Usable PL: Human-Computer Interaction for PL Research
summary
Argues that HCI methods let PL research pick higher‑impact problems, refine designs, and check real effects by using three concrete study types: reconnaissance studies (watch users do tasks to find what’s broken or working), formative studies (quick prototypes and mockups to choose between solutions), and evaluative studies (controlled measures like time‑to‑correctness). Shows a GUI that edits a JavaScript program as users interact, stresses observing users for minutes rather than asking, choosing a clear research question, and iterating early and often.
AI-written summary · about
chapters (14)
- 0:00PL and HCI connection
- 1:45Study users in PL research
- 4:24User studies for PL tools
- 7:02Design effective user studies
- 9:07Research question templates
- 16:01Needfinding studies and observations
- 18:30Observation vs interviews
- 22:22PLHCI without user studies
- 23:23Quick observation benefits
- 24:24User study types and impact
- 25:11Q&A: Human interaction vs technical claims
- 26:56Q&A: Unexpected study results
- 28:43Q&A: Recruiting participants
- 31:55Q&A: PLHCI without user studies
related talks
27mResearch Like You're Wrong
11mEvaluating the Usability of Privacy Choice Mechanisms
39mHacking usability
24mUsability Testing for Developers
40mGames User Research Throughout the Development Process and Beyond!
25mThe hacker's guide to usability testing
06mRataplan: Resilient Automation of User Interface Actions with Multi-modal Proxies
48mUser testing on any budget