I don't like notebooks.- Joel Grus (Allen Institute for Artificial Intelligence)
What it is
Joel Grus's JupyterCon talk explaining why he dislikes Jupyter notebooks. He covers hidden state, trouble with reproducibility and testing, and how notebooks frustrate teaching. He shows his preferred workflow and what Jupyter could change.
Why Hacker News recommends it
It is the go-to funny, irreverent critique of notebooks, linked whenever HN debates notebooks versus testable code.
- 6 people recommended it
- 6 different months
- 2 strong recommendations
- first 2018, last 2025
What HN commenters said
“they're definitely irreverent, but I find myself strongly agreeing with everything he says.”
Irreverent critique of notebooks the commenter fully agrees with.
tcpekin, 2018-10 · Strong recommendation · in “Why Jupyter is data scientists’ computational notebook of choice”
“I highly recommend it.”
Talk critiquing Jupyter notebooks against engineering best practices.
kevsim, 2019-04 · Strong recommendation · in “How to separate your data from your code”
“I think both talks are very worth watching”
Critical counterpart to Jeremy Howard's pro-notebook talk; both show strengths and shortcomings.
tosh, 2020-10 · Recommendation · in “I Violated a Code of Conduct”
“I would recommend the talk”
Argues for moving code from notebooks into testable libraries.
softwaredoug, 2021-06 · Recommendation · in “Ask HN: How do you run long computations without interruption?”
“it's a really funny talk, a lot of the points are IMHO invalid as Joel is misinformed on how things can be configured, but still a great watch.”
Funny talk on notebook flaws, though some points reflect misconfiguration.
carreau, 2022-01 · Recommendation · in “IPython 8.0”
“The parent article links this great talk that goes into more detail than the parent post and is much easier to understand”
Detailed explanation of why notebooks make state hard to track.
lelandbatey, 2024-06 · Recommendation · in “Notebooks Are McDonalds of Code”
Excerpts from Hacker News comments, linked to the originals. Summaries and labels were written by a language model from those comments.
Topics: jupyter, python, data science, reproducibility, software engineering
Also recommended by the same people
explainerThe US Has More Tribes Than the World Has CountriesCGP Grey · 4 HN recommenders · 2020–2026
talkRails Conf 2012 Keynote: Simplicity Matters by Rich HickeyConfreaks · 8 HN recommenders · 2015–2024
podcastJoe Rogan Experience #1342 - John CarmackPowerfulJRE · 3 HN recommenders · 2019–2020
talk"Programming Should Eat Itself" by Nada AminStrange Loop Conference · 3 HN recommenders · 2015–2024
talkClojure for Lisp Programmers Part 1 - Rich HickeyClojureTV · 1 HN recommender · 2017–2019
talk"The Mess We're In" by Joe ArmstrongStrange Loop Conference · 10 HN recommenders · 2014–2025
talkI Like NotebooksJeremy Howard · 4 HN recommenders · 2020–2025
podcastWhat Alcohol Does to Your Body, Brain & HealthAndrew Huberman · 7 HN recommenders · 2022–2023