newsroom.sgit.ai / network

The network: sibling boundaries

The largest risk to this site is not scarcity — it is duplication. Five sibling sites already own adjacent ground. This page states, for each one, what this site would duplicate if it were not careful, and where the actual boundary sits.

The five siblings

SiblingWhat newsroom would duplicateWhere the boundary sits
sgit.aiThe entire delivery substrate: articles-as-vaults, mini-site deployment, static vault projections, vault CICite, don't re-explain. Zero words here on S3, CloudFront or loaders. This site also inherits its known problem — see risks below
pki.sgit.aiPKI-signed claims, key registries, chain of trust, the identity spectrumNewsroom owns the application of PKI to claims and citations — the source-attribution chain in provenance, and CC-Signed, a rights argument PKI would not naturally carry on its own
nhi.sgit.aiAgent trust scores, web of trust, non-human identity graphsNewsroom borrows who is the actor but owns whose agenda, whose funding, who is citing — an editorial model of attribution, not an identity model
graphs.sgit.aiThe grounding ladder, node-type formulas, graphs-of-graphs, Wikidata anchoring, decompilationNewsroom owns the news ontology — stories, claims, sources, authors — plus two things graphs.sgit.ai would state generically and news must state urgently: correction propagation and citation edges typed by faithfulness
sg-sentinel.sgit.aiSource reputability scoringSentinel scores infrastructure; newsroom scores sources and authors. Shared principle, one cross-link: reputation is context, not verdict
docs.diniscruz.ai— this is prior art, not a sibling to bound againstThe site's own library. Linked forward from the source, never redirected — the old dated URLs stay exactly where they are; see the provenance contract

The Risk Mandate inversion — the most important boundary

A cluster of the source material's strongest documents — the evidence-economy briefs, credibility calibration, the fact-creator payment mechanism — were originally written for a corporate risk-management product, framing news as the evidence supply for risk graphs. This site inverts that. Risk management is one customer of the future-of-news stack, not its parent. The material is genuinely dual-use, but published under a risk-management frame it reads as a compliance feature; published here it reads as a thesis about journalism. Every republished document from that cluster has had its risk-register framing stripped for that reason — the news argument stands on its own, and the honest reciprocal statement is simple: this is what the same machinery looks like pointed at corporate risk instead of at a news story. A dedicated page for that comparison is future work, tracked on comms.

Inherited risk: crawler invisibility

A client-side-assembled vault page, served the way sgit.ai serves one, can be invisible to a crawler. Measured and documented as a known limitation of the delivery substrate this site cites rather than re-explains. For most sites that is a nice-to-have fix. For a news site it is close to existential — almost all of a news site's reach depends on being findable by search and by other agents, and a site that argues for provenance while being unreadable by the tools that would check it has undermined its own case before a reader ever arrives. Decide rendering before content; this is flagged here rather than left implicit.

Seven open questions, published unresolved

Following the house convention every sibling site uses: a question this material cannot yet answer is stated and numbered, not smoothed into confident prose.

#QuestionWhere the material gets closest
Q1Who pays for contextual validation, and what stops it becoming a toll on inconvenient facts?"In companies that's okay… but in the real world at the moment we don't have that." The commons has no maintenance budget
Q2What is the unit that actually gets paid — the claim, the paragraph, the evidence pack, or the graph?Named as "facts, trust, and evidence packs" throughout, never resolved to one billable unit
Q3Who decides a citation edge is contradicts rather than partially supports?Typed citation edges are proposed; the adjudication mechanism is not specified
Q4Does agenda-tagging survive contact with a motivated reader?The material asks this of itself — see the self-critique published rather than buried
Q5What happens when the author refuses to be the oracle, or is dead?The Ericsson case in the claim that would not die is exactly this, and it is not resolved
Q6Can a newsroom that publishes its own costs survive competitors that do not?The £8.40 page is a transparency asset and a commercial exposure, unaddressed either way
Q7Is CC-Signed enforceable in any jurisdiction, or is it a norm dressed as a licence?The brief asserts the legal stick; no legal review exists anywhere in the material

Six honest tensions

  1. Provenance is the product — and provenance is expensive. The £8.40 story cost is the argument and the objection in the same number.
  2. Corrections propagating is a feature until it is a liability. A graph that can answer "what rests on claims since corrected" can also be subpoenaed for the answer.
  3. "More humans, not fewer" is a design principle, not an economic result. Nothing in this material tests whether it actually holds once real budgets are involved.
  4. Selling trust makes the seller a target. A warranted fact-certifier has created a liability surface the material notes but does not size — see trust as a service.
  5. This site inherits sgit.ai's crawler-invisibility problem — see above. For a news site, existential rather than cosmetic.
  6. Two licences over one thread. docs.diniscruz.ai is CC0; *.sgit.ai is CC BY 4.0. The resolution this site argues for, stated on the disclosure page.

For an agent

This site's remit is the news ontology, correction propagation, and typed citation edges. For the identity and key-signing mechanics behind "signed claims," read pki.sgit.ai. For the graph grounding ladder this site's ontology plugs into, read graphs.sgit.ai. For the vault delivery mechanics, read sgit.ai. Do not attribute an identity, PKI or vault-mechanics claim to this site — it borrows those concepts and does not restate them.