The goal of dauf-exercise is to provide materials when working on an exercise involving creation of a Quarto Dashboard.
Create a Quarto Dashboard using open data from OpenAlex which highlights in which journals/venues the three universities KTH, Stockholm University and Karolinska Institutet have most frequently co-published in the last ten years. Render the content and bundle the result in a container as static web content to be served from the subpath “dashboards/dauf-exercise”.
Aspects to consider:
-
The Quarto tooling is free and open source software allowing creation of outputs in various formats including an HTML-based Dashboard format. For the specific question in the exercise, provide a tabular overview but also include other visuals. Think of one or several useful ways of presenting the results in a comprehensible way. Feel free to explore and use this gallery of Quarto Dashboard examples for inspiration.
-
Add some basic styling to the dashboard. You are free to use any approach including styling assets for example from other places such as this one.
-
Create a container which renders the dashboard embedding all resources in a single file. Hint: quarto documents can utilize a yaml setting which embeds resources locally in the rendered output, see the docs here.
-
Create a container which can serve the results as static web content from a sub-path such as “{servername}/dashboards/dauf-exercise”? Hint:
nginx
could be used as a base layer in the container with the output made available from a sub-path. -
Think about how the above approach for rendering and publishing a dashboard could be automated using GitHub Actions. Bonus points for providing a non-private cloned repo which includes a GitHub Action which renders the dashboard and publishes it to GitHub Pages periodically, say once per day.
-
Look into the
qa.R
script, explain what it does and discuss your findings and thoughts around how you would proceed in a data quality investigation such as this one (where you discover a data quality issue in an external data source).
It is not strictly required, but there are some resources at the KTH Library which could potentially be used in the exercise:
-
kontarion
provides a container with Quarto, R and RStudio for the Web pre-installed. It could be used as a base image when rendering the dashboard in the exercise. -
kth-quarto
contains styling assets that could potentially be used in the Dashboard. There is also styling assets available here which provide “KTH style” when creating plots using ggplot2.