Introducing Datapane Text Reports — a better way to write data science documents
Imagine you have created a beautiful report in Python from your plots and datasets, but your audience needs more context to understand your analysis. You want to go from a set of visualizations and data to a long-form report, article, or blog-post which you can use to tell a data story.
You start using Datapane’s Text component to add snippets of text in your Python code, and you end up with something like this:
Wouldn’t it be nice if you could write this text in a proper editor within the browser, and easily integrate your plots and datasets from Python?
In the Datapane open-source library, Text is a component which you include in Python, in the same way you would include an Altair plot or a pandas DataFrame.
This is the good fit for visualization and data heavy dashboards and reports, where you need to add a heading or a caption, which was the first way we saw people using the library. But we soon realised had neglected a large subset of the documents our users want to build, which were less “PowerPoint” and more “Word”.
These users faced a problem: Python is the best platform for analysing data, but it definitely wasn’t designed to be a way to write the text of a tutorial or blogpost. For the latter group, we noticed that most users instead wrote the bulk of their text in Medium, and then embedded individual plots (like they would with GitHub gists and YouTube videos).
