DevOps for DataVis: a survey and provocation for teaching deployment of data visualizations

Left: bar chart showing mentions of devops keywords, with no keyword groups represented in a majority of syllabi. Right: bar chart showing nearly a five-way tie in languages / toolings mentioned in syllabi between R, Python, Tableau, Excel, and web tools like d3.
Syllabi mentioning devops and tooling keywords. No DevOps keyword groups were mentioned in even close to a majority of syllabi. Is this because they are assumed as a prerequisite? Not taught? or taught but not considered noteworthy? Regarding tooling keywords, note the non-dominated front: no one tooling group was mentioned on much more than a third of all syllabi. Some courses mention multiple toolings – does this improve breadth of understanding, or introduce confusion?
Abstract
We present a provocation towards teaching development operations ("DevOps") and other infrastructure concepts in the course of collegiate data visualization instruction. We survey 65 syllabi from semester-long, college-level data visualization courses, with an eye toward languages and platforms used, as well as mentions of deployment related terms. Results convey significant variability in language and tooling used in curricula. We identify a distinct lack of discussions around ‘DevOps for DataVis’ scaffolding concepts such as version control, package management, server infrastructure, high-performance computing, and machine learning data pipelines. We acknowledge the challenges of adding supplemental information to already dense curricula, and the expectation that prior or concurrent classes should provide this computer science background. We propose a group community effort to create one free ‘course’ or ‘wiki’ as a living reference on the ways these broader DevOps concepts relate directly to data visualization specifically. A free copy of this paper and all supplemental materials are available at osf.io/bxaqz.
Materials
PDF | Preprint | DOI | Supplement | BibTeX | alt.vis "Most Frustration-motivated" award!
Authors
Citation
Thumbnail image for publication titled: DevOps for DataVis: a survey and provocation for teaching deployment of data visualizations
DevOps for DataVis: a survey and provocation for teaching deployment of data visualizations

Jane Adams. Proc. alt.VIS workshop at IEEE VIS—alt.VIS. 2023. DOI: 10.31219/osf.io/ub539

PDF | Preprint | DOI | Supplement | BibTeX | alt.vis "Most Frustration-motivated" award!


Khoury Vis Lab — Northeastern University
West Village H, Room 302
440 Huntington Ave, Boston, MA 02115, USA