What heraldry and Japanese mon can teach about visual-identity generators
Comparing Western heraldry and Japanese mon reveals how procedural generators should model operations and rules rather than static asset catalogs.
read it A thoughtful architectural deep-dive that contrasts linguistic blazons with geometric mon systems for procedural generation.
summary
The author examines how to upgrade a fantasy crest generator by comparing Western European heraldry with Japanese mon. Western heraldry relies on a linguistic grammar resembling abstract syntax trees, whereas Japanese mon use a compositional geometry closer to scene graphs. Rather than treating historical traditions as static asset libraries, generators should model underlying operational rules. Combining these linguistic and geometric approaches offers a superior blueprint for procedural visual identity systems.
key points
- Western heraldry maps well to abstract syntax trees because coats of arms can be formally described using linguistic blazons.
- Japanese mon rely on compositional geometry, utilizing base motifs, transformations, and scene graphs rather than descriptive text.
- Procedural generation systems should focus on operations like rotation and repetition rather than static symbol catalogs.
who should read it
Game developers and software engineers interested in procedural generation and generative art systems.
discussion warm
consensus
Commenters thoroughly enjoy the deep architectural breakdown, discussing related visual identity projects like Urbit sigils and generative shields.
pushback
A commenter notes that monshō is a more precise term than mon when discussing heraldry broadly.
Reminds me of Urbit sigils. They had a really cool system for generating unique visual identities. The write up is one of my favorite design/tech articles:
https://urbit.org/blog/creating-sigils.
It introduces a parallel real-world system for generating unique visual identities using similar generative constraints.
Generated by an LLM from the article and thread. Check the sources before relying on it.