Most Influential Paper - Don’t Underestimate the Importance of Experime...
summary
A controlled experiment compares 19 students working 90 minutes on four C systems (web server, chart, IRC client, terminal) with refactored versions that convert fine‑grained preprocessor annotations into block‑level, “disciplined” forms; a custom controlled programming environment records edits and scrolling. Measured tasks are mental simulation, maintenance and bug detection; analysis finds no significant correctness difference and a slight speed advantage for undisciplined code, contradicting the prior assumption that undisciplined annotations are categorically worse.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:11Authors and context
- 3:43Software product lines history
- 6:50Preprocessor variability
- 12:33Experiment setup
- 17:29Experiment results
- 26:18Impact and legacy
- 29:49Takeaway for programmers
- 31:15Q&A disciplined vs undisciplined
- 34:08Q&A experiment validity
- 37:19Q&A experiment design
- 40:31Q&A modern languages
- 42:00Q&A code size
related talks
28m7±2 Reasons Psychology Will Help You Write Better Code
01h 03mRewiring your brain with test driven thinking in C++
49mReading code
18mPreview: Verification (Correctness & Debugging)
07mDevelopment setup: how an important part of your toolset is often overlooked
47mReading code
35mDebugging the Mind: A Programmer's Guide to Critical Thinking
24mDispelling The 'Genius Programmer' Myth Through Code Review