Show HN: What is HN thinking? Real-time sentiment and concept analysis

HN builds a mood ring for itself — users cheer, shudder, and nitpick

TLDR: A real-time Hacker News dashboard tracks comment vibes and trending topics. Reactions split between “this looks like government surveillance,” excitement about entity tracking and DIY dashboards, and nitpicks about sloppy sentiment and samey green colors — a fun yet uneasy mirror of the community itself.

Hacker News just launched a real‑time “what are we thinking?” dashboard, and the comments instantly turned into, well, a dashboard of human feelings. The tool tracks activity spikes, shows happy/sad vibes, and highlights trending names and topics. Cue the drama: one top comment compared it to government surveillance tools, dropping a sarcastic “congrats, I guess” and sending a chill through the thread. That’s the privacy panic camp.

On the hype side, builders are thrilled. One user says large language models (AI that reads text) make it easy to turn messy comments into pretty charts, even sharing their own job post tracker. That’s the maker energy speaking. Then came the design police: “Great work… but please fix the greens.” Yes, the sentiment pie’s similar shades of green became a mini‑meme, aka Fifty Shades of Meh.

The spiciest critique hits quality: folks love the entity tracking (seeing which people and topics are hot) but call the actual mood scores “sloppy,” blaming budget‑level AI. Translation: the labels impress, the feelings guesswork… not so much. So where does HN land? A split screen: fascination with watching HN analyze itself, fear that it looks like surveillance tech, and fussy love for color palettes. The controversy meter says “medium,” but the comment energy says “peak meta” — HN is staring in the mirror, arguing about the lighting, and asking if the mirror works at all.

Key Points

  • A real-time dashboard analyzes Hacker News discussions.
  • Activity Over Time charts show story and comment counts across hours.
  • Sentiment is tracked both as a time-series trend (-1 to 1) and as a categorical split (neutral to very positive/negative).
  • Controversy and Depth gauges summarize the intensity and analytical depth of discussions.
  • A Discourse Breakdown classifies contributions (e.g., metacommentary, counterargument, technical insight, questions, humor, resources).

Hottest takes

"virtually identical to tools the US Department of Homeland Security uses" — esseph
"LLMs open up space for transforming unstructured raw data into visualizations and dashboards." — dk8996
"the sentiment ratings themselves seem pretty sloppy" — sdwr
Made with <3 by @siedrix and @shesho from CDMX. Powered by Forge&Hive.