A Data-centered User Study for jsCoq (short talk)
summary
A lightweight data pipeline records every interaction with a modified jsCoq front end inside Docker Compose, sends click and shortcut events to a Node.js event logger, and persists timestamped snapshots of the scratchpad, goals tab, messages, worker status and user agent to MongoDB records. The system can replay user steps, surface browser-specific bugs, reveal accessibility issues and liveness bottlenecks, but currently lacks anonymization, requires manual image updates and faces MongoDB scaling limits.
AI-written summary · about
chapters (11)
- 0:00A Data-Centered User Study for jsCoq
- 0:21Are the tools we use to write proofs good?
- 0:32What is a good tool?
- 1:23Usability in programming environments for imperative languages
- 2:28Detecting usability issues: user studies
- 3:14Usability in UITP?
- 4:25Data Collection
- 5:15Generated data
- 6:10Analyzing the data
- 7:56Summary and Future work
- 10:18References
related talks
20mThe ongoing design and evolution of cyber-dojo
29mDeveloping JS Applications Using Docker Containers
35mMaking a lot of things
26mTales from the QA crypt
22mFrontend Observability Odyssey
49mHow to get smarter while doing your job
47mDebugging, Profiling, & Rocking Out with Browser-Based Developer Tools!
57mPanel 7: Testing and tooling