Loomworks and Indigenous Knowledge

1. The observation that started this

A YouTube series on Ancient Australia — landscape, deep history, the connection of indigenous peoples to country. Interesting, well made, and narrated entirely by a non-indigenous voice reading researched facts. The people whose history it is appear as subject matter, not as tellers.

The technology to change that is already in those communities: phones, cameras, connectivity. What is missing is not equipment. It is a place to put the knowledge that the community itself owns and governs — where recording something does not mean surrendering it.

2. What Loomworks is, in one page

Loomworks is a platform for what DUNIN7 calls an engagement — a long-running body of work with a memory that does not forget. Each engagement has four rooms:

• Memory. Everything contributed — spoken accounts, documents, photographs, corrections — accumulates here. Every entry carries its provenance: who said it, when, from what source. Nothing is silently overwritten. When an account is corrected or a later telling supersedes an earlier one, both remain, with the relationship named.
• Manifestation. The memory organized at a moment in time — the current shape of what is known.
• Shaping. The knowledge arranged for a specific reader. The same memory can be shaped one way for the community's own young people, another way for researchers under agreement, another way for the general public. These are different shapes of the same body, each deliberate.
• Rendering. Producing the actual artifact — a document, a website, an audio archive page — from a shape. Nothing renders outward without approval.

Two governing principles matter most for this idea:

• Operator-authority. A human authority — here, the community or its designated custodians — approves every consequential step. The system surfaces and suggests; it never publishes, releases, or changes state on its own.
• Corrections preserved, not smoothed. Oral traditions carry variant tellings and evolving accounts. Loomworks keeps the trajectory, not just a flattened "final version." That is a feature, not a compromise.

Access control uses a model (GRANTHA) whose single primitive is the grant: a person holds a grant to see a thing, and the list of who holds grants is never itself readable. This matters more than it sounds — see §4.

3. Why the fit is real

Provenance is the point, not a feature. Indigenous knowledge systems care intensely about who has the right to tell this — which elder, which family, which country. A platform where every assertion carries its tie-to-source matches how the knowledge already works. Most digital archives strip that; Loomworks is built around it.

The community holds the authority seat. The Operator role puts approval of every outward-facing step in the community's hands structurally, not as a policy promise. Nothing becomes public because a default was left on.

Restricted knowledge stays restricted — and invisibly so. Much Aboriginal knowledge is lawfully secret: men's business, women's business, sacred sites, knowledge tied to initiation status. A conventional permissions system (groups, access lists) leaks the shape of the secret — outsiders can see that a restricted category exists and who belongs to it. The grant model does not: holding a grant is private, the holder-set is unreadable, and to a person without the grant the material simply does not appear. Loomworks' surfaces follow the same rule — nothing disabled or grayed-out; what you cannot access does not exist for you.

Graduated disclosure is native to the design. The Shaping room is precisely the mechanism for "the community sees everything; the school program sees this; the public sees that." These are not exports with things deleted; they are deliberate arrangements of the same governed memory.

Variant tellings are respected. Because corrections and supersessions are preserved rather than smoothed, an archive can hold multiple accounts of the same story from different families or different decades without forcing one to be "the truth."

4. Where the naivety risk sits — and the correction

The instinctive tagline was: taking the threads of verbally passed-down history and weaving a fabric for public knowledge.

That framing has the destination wrong, and communities will hear it instantly. For much indigenous knowledge, the destination is protection and controlled transmission — to the next generation, on the community's terms. Public rendering is a small, carefully gated subset, and for some material the correct amount of public rendering is zero, forever.

If the pitch leads with the public layer, it reads as "extraction with better tooling" — and two centuries of anthropologists, missionaries, and film crews have made that reflex entirely earned. The corrected framing:

The community owns a memory that does not forget. What leaves it is theirs to shape and approve. Public knowledge is one possible output, never the mission.

The public interface the original idea imagined — a public information layer integrating with the Render layer — is exactly right architecturally, but it is a render-type the community may choose to switch on, not the reason the platform exists.

