Hackathon Trace Ethnography
Research on hackathons using "traces" or artifacts of the development and documentation process.
What is this thing?
I am, among other things, a PhD student in information science, and I research scientific hackathons. This repo contains code used to create visual representations of hackathon communication in the form of Slack channels, Github Repos, and Google Docs. In this project, I am looking at different patterns of communication among teams and whether these have an impact on outcomes and outputs of the hackathon. Here is an example of the chart I have created to do this.
Why should I use it?
Probably you shouldn't. I can't imagine this has any usefulness for any one other than me, but I've been learning about pipelines and Github and Git at hackathons so I thought it would be fun and interesting to put my code here. Also, working on this was a lot more interesting than working on my lit review.
Parts of this repo
Essentially this repo contains a pipeline consisting of several parts:
- get_github_data.R: takes as input names of Github repos and a Github API token, gets back and parses data on commit history
- download_google_doc_data.R: using the Google auth functionality from the googlesheets package, this takes as input document IDs for google docs and gets back and parses data on revision history
- slack_data.R: takes as input a Slack API token and Slack channel names and gets back and parses data on Slack channel chat history
combined_charts.R takes as input a data frame containing Slack channel names, Google Doc IDs, and Github repo names, and runs all three of the above, returning a list of data frames containing the combined data, ready for chart building. The second part of this, which I may separate out later, builds the chart for each of the teams. implementation.R shows an example of how I used these various parts to get data and make charts for three hackathons.