About
Sources, rights, and discovery
Poem Explorer is built around copyright-free reading: older source editions, clear attribution, and local discovery metadata that helps readers move from one poem to the next.
What is in the corpus
The poems come from public-domain-oriented sources such as PoetryDB, Project Gutenberg, and Wikisource. Each poem keeps source metadata so a reader can see where the text came from.
The collection includes English-language poems, older translations, and a growing world poetry map with China, Japan, Korea, India, Australia, Brazil, Chile, South Africa, Morocco, Russia, and Sufi traditions.
Rights and translations
The app is conservative about rights. A medieval poem may be old enough to reuse, but a modern translation of that poem can still be protected.
That is why translated poets such as Rumi, Hafiz, Li Po, Hwang Jini, and Gabriela Mistral are represented through older source editions, public-domain translations, or public-domain transcriptions.
The Reading Room follows the same rule. A copyrighted essay that is free to read on another site remains a link, not a locally reproduced text. Public-domain labels apply to a specific original edition and jurisdiction, not every later edition or annotation.
How recommendations work
Search combines exact text matches with tags, keywords, and lightweight semantic vectors generated locally from poem language. The goal is discovery, not authority.
The recommended next poem is chosen from shared themes, forms, moods, and contrast with the current poem, so the path stays related without becoming repetitive.