frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Show HN: What 180k words look like as a temporal knowledge graph (Oz series)

https://synaptale.com/graph?ch=100
8•ald0r•1h ago
The graph is free to explore and requires no registration.

SynapTale builds a model of a story as a temporal graph made up of nodes (entities) and edges (their actions and relationships). The graph is not a visualization of the wiki. The wiki, timelines, relationship histories, and analytics are projections of the graph.

The current demo contains 232 entities, 1,852 edges, and a snapshot of the story’s state at every chapter. By chapter 100, it still remembers a promise made in chapter 8 and turns the story into a set of source-verifiable facts.

The most interesting things can be found in the graph itself and in the Analytics tab. A few things I found:

1. The character with the highest kill count is the Tin Woodman—the same character who cries over a beetle he accidentally crushed. Dorothy comes second, with three killing events. 2. Dorothy never deceives anyone during the first 100 chapters of the series. 3. The Scarecrow’s debt to the stork has remained active for 92 chapters, starting in chapter 8. 4. The Cowardly Lion ranks third by number of threats. 5. The first 100 chapters contain 60 secrets and 254 dialogue events.

Technical details

1. Five different multi-agent pipelines combining LLMs and NLP: a prescan, ontology construction, chapter-by-chapter graph extraction, retrospective validation over spans of dozens of chapters, and a linguistic prescan for speech profiles and linguistic edges.

2. A living story needs a living graph. It has to account for time, because entities and the relationships between them evolve. A simple is_active field is not enough.

I ended up with three types of edges:

event: an instantaneous action; identity: a fact; state: a persistent action whose termination requires justification and a supporting quote from the text.

The vast majority of edges are events and end in the same chapter in which they began. This allows the system to scale well, since only a minority of state and identity edges remain continuously active.

3. Ontology. You cannot simply ask an LLM to extract entities and relationships into a graph. With every chapter, even the smartest model will keep inventing unimportant fields, creating new aliases for existing fields, and representing the same fields inconsistently.

Before extracting the graph, the system therefore performs an ontology scan across the entire story. It captures story-specific entity and edge types, along with their fields and descriptions.

4. Epistemics. Events are only one part of a story. It is also important to understand how information is distributed, which is difficult to represent using event edges alone.

I addressed this by introducing a new node type: epistemic nodes, which capture different entities’ perspectives on the same fact. Subtle hints can still be missed, the system is not yet perfect in this area.

JetZero

https://www.jetzero.aero
72•lisper•1h ago•55 comments

Show HN: I mapped every US golf course – 16k+ courses, free, no signup

https://golfcoursebrowser.com/
43•rickmf•1h ago•7 comments

An ESP32 based plane radar for my desk

https://blog.ktz.me/esp32-plane-radar/
25•alexktz•1h ago•9 comments

Humans Haven't Stopped Evolving

https://www.harvardmagazine.com/research/harvard-human-evolution-genes-selective-pressure
11•ilamont•57m ago•3 comments

Stolen Buttons

https://anatolyzenkov.com/stolen-buttons
634•Gecko4072•5d ago•154 comments

Inflect-Micro-v2: complete voice in 9.36M parameters

https://huggingface.co/owensong/Inflect-Micro-v2
42•nateb2022•3h ago•3 comments

Systems and Delays

https://martin.janiczek.cz/2026/07/24/systems-and-delays.html
37•vinhnx•3h ago•10 comments

Clinical failure rates over the decades: yikes

https://www.science.org/content/blog-post/clinical-failure-rates-over-decades-yikes
70•EA-3167•5h ago•39 comments

Cloudflare's new AI traffic options for customers

https://blog.cloudflare.com/content-independence-day-ai-options/
62•alphabetatango•5h ago•40 comments

The new rules of context engineering for Claude 5 generation models

https://claude.com/blog/the-new-rules-of-context-engineering-for-claude-5-generation-models
204•mellosouls•7h ago•135 comments

Rethinking Legal Education in the AI Era

https://www.law.uchicago.edu/news/ai-strategy-statement
26•jjwiseman•2d ago•1 comments

Git rebase -I is not that scary

https://cachebag.sh/journal/interactive-rebasing/
37•vinhnx•3h ago•31 comments

GM Backs Sodium Ion Batteries for U.S. Grid Storage

https://spectrum.ieee.org/sodium-ion-battery-peak-energy
150•rbanffy•6h ago•58 comments

Running a 28.9M parameter LLM on an $8 microcontroller

https://github.com/slvDev/esp32-ai
106•boveyking•9h ago•24 comments

DeepSeek pause fundraise after comments on compute gap to US leaked (transcript) [pdf]

https://github.com/demo-zexuan/liang-wenfeng-investor-meeting-2026-7-22/blob/master/%E6%A2%81%E6%...
107•oliculipolicula•4h ago•69 comments

Show HN: What 180k words look like as a temporal knowledge graph (Oz series)

https://synaptale.com/graph?ch=100
8•ald0r•1h ago•0 comments

SIMD for Collision

https://box2d.org/posts/2026/07/simd-for-collision/
72•birdculture•3d ago•22 comments

Agatha Christie's Disappearance: Amnesia, Suicide, or Despair?

https://www.historicmysteries.com/unexplained-mysteries/agatha-christie-disappearance/23136/
4•thunderbong•3d ago•1 comments

LLM Usage in Debian: Three Proposals

https://www.debian.org/vote/2026/vote_002
106•zdw•8h ago•97 comments

Open-weight AI is having its Kubernetes moment

https://tobi.knaup.me/2026-07-25-open-weight-ai-is-having-its-kubernetes-moment/
335•tknaup•13h ago•270 comments

Producing ammonia and fertiliser using wind power in Morris, Minnesota

https://ammoniaenergy.org/articles/flexible-renewable-ammonia-demonstrator-now-operational-in-min...
107•gritzko•8h ago•70 comments

What if they are all wrong? (2020)

https://igorpak.wordpress.com/2020/12/10/what-if-they-are-all-wrong/
8•kkoncevicius•4d ago•2 comments

Alien World Chemistry Found Inside Meteorite That Struck New Jersey Home

https://www.seti.org/news/alien-world-chemistry-found-inside-meteorite/
9•spzx•2h ago•1 comments

Show HN: I made some transistor animations

https://brandonli.net/semisim/animations
144•stunningllama•1d ago•16 comments

GDID Windows – Cut the tracker that follows you even under VPN

https://korben.info/en/gdid-windows-cut-tracker-vpn.html
122•rfarley04•4d ago•87 comments

Memory Safety Absolutists

https://itsallaboutthebit.com/memory-safety-absolutists/
68•drogus•9h ago•87 comments

Turn And Face The Strange

https://fly.io/blog/kurt-scott-money-sprites/
160•subarctic•7h ago•117 comments

Zero roadkill as Amazon canopy bridges secure 15,000 crossings

https://news.mongabay.com/2026/07/zero-roadkill-as-amazon-canopy-bridges-secure-15000-crossings/
315•hn_acker•3d ago•96 comments

Show HN: Brolly, a plain-text weather forecast site

https://brolly.sh/forecast/RWFP2qW8
144•jsax•10h ago•45 comments

Did They Ghost You?

https://didtheyghostyou.com/
331•mooreds•7h ago•149 comments