A curated archive of speculative fiction and technology writing, indexed not by title or author but by idea. Every book is broken down at the chapter and passage level, connected through a knowledge graph that maps how concepts — surveillance, post-scarcity, cyborg identity, algorithmic governance — evolve across works and decades. You can browse it, search it, or have a conversation with it.

The physical collection exists — thousands of books on actual shelves. The digital system indexes those books at a depth that a physical catalog never could: passage-level semantic search, concept-level connections, and an AI librarian that can synthesize across the entire collection in response to a question. Think of it as the building’s nervous system.

A general-purpose AI draws on everything it was trained on, which means it gives you the consensus view — the Wikipedia version. The Tronix Librarian is grounded in a specific, curated collection. It cites actual passages. It knows what’s in the archive and what isn’t. It can trace how an idea evolves from Ursula K. Le Guin through Octavia Butler through N.K. Jemisin because it has the conceptual graph connecting those works at the chapter level. It has a point of view. General AI is a search engine; this is a reader.

Yes. The conversational interface is powered by a large language model, grounded in the library’s indexed collection through retrieval-augmented generation. It has access to the knowledge graph, the full-text search index, and the concept evolution narratives. It doesn’t make things up about what’s in the collection — when it cites a passage, that passage exists. When it doesn’t know something, it says so.

A network of ideas and the relationships between them. Each node is a concept (like “post-human identity” or “corporate sovereignty”), a work, or an author. The edges between them represent connections: influence, contradiction, evolution, co-occurrence. When you explore the graph visually, you’re seeing the conceptual DNA of the collection — how ideas migrate between books, mutate across decades, and sometimes predict the future.

Fiction that makes itself real. The term comes from the Cybernetic Culture Research Unit (CCRU) and describes how speculative ideas can escape the page and reshape the actual world. Gibson imagined cyberspace; engineers built it. Stephenson imagined the Metaverse; companies spent billions trying to create it. The library takes this seriously: it treats speculative fiction not as entertainment but as a kind of distributed R&D for civilization.

It’s not only science fiction. The collection includes critical theory, philosophy of technology, cyberculture history, futurism, and design futures. The common thread isn’t genre — it’s a shared preoccupation with how technology reshapes human experience. A chapter of Donna Haraway sits next to a chapter of William Gibson because they’re asking the same questions from different angles.

By a human, with opinions. The collection reflects a specific curatorial sensibility — it gravitates toward work where speculation and technology are in genuine dialogue, where ideas matter as much as narrative. There’s no advisory board, no algorithm, no popularity metric. A book gets in because a person read it and thought: this belongs in the conversation.

Check the archive. The collection is curated, not comprehensive — if a book isn’t there, it either hasn’t been indexed yet or it falls outside the collection’s scope. The archive is growing continuously.

Not yet through the interface, but the collection is actively growing. Suggestions that fit the collection’s sensibility — work that sits at the intersection of speculation, technology, and the human condition — are welcome. The best way right now is to mention it in a conversation with the Librarian; it won’t add the book, but the suggestion gets heard.

That’s exactly what it’s for. The system is designed for writers, researchers, foresight practitioners, students, and anyone who thinks seriously about futures. The Librarian cites its sources. The knowledge graph shows its reasoning. Every claim can be traced back to a specific passage in a specific work. It’s a research companion, not a summary generator.

Conversations are not stored on our servers. Messages pass through third-party AI providers to generate responses. See the privacy policy for details.

One person, slowly, in public. The collection, the knowledge graph, the indexing pipeline, the conversational system, and the daily dispatches are all the work of a single curator-developer. The library reflects that — it has the coherence and the idiosyncrasies of a personal project with a clear editorial vision.

The codebase is on GitHub. The corpus itself — the indexed texts — is not publicly distributed for copyright reasons. The knowledge graph data, concept evolution narratives, and the system architecture are all visible.

Continuously. New books are ingested, indexed at the chapter and passage level, woven into the knowledge graph, and made available to the Librarian. The pace depends on the complexity of the work — some books connect to dozens of existing concepts; others open entirely new threads.

The archive, knowledge graph, timeline, and daily dispatches are free to browse. Conversations with the Librarian require a login. The system runs on real infrastructure with real costs — how that gets sustained is an evolving question.