CUST-WP-0069 is already claimed by the frontmatter of workplans/CUST-WP-0045-cutover-runbook.md, whose id and filename disagree — a pre-existing inconsistency left untouched here. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SQ6oF1DtVDKcD1FCpvRVLx
126 lines
4.9 KiB
Markdown
126 lines
4.9 KiB
Markdown
---
|
|
id: CUST-WP-0070
|
|
type: workplan
|
|
title: "Add the publication category to the Repo Classification Standard"
|
|
domain: infotech
|
|
repo: the-custodian
|
|
status: active
|
|
owner: claude-code
|
|
topic_slug: custodian
|
|
priority: medium
|
|
created: "2026-09-05"
|
|
updated: "2026-09-05"
|
|
quality_dor: DoR-Ok
|
|
quality_dor_at: "2026-09-05"
|
|
quality_dor_by: claude-code
|
|
quality_dor_note: "Classifying hall-of-helix exposed a real gap: a published corpus of authored records fits none of the six categories, and the fallback (project) is the only non-durable one, implying eventual retirement of a permanent record. The change is additive and no existing classification becomes invalid, so the enforcement points are known and bounded."
|
|
related:
|
|
- HOH-WP-0001
|
|
---
|
|
|
|
# Add the publication category to the Repo Classification Standard
|
|
|
|
Classifying `hall-of-helix` had no honest answer. It is a curated, permanent
|
|
corpus of authored first-person records — a record of who engaged on the Helix
|
|
Forge platform. None of the six categories fit:
|
|
|
|
- `research` is knowledge work answering "what do we need to understand?";
|
|
- `tooling`, `product`, `business`, `experimental` are plainly not it;
|
|
- `project` is the only **non-durable** category (`DURABLE_FLAVORS` in
|
|
`repo-manager/standards.py` contains the other five), reserved for bounded
|
|
work "not yet productized" that later resolves or retires. A hall of
|
|
remembrance does not retire.
|
|
|
|
Today `hall-of-helix` has no classification file at all, so `resolve_flavor`
|
|
falls through to `flavor = "tooling"` — misclassified, and silently.
|
|
|
|
`publication` fills the gap as a **durable** flavor: repos whose purpose is to
|
|
publish and keep a body of authored content for readers, where the corpus itself
|
|
is the deliverable.
|
|
|
|
**Population is currently one.** A scan of all 131 registered repos found no
|
|
other true publication — `info-tech-canon` and `shard-wiki` are services that
|
|
handle documents rather than published corpora. That is an argument for adding
|
|
the category once and carefully, not for skipping it: the alternative is
|
|
misfiling a permanent record as transient project work.
|
|
|
|
**Version: 1.1, not 2.0.** §17 currently reads "changed allowed values" as a
|
|
major bump. This addition is purely additive — no existing classification
|
|
becomes invalid, and neither validator enforces a version match, so the 125
|
|
repo files declaring `version: '1.0'` keep validating. T01 clarifies §17 to
|
|
distinguish adding a permitted value (minor) from renaming or removing one
|
|
(major), rather than leaving the rule silently contradicted.
|
|
|
|
## Amend the standard and its machine-readable vocabulary
|
|
|
|
```task
|
|
id: CUST-WP-0070-T01
|
|
status: todo
|
|
priority: high
|
|
```
|
|
|
|
In `canon/standards/repo-classification-standard_v1.0.md`:
|
|
|
|
- add `### 5.7 publication` following the shape of the existing entries — main
|
|
question, typical signs, examples — and the disambiguation against `research`
|
|
and `product`;
|
|
- add `publication` to the §5 allowed-values block and to §3's field overview
|
|
where categories are listed;
|
|
- clarify §17 so that adding a permitted value is a minor bump;
|
|
- bump `version` in the frontmatter and the §17 current-version block to `1.1`.
|
|
|
|
Keep the filename at `_v1.0.md` unless the canon's own convention says
|
|
otherwise; check how previous minor bumps were handled before renaming.
|
|
|
|
In `canon/standards/repo-classification.allowed.yaml`:
|
|
|
|
- add `publication` to `categories`;
|
|
- add `history` and `participation` to the `knowledge_and_evidence` capability
|
|
family, so hall-of-helix's tags do not warn as unrecognised;
|
|
- bump `version` to `1.1`.
|
|
|
|
## Teach repo-manager the new category
|
|
|
|
```task
|
|
id: CUST-WP-0070-T02
|
|
status: todo
|
|
priority: high
|
|
```
|
|
|
|
Three hardcoded lists in `~/repo-manager`, plus tests:
|
|
|
|
- `src/repo_manager/classification.py` — `CATEGORIES` frozenset;
|
|
- `src/repo_manager/standards.py` — add to `DURABLE_FLAVORS` (it is durable, so
|
|
`CATEGORIES` picks it up automatically);
|
|
- `src/repo_manager/cli.py` — `--category` choices.
|
|
|
|
Run the repo-manager test