The three tiers
Every entry in this register carries a tier, and the tier is a statement about evidence, not about importance. A STATED influence may well be the deepest one on the list — the tier says only how well the site can currently show it. 25 entries: 15 traced, 7 stated, 3 discovered and awaiting confirmation. Those counts are recomputed from data/influences.json on every build.
TRACED
Corpus evidence exists today. The entry lists the files, and the claim can be checked against them.
STATED
On Dinis Cruz's own list, and thin or absent in the corpus. Published as a stub with the research plan visible.
DISCOVERED
Found by mining, never claimed by Dinis Cruz. A falsifiable claim about him, pending his confirmation.
TRACED — corpus evidence exists
The mining run found files. The entry lists them by path, and a reader with the repositories open can go and check whether they say what the entry says they say.
What TRACED does not mean. It does not mean the influence is proven, and it does not mean it is important. It means the site can show you something. A trace table shows that a pattern from an anchor work appears at a named place in the estate; it cannot show that the anchor work caused it. That limit is stated on the Popper entry, which is the influence that explains why this site is shaped the way it is, and again on the disclosure page.
7 of the entries in this register were surfaced by mining rather than named by Dinis Cruz. Four of those now sit at TRACED — three because the evidence was too strong to leave them anywhere else, and one, Design, because Dinis Cruz confirmed it in writing on 25 August 2026. Those carry a discovered → traced badge, because how an entry got here is part of what it is.
STATED — on Dinis Cruz's list, and not yet in the corpus
7 entries. Most of them have zero corpus evidence: the mining run found nothing at all, and the entry publishes an empty evidence block saying so.
This is the roadmap, not the debt. Dinis Cruz's own plan was to write a detailed briefing document about each one, and the site's job is to make that plan addressable rather than to wait quietly. Every stated entry has an open thread in the briefing queue, a published research plan, and a tier that will move when the document arrives.
Two reasons the corpus is empty, and they are different problems:
- The material exists and was not reachable. Two decades of earlier writing — the O2 Platform-era blog, OWASP talks, conference decks, the earlier books — were outside the scan. This is a tooling problem with a known fix, and it is R3.
- The material never existed. Playing in a band has no public record, no publication to mine and no third party to ask. No amount of access closes that one; only Dinis Cruz can.
A stub is published, not hidden. Every one carries Block 1 and Block 3 at best current knowledge, an explicit awaiting-briefing marker, and the research plan — and where a principle is a hypothesis constructed from public material rather than from Dinis Cruz's account, the entry says so in the principle itself.
DISCOVERED — the corpus said so and Dinis Cruz never did
These entries were found by mining and were never on anybody's list. The claim that they shaped the work rests entirely on the evidence in the entry, and they stay at DISCOVERED until Dinis Cruz confirms or corrects them.
This is the tier the site is proudest of and the one that most needs explaining. A register that only records what its subject already believed about himself is an autobiography. Publishing claims about him that he has not agreed to — and asking him in public to strike the ones that do not belong — is the site's falsifiability applied to itself.
It has already worked once, before this site existed. Design, with a capital D was surfaced by mining a role definition and three briefs; Dinis Cruz confirmed it on 25 August 2026, the day the commissioning pack shipped. That was the first DISCOVERED → confirmed transition, and it is the model for the rest.
Tier movement is this site's changelog
The interesting events here are not new pages. They are transitions, and each one is a dated row on the movement log:
| Transition | What causes it | What it changes |
|---|---|---|
| STATED → TRACED | A briefing document arrives and is integrated, or the earlier material becomes reachable and the scan finds something | The stub becomes a full entry: evidence, a trace table, gaps and a checklist. Its briefing thread closes |
| DISCOVERED → TRACED | Dinis Cruz confirms the influence is real | The tier moves and the entry keeps a discovered → traced badge — how it got here stays visible |
| DISCOVERED → removed | Dinis Cruz says it was a citation, not an influence | The entry goes, and the movement log records that it was here and why it left. This has not happened yet |
| A trace row: absent → implemented | The estate builds one of the gaps an entry specified | The row flips, and an influence page turns out to have been a backlog that got worked |
A provenance site whose provenance changed silently would be self-refuting, which is why the counts on this page are recomputed from the register on every build rather than typed, and why the release gate fails if a page disagrees with the data.
How the counting works, exactly
Two numbers on this site can be read two ways, so both are published:
- Tier is where an entry sits now: 15 / 7 / 3.
- Origin is how it got into the register. 7 entries were surfaced by mining rather than named by Dinis Cruz — the 3 still awaiting confirmation, plus four already promoted.
The commissioning pack's prose used an earlier snapshot (12 traced, 7 stated, 5 discovered) taken at mining time, before the promotions and before the Design confirmation. The pack says its own seed JSON is authoritative and that the counts should be regenerated from it, and this site does exactly that — which is why the numbers here differ from the numbers in the document you can read at /documents/. Both are published; the computed one wins.
Tier is not the same as status
A second axis, easy to confuse with the first. Tier is about evidence; status is about which of the seven blocks an entry has earned:
| Status | Count | What it is |
|---|---|---|
| full | 13 | All seven blocks, or as many as the evidence supports |
| link-out | 3 | Blocks 1–3 plus a pointer where the trace table would be. The trace of that influence is an entire sibling site, and duplicating it would break the deconfliction rule |
| stub | 9 | Blocks 1 and 3 at best current knowledge, plus the research plan |
The two axes are independent. A TRACED entry can be a link-out — Wardley is the most heavily evidenced influence in the corpus and has the shortest entry, because its material belongs to a sibling. A DISCOVERED entry can be full: Christopher Alexander has a trace table with one implemented row and two absent ones, which is a more useful thing to publish than nothing.