We have identified two main patterns that take advantage of Curvenote’s computational capabilities: Computational Articles and Computational Reports.
- Computational Articles
- Structured around a core manuscript with attached supporting materials, computational articles focus on providing in-context interactive figures and notebooks with a strong narrative flow. Interactive figures can be executed independently, and notebooks are integrated into the main content. A separate JupyterLab environment can be launched by the reader, allowing them to explore the content fully and encourage re-use and further experimentation.
- Examples: pyUserCalc
- Computational Reports & Tutorials
- Computational reports include the same interactive features as articles, they are structured more like a “book”—presenting a collection of notebooks within a broader narrative. These reports emphasize comprehensive documentation and exploration of the computational environment. Readers can launch a JupyterLab instance to interact with the notebooks or, optionally, execute the notebooks in place within the report.
- Example: Tilt-Corrected BF-STEM
In both cases, it’s important to carefully consider the computational environment for your article or report. Whether you’re working in Python, Julia, or R, best practices for dependency management and reproducibility are essential.
