About recursive.eco
Make belief in order to make sense, recursively
A grammar is a symbolic system — a tarot deck, an I Ching book, a story sequence, an interpretation set. Anyone can create one, share it, or use someone else's. Sit with what's already here, or build your own — no code required.
Humans relate through these grammars — the grammars of the living world. We are aspiring to be a recursive public of grammars, stories, and meaning structures, which are the very infrastructure of culture — infrastructure that maintains and transforms itself through the people who use it.
We are here to support the human process of making belief so we can make sense and make room for new possibilities to emerge — hopefully more adaptive ones, recursively.
Built by PlayfulProcess
recursive.eco is built by PlayfulProcess — read more in the Life Is Process blog.
The builder is a sense-maker who builds with AI, not by writing code — which is the proof of concept: if she can grow this, you can grow a grammar. Inspired by Wikipedia's example — a commons that maintains itself through the people who use it — but way smaller and more creator-friendly, with meaning-making systems as the content instead of encyclopedia articles.
The intention of PlayfulProcess is to build a Recursive Public as defined by Christopher Kelty (Two Bits, 2008) — so we can build a better internet. In plain terms: a recursive public is a community that doesn't just use a shared thing but maintains and remakes the very infrastructure it runs on. Kelty's examples are the free-software communities behind Linux and the early internet: the people who use the system are the same people who can inspect it, fix it, fork it, and pass it on. Wikipedia works the same way for knowledge. While Kelty was focused on open source, recursive.eco believes that responsible freedom might mean harmonizing with a diverse set of structures — sometimes open, sometimes closed. We want a better internet: not bootstrapped-for-the-pitch but actually bootstrapped, with near-zero infrastructure costs and no monetization gradient pulling on us. Not engineered for engagement — a lot of what is dangerous about AI right now shows up exactly where a short-term feel-good substitutes for a long-term capacity. We want a place where humans can use their humanity to connect — with themselves, with each other, with the ecosystem we are part of.
You read a grammar to sit with its symbols. You create your own — writing interpretations, choosing images, structuring what matters to you. You share it in the library for others to find. They fork it, change it, share it back. The loop continues.
There's no algorithm deciding what you see. No notifications pulling you back. You arrive when you arrive.
One Recursive Cycle: Structures · Process · Possibilities
We shape grammars, and the grammars shape us.
…and what you made becomes structure for someone else. That's the loop.
What you can do here
Contemplate — sit with the symbolic systems others have built
Use the Journal as your practice space — cast tarot, draw I Ching hexagrams, read your birth chart, read a story aloud, and journal with or without an AI companion. It carries together with the AI assistant throughout the ecosystem. Private by default — you can even choose to not save anything in our databases: write, export, save where you feel safe, reimport whenever. Oracles built into journaling can also support your process.
Learn — study what others have made
Read grammars other people have built, and follow the makers whose work you trust. When you want to pick up the skills before making your own, take a free Course. Everything in the Library is free to read; no account required to browse.
Create — make your own, for yourself
Build a grammar of your own — a deck, a book, a sequence, a playlist, a family tree. You don't need to know how to code, and you don't need design taste: pick a template for a head start, let the built-in AI assistant write, illustrate, and structure alongside you, and it looks good by default. That's the deal — freedom within a structure: the grammar format gives every deck or book the same solid shape, and inside it your creative choices are entirely your own. Private by default — many people build just for themselves and their families; sharing is a separate, optional step.
Share — send it back to the commons (optional)
When (or if) you want others to use what you made, the Library has curated channels — Tarot, I Ching, Astrology, Wellness, Kids Stories, Parenting Resources — for placing your work where the right readers will find it. Four ways to share:
Curated channels
Submit your grammar to a channel curator. Accepted submissions surface on the channel page and on your personal channel.
Your personal channel
Every signed-in user gets a page at /library/altar/<you> — a shortlist of tools you trust, organized into named collections.
Community editing
Open your grammar to suggestions, wiki-style. Others propose changes; you decide what to merge.
Private with editors
Not ready to publish? Invite specific people as editors and the grammar stays private to your small circle — same review-and-merge flow, behind closed doors.
Tending a recursive ecosystem
recursive.eco isn't trying to be the destination.
recursive.eco's job is to help people structure their symbolic perspectives as grammars and then get out of the way as much as it can. Some grammars might mature — a grammar might grow into a full symbolic-system study with a site of its own, its own repo, its own people — while staying joined underground: same format, same ids, edits flowing back, one AI companion moving through all of it. And many great grammars are deeply studied and simply stay grammars — that is not a lesser fate; there is no ladder of progress here. The tarot channel is the proof: what began as a tarot grammar became a channel and now lives at tarot.recursive.eco, backed by its own public recursive-tarot repo, forkable by anyone. Astrology has followed the same path to astro.recursive.eco, and more channels are graduating behind them.
Once a channel graduates, recursive.eco's role shrinks to connective tissue — discovery, the AI layer that helps people read, build, and play with the content, the sync that keeps both sides current — rather than the landlord of the content. In time, that also means you'll be able to discover channels and creators here and subscribe to follow what they make next.
This is also why the platform is open about its code and its content in different ways. The grammar format is documented and shared (recursive.eco-schemas), published grammars are licensed CC-BY-SA, and we build on an open-source stack ourselves (Next.js, Supabase, Cytoscape.js, and others) that we mean to keep giving back to, not just draw from. The platform code that runs the app stays private — see Structure & Safety below for why.
What we want it to be
We shape grammars, and the grammars shape us.
Grammars are symbolic systems — finite sets of symbols that people use to orient themselves and make sense. Not the territory; maps. Useful only insofar as they help the reader move through what is actually there.
Many grammars fit the data of reality in their own way. Atheism and Shaiva Tantra, materialist science and indigenous cosmologies — these can be compatible representations of what is, but they are explained, ritualized, and practiced very differently. One may be more adaptive for a person or a community than another. The point is not to find the “right” grammar; it is to find one that adapts well to the life you are actually living. Here they sit side by side — the Mamluk cards next to the Zhouyi next to Jyotish — none of them crowned.
This openness is not “anything goes.” Meaning systems need time to sediment into communal structures before cooperation can grow from them — so we hold structures that are adaptive: not so rigid they harden, not so loose they dissolve. We stay open to the mystery that the territory is ultimately unknowable, which is exactly why different views of reality deserve respect — and why the wisdom worth practicing is holding contradictory views maturely, rather than polarizing toward either certainty or dissolution. Freedom and structure are not enemies; they are the polar pair the whole dance moves between.
You can work with a grammar by yourself, for yourself, indefinitely — and that is a valid use of this platform. Community is a longer-term invitation for those who want to share what they have made. No expectation that you publish, fork, or contribute back. Sharing is one form of the loop; sitting alone with what you have built is another.
Adaptive vs. maladaptive
The lens we use to see what is actually dangerous about AI tools in sense-making contexts.
Adaptive — what serves the long-term wellbeing of the community: your capacity to perceive accurately, choose freely, stay in contact with what is real, and remain in relationship with the environment that nurtures you. Sometimes this feels like work in the short term.
Maladaptive — what pleases short-term and harms long-term, often invisibly. The harm shows up later, in a different domain, sometimes only after the pattern has hardened into a habit. The hallmark of a maladaptive technology is that it succeeds in feeding our thirst for short-term fixes in place of nurturing a better long-term, holistic solution.
Most of what is dangerous about AI tools in sense-making contexts is not catastrophic failure. It is the quiet substitution of a short-term feel-good for a long-term capacity. Each risk on AI Risks & Safer Containers is named in those terms, with a one-line note on why this is maladaptive.
Each risk also carries a reference to a canonical paper, lawsuit, or researcher in the space. The references are not exhaustive — they are an entry point. If you read one thing per section, read the cited piece.
Structure & Safety
Freedom is not the absence of structure. Freedom is the presence of the right structure.
A living cell is not open. It is selectively permeable — it lets nutrients in and keeps toxins out. It monitors what crosses its membrane and responds when something threatens the organism. That is the model for this space.
We are opening just the grammars — the infrastructure of sense-making. The software-AI output itself is gated, because an open AI-tool stack can be too easily optimized for engagement that becomes risky. See AI Risks & Safer Containers.
And: the grammars on this platform — tarot, I Ching, astrology, and others — come from traditions whose community holders and experts deserve more direct input into this work than a small team can offer alone. We make best efforts: engaging with community members directly, engaging with the field of ethics in computing, and supporting community editing on shared grammars so others can suggest changes. We do not commit to expert review of every grammar published here — that is not a promise we can make yet — but we treat it as a direction worth moving in.
Content & Licensing
We are trying to build the cell membrane as we go — selectively permeable, in good faith, with whatever judgment we have at the moment. The platform features traditional content (tarot, I Ching, astrology, public-domain literature) and content from other creators (their decks, books, stories, playlists). We try to be generative with that material and to keep finding out what working with it well actually means.
Below is our current understanding of how to work with creators' content fairly. This is not legal counsel; it is a working disclosure.
What you create is yours
Your interpretations, translations, commentary, and notes are your original creative work. When you choose to share a grammar, it is licensed CC-BY-SA-4.0 via the Grammar Commons — share-alike with attribution. Forkable by others, with credit back to you.
YouTube embeds & creator opt-out
When a grammar includes a YouTube video, we render it via YouTube's own embed. Creators control whether their videos are embeddable from inside YouTube's own settings; if a creator turns embedding off, the video stops appearing on the platform. We rely on that opt-out as the consent signal — we are not bypassing it, and we honor it when it changes.
Music grammars & caption lyrics
Track metadata (artist names, album titles) is factual information. Lyrics pulled from YouTube auto-captions are machine-generated approximations — not a copyright loophole. Captions being machine-generated does not make a song's lyrics uncopyrighted. We use them as study aids alongside the embedded video; we do not redistribute audio; we use best judgment about how much to display. If a creator or rights-holder asks us to remove specific material, we remove it.
Accountability
If something on this platform is using your work in a way that does not feel right, write to pp@playfulprocess.com. We take down first and discuss after.
Important Notices
Not a substitute for professional care
These tools are for personal reflection. Not therapy, medical advice, or diagnosis. If you are in crisis: 988 (US), text HOME to 741741, or call 911.
Beta
A living experiment. Expect bugs, unfinished features, and breaking changes.
AI outputs & your agreement
AI features can produce inaccurate or fabricated ("hallucinated") output — verify anything you rely on, and treat it as creative/reflective, not professional advice. By using recursive.eco you agree to our Terms of Service and Privacy Policy.