From d0c8194dab69c28c2a63c3ff486c3fb5ad9d508f Mon Sep 17 00:00:00 2001 From: Matt McKay Date: Mon, 7 Sep 2026 10:05:59 +1000 Subject: [PATCH] AGENTS: the manifests are the upstream-delta register of record The register issue (#39) served while the static migration ran. The migration completed 2026-08-18 (40 of 40 repointed, strict audit green 2026-08-31) and the review step ran 2026-09-01, so the issue closes on 2026-09-07 (ruled by @mmcky). The policy line now says the manifest's integrity.upstream field and the generated catalog are the register, and cites #39's history as the worked example of the three steps. Co-Authored-By: Claude Fable 5.1 --- AGENTS.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/AGENTS.md b/AGENTS.md index 13483c3..9505979 100644 --- a/AGENTS.md +++ b/AGENTS.md @@ -55,7 +55,7 @@ Cross-repo repoints are worked from [`QuantEcon/workspace-lectures`](https://git ### Repointing a lecture — one scope rule -- **A migration moves bytes; it does not update them.** Land the copy the lectures already consume, validated byte-identical — that is what makes a repoint provably unable to change a figure. If the committed file differs from what upstream publishes today, migrate it unchanged anyway, record the delta in `integrity.upstream` **and** in the register at [#39](https://github.com/QuantEcon/data-lectures/issues/39), and leave the decision for after the migration. Adopting a newer vintage changes lecture output and is an author's call, not an infrastructure one — and per "Corrections vs vintages" below it gets a **new filename**, never a silent replacement. +- **A migration moves bytes; it does not update them.** Land the copy the lectures already consume, validated byte-identical — that is what makes a repoint provably unable to change a figure. If the committed file differs from what upstream publishes today, migrate it unchanged anyway, record the delta in `integrity.upstream` — **the manifest is the register of record**, and the generated catalog shows the delta from the day it is found — and leave the decision for after the migration. The upstream-delta register issue [#39](https://github.com/QuantEcon/data-lectures/issues/39) was that register while the migration ran; it closed on 2026-09-07 once the migration completed (40 of 40 repointed 2026-08-18, strict audit green) and its review step had run (2026-09-01: `mpd2020.xlsx` and `life-expectancy-vs-gdp-per-capita.csv` kept deliberately, `business_cycle_data.csv` refreshed by #112). Its history is the worked example of the three steps; new deltas need no issue — set `integrity.upstream` in the same PR that finds them, and the catalog is where a reviewer looks. Adopting a newer vintage changes lecture output and is an author's call, not an infrastructure one — and per "Corrections vs vintages" below it gets a **new filename**, never a silent replacement. ### Corrections vs vintages