvas API in order to generate codespaces on-demand for students in his introductory web programming course (of ~300-400 students). This allows students to focus on coding and design rather than wasting time manually setting up codespaces. Hear Kyle's story below.
Another use case comes from Senior Lecturer Hadas Ritz who uses Python code with the Canvas API and MATLAB in order to generate large question banks for quizzes in Canvas. This allows her to put together practice quizzes for students which they can attempt multiple times without encountering the same practice problems more than once. Learn more about Hadas' story below.
Additionally, teaching associates in the Department of Chemistry and Chemical Biology utilize the Canvas API to migrate assignments submitted to Canvas to Gradescope and back. This allows them to utilize the Turnitin plagiarism check in Canvas and the grading efficiency of Gradescope. This process is improved by the associates being able to interact with the API via a Streamlit app developed by post-doc Peter Sun. Learn more about enhancing grading consistency via the Canvas API & Gradescope below.
These use cases were identified by running the User Access Tokens report available to account admins. This report was then filtered to only show User-Generated tokens that had not yet expired. With API users identified, the Canvas@Cornell team sent out a survey to learn more about individual use cases. Some of these uses were further explored at an in-person meeting where CAPICO@Cornell (Canvas API Community at Cornell).
At Cornell, API access token generation has also been gated. This allows the team to be more aware of API use cases and to intervene where access token security might be threatened. An example of this is the application Speechify that requests users generate an access token and then provide it to the app (allowing Speechify full access to their entire Canvas account).
The team seeks to further support API usage by sharing current use cases and developing API webinars. Additionally, they hope to continue having CAPICO@Cornell events that may also include users and use cases from other institutions. Finally, they hope to develop a curated dashboard of Canvas APIs that can be interacted with in a secure manner. They imagine this will allow them to introduce the concepts and features of various API calls without requiring prerequisite coding knowledge.
The Innovation Greenhouse at Cornell seeks to build on scalable API use cases. In collaboration with Josh Armes from Wayne State University, they have begun this work with the development of an automated due date reminder LTI called NudgeBot. The team hopes this is the first of many internal developments/collaborations, including possible work on an LTI to map SDS required accommodations to assessments in Canvas, as well as bringing slip day functionality to activities.
Conclusion
Participants are invited to explore current API use cases at their own institutions and to see how they might support and grow positive use cases, while securing access tokens from improper use. The speakers also invite participants to reach out regarding getting Canvas API communities together and sharing resources.
