4 HN users recommended this talk4 months2018–2021score 5
Years with an HN recommendation, 2007–2026
HN commenters love it as a witty and insightful tour of design anti-patterns, applicable to DSLs, OOP and API design.
Stuart Halloway's satirical talk describes how to protect yourself from industry change by spreading complexity through systems while following idiomatic practice. It uses this inversion to catalog software design anti-patterns.
“a great talk by Stuart Halloway called Narcissistic Design.” (HN, 2018)
Along a similar line, I love Stuart Halloway's talk on Narcissistic Design - [this video] Prefer APIs over data. Start with DSLs. Always connect (and never enqueue). Create abstractions for information. Use static typing across subsystem boundaries. Put language semantics on the wire. Write lots of unit tests. Leverage context. Update information in place.
Nobody seems to have posted Stuart Halloway's Narcissistic Design yet, so I'll just leave this here. [this video] Has some great insights and quips. "Nothing says 'screw you' like a DSL"
[this video] This talk goes into many anti-patterns, but he gets to the real problem with SQL, that it is ONLY a DSL and lacks either an API or a fundamental information model to program against. Composability when working with a DSL is a nightmare. I also have been moving away from ORM to directly writing SQL, but lets be honest about the reasons that programmers are wary of working directly with SQL as they are legitimate issues.
Covers anti-patterns including why SQL is only a DSL.
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
Plays via YouTube's privacy-enhanced embed
4 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.