5. What already exists (prior art worth knowing)

This need is validated; tools exist; none has Loomworks' architecture.

• Mukurtu CMS (USA/Australia, Washington State University) — a content management system built specifically for indigenous communities, with cultural protocols controlling who sees what. Widely used, well respected. It is an archive with access rules.
• Ara Irititja (Australia) — a digital archive run for and by Aṉangu communities in the APY Lands since the 1990s. Decades of proof that communities will contribute and maintain a digital knowledge store when they own it.
• Te Hiku Media (New Zealand) — Māori-governed media and language organization; the strongest reference for the sovereignty posture. They famously refused to hand their language data to large tech companies and built their own speech tools under Māori data-sovereignty principles (kaitiakitanga — guardianship).
• Governance frameworks — the CARE Principles for Indigenous Data Governance (Collective benefit, Authority to control, Responsibility, Ethics) internationally, and OCAP (Ownership, Control, Access, Possession) in Canada. Any credible offering will be measured against these; Loomworks' Operator-authority and provenance model align with them naturally.

The gap: existing tools are archives. They store and gate. None has a governed path from memory through deliberate shaping to living artifacts — school materials, community sites, audio collections — with provenance intact the whole way and the community approving each step. That is the actual opening.

6. Structural realities of a non-profit path

Indigenous-led, or it does not happen. The delivery organization has to be governed by the communities it serves — in Australia that means working through bodies like land councils and community-controlled organizations; in New Zealand, iwi trusts. DUNIN7's correct position is substrate provider — the technology under an indigenous-led front door — not the front door itself.

Data lives where the community says. Data sovereignty is non-negotiable in this space. The memory should sit on infrastructure the community (or its trusted indigenous-led organization) governs. Loomworks' existing self-hosted / bring-your-own-keys posture fits; a "we host everything in Miami" posture does not.

Funding channels exist. Australia has indigenous cultural heritage funding streams (federal and state), philanthropic foundations active in language and culture preservation, and native title bodies with cultural obligations. New Zealand has iwi settlements funding cultural projects and government language-revitalization money. A non-profit vehicle carrying the platform to communities is fundable — but it must be indigenous-controlled to access most of these channels credibly.

Elders and time. The knowledge holders are aging. Every year of delay loses accounts permanently. This is the same urgency that drives DUNIN7's legacy-systems knowledge work in banking — retiring experts, undocumented knowledge — transposed to a domain where the stakes are cultural survival rather than system maintenance. The mechanics of capture (spoken contribution, interpretation into held assertions, custodian review before anything enters memory) are the same mechanics Loomworks already builds.

7. What a first step looks like

Not a product launch. A conversation.

1. Find one community-controlled organization already doing cultural preservation work — ideally one already frustrated with the limits of a plain archive — and listen to what they actually need. Sydney is a good place to start that search; the Australian institutional landscape (AIATSIS, state library indigenous units, land councils) is navigable from there.
2. Show, don't pitch. A small demonstration engagement — with placeholder content, never real cultural material — showing memory-with-provenance, restricted material that is simply invisible without a grant, and one deliberate public render. Twenty minutes of demonstration beats any document.
3. Let the governance question lead. Open with "you would hold the authority seat; nothing leaves without your approval; the data sits where you say" — and let the community test whether the structure honors that claim. It does, and that is the entire pitch.
4. Structure follows interest. If a genuine partner emerges, then the non-profit vehicle, funding applications, and hosting arrangements get designed with them, per CARE/OCAP principles — not presented to them ready-made.

8. The honest answer to "am I being too naive?"

No — about the fit. The architecture genuinely matches the problem, the need is real and urgent, and the existing tools leave the gap open.

Yes — if the framing stays "weaving a fabric for public knowledge." Inverted — the fabric is woven for the community first; the public sees only what the community chooses to render outward — it stops being naive and becomes the most defensible use of the platform that exists.