Foraging among an Overabundance of Similar Variants
summary
Study of eight novice programmers examines reuse across many near-duplicate program copies and refines Information Foraging Theory by adding a variation dimension. Participants receive timestamped, chronologically ordered variants with change logs and use timestamps, change-log descriptions and visible UI differences to select source and destination copies. They perform between-variant foraging to find the right copy and within-variant foraging to find patches, reuse matched “information features” across copies sometimes successfully, sometimes leading to abandonment. Open questions: useful auto-save points and appropriate variant granularity.
AI-written summary · about
chapters (6)
- 0:00Introduction and motivation
- 2:41Research questions and study setup
- 6:38Foraging process and patch identification
- 11:35Integration and evaluation of variants
- 14:26Implications and open questions
- 16:13Conclusion and Q&A
related talks
14mPFIS-V: Modeling Foraging Behavior in the Presence of Variants
09mMicro-Versioning Tool to Support Experimentation in Exploratory Programming
50mGuide Refactorings With Behavioral Code Analysis
22mShifting Left, Shifting Right
32mPractical code archaeology
20mDevX: History of Code Search
30mVisualizing process evolution
18mGrounded Copilot: How Programmers Interact with Code-Generating Models
