Debugging with the Scientific Method - Stuart Halloway

Channel:ClojureTV · Length:49:09 · Published:2015-11-17 · Views:25219

Watch on YouTube

Summary

Stuart Halloway of Cognitect argues for debugging with the scientific method: form hypotheses, try to falsify them, and write everything down. The talk is given to the Clojure community but applies to any language.

Why HN recommends it

HN recommends it as a talk that changed how people debug, worth rewatching periodically.

changed how I debug problems. I try to watch it once every year or two. — HN, 2022

Functional programming · Testing, debugging & developer tools

Subtitles (excerpt)

[Applause] all right so testing two different mics here we all good all right so um I guess I should make uh a couple of clarifications on the introduction uh before we begin I'm pretty sure that Alex just drew a chain of causality whereby I can say that I'm responsible for strange Loop so I'm pretty I'm pretty excited to have discovered that and and uh can't wait to see my checks in the mail um uh the second thing that he implied a little bit was that um there's a lot of opportunity uh to watch debugging uh happening on the product team uh if you work at cognitech and actually you know datomic is really no bugs at all so I don't uh I don't know what he's talking about there uh either uh but seriously um you know uh in a closure audience I really want to begin this talk uh with a rationale right this is something that uh Rich has instilled uh as a value in this community and so uh I put a lot of effort in having a good rationale or trying to and I'm going to have a actually a multi-part rationale for this talk the first one is debugging why should we care about debugging and I think there are uh there's an obvious reason because bugs right we have bugs that is a um you know there are some estimates that that

On Hacker News

This talk, "Debugging with the Scientific Method" changed how I debug problems. I try to watch it once every year or two. tl;dr, when debugging people naturally form a hypothesis about what is going wrong and then set about gathering evidence to support or refute that hypothesis. When you do so unconciously you are very prone to biases of all sorts, most importantly confirmation bias. When you do… — seanc, 2022-09
Stu Halloway (also a member of the Clojure core team) has a couple of great talks on using the scientific method in programming: [VIDEO] [LINK] — simon_acca, 2023-05
Stuart Halloway from the Clojure community has some great a talks using the REPL to debug. Links: Debugging with the Scientific Method - Stuart Halloway ( [VIDEO] ) REPL DEBUGGING: NO STACKTRACE REQUIRED ( [LINK] ) scope-capture ( [LINK] ) — grierson, 2018-09
I think your criticisms are definitely valid. The learning curve is steep (intentionally so) but definitely worth it. Your programs will become more reliable and you'll get more done as a consequence. Please push through! The error messages are terrible, and there are a lot of ongoing conversations about them at the moment in the Clojure community[1]. The Clojure[script] toolchain does tend to be… — chrislloyd, 2015-12

Related videos