For the past four years (starting January 2021), I have been involved in a group of faculty and graduate students producing a set of tools for both the authorship and learning of mathematical content, easing our department’s reliance on third-party publishers. Ultimately, we envision a more affordable, accessible future for higher mathematics education. The project leader is Dr. Jan Reimann of Penn State University, and we sometimes refer to our group as Calculus with Jupyter.

We pursue expanded accessibility and affordability of mathematics content by developing open-source tools in Python, JavaScript, and similar to (1) author new or revamp existing materials, (2) automatically generate textual descriptions of existing, visual, mathematical content, and (3) design educational notebook environments with randomization and personalization capabilities.

In particular, I have already helped to convert the assignments, assessments, and textbook of our popular Calculus I course from a third-party publisher to completely free, accessible, HTML or Canvas-based formats. The original publisher’s site – which hosted the course textbook, homework, and even assessments – costed over $100 to access per student, was not able to be read by screen-reading devices, and used a proprietary coding language to author problems. Thanks to our intervention, students can have tailored learning experiences with randomized practice problems and a very clean interface: all for free. Here is a permalink to the textbook we composed in Jupyter Books from existing teachers’ notes: MATH 110: Techniques of Calculus, first deployed in Fall 2021.

Authors benefit from our tools as well. We have developed a universal problem scripting language suitable for and easily transferrable to various platforms. We have collected a library of GUI components and plug-ins which optimize for accessibility, especially for those using screen-readers. Furthermore, machine learning has played a part in the generation of long/alt-descriptions of the large corpus of visual materials, as well as in augmenting the interactive capabilities of our notebooks in responding to natural language.

From my point of view, mathematics is inherently an historical, human process of developing and preserving human-conceived interventions, meaning mathematics cannot be divorced from its pedagogy. And for this reason, I aim to teach mathematics in a human-focused way by centering the needs of the students. To me, the student outcomes of confidence and competence in the course material are what drive me to continuously improve my teaching practices. I aim to help students understand the need of the mathematics, relate it to their own experience, witness its use cases, and feel welcome to succeed and make mistakes without penalty. This ideal environment for mathematics education cannot exist without providing the proper accommodations and accessibility in an equitable manner to the students. I view my work under Calculus with Jupyter as aligned with my vision to make a more affordable and accessible experience for mathematics students at Penn State. I hope to expand the scope and reach of this project to inspire similar efforts across the various departments within and beyond my university.