---
title: Status and freshness — Gwinnett Index
as_of: "2026-08-13"
canonical: "https://www.gwindex.net/status"
licence: "CC0-1.0"
---

# Status and freshness

What is loaded, what has actually been checked, and what is knowingly missing. An index that
hides its staleness is worse than one that admits it.

## Resolver accuracy

Scored against 0 probe points taken from the county's own zoning layers.

| Confidence | Probes | Correct | Rate |
|---|---|---|---|
| high | 1546 | 1546 | 100.00% |
| low | 361 | 347 | 96.12% |
| unresolved | 8 | 0 | 0.00% |
| overall | 1915 | 1893 | 98.85% |

When the resolver reports **high** confidence it has never been wrong. Every error falls inside
the low-confidence band, and every one was within 141 m of a jurisdiction line — annexation lag,
since cities annex continuously while Census boundaries update yearly.

## Coverage

| Metric | Value |
|---|---|
| Jurisdictions | 19 |
| Land-use cases | 11849 |
| — county (Board of Commissioners) | 11739 |
| — City of Duluth (agenda-mined) | 110 |
| Cases with a source link | 11598 |
| Cases with a parcel PIN | 10804 |
| Code sections | 858 |
| Code tables | 18 |
| — verified against the source page | 1 |
| Meeting documents | 0 |
| Build data source | supabase (postgrest) |

## Known gaps

- **17 of 18 code tables are unverified.** Only Table 2-B has been checked
  against the rendered source page, and even its merged PUD and CBD rows remain unreliable.
- **0 applicant name pairs await human review**, so developer case counts are lower bounds.
- **Duluth minutes are 57–69% scanned** with no text layer. OCR text is stored separately and
  marked as a reconstruction, never as a quotation of the record. The OCR pass is incomplete.
- **Duluth case fields are agenda-mined, not database-sourced.** The city runs no case tracker,
  so these rows are pattern-matched out of agenda and minutes PDFs. Field-level quality is poor,
  measured across all 110: 68 have a "location" with no street number
  (15 are plainly not addresses at all), 58 have a "request" that is just
  the word ORDINANCE, 24 have an applicant field that ran on into a mailing address, and
  only 16 carry a zoning district. Treat each as an index entry pointing at a PDF, not as
  a structured record: the linked document is the authority, the fields are a finding aid.
- **16 municipalities have boundaries but no corpus.** They resolve correctly; there is no code
  or case data behind them yet.
- **Meeting body text is not published here** (12.5 M characters); only document metadata is indexed.

## Legal posture

| Content | Posture | Basis |
|---|---|---|
| Ordinance and plan text | Mirrored in full | Georgia v. Public.Resource.Org (2020) — edicts of government |
| Case records | Mirrored in full | Facts; Georgia Open Records Act |
| County parcel and zoning geometry | NEVER rehosted | Gwinnett GIS licence forbids redistribution — queried live, linked out |
| Jurisdiction boundaries | Hosted | US Census TIGER — federal, public domain |
| IBC/IRC base text | NEVER indexed | ICC copyright; Georgia's amendment packets indexed instead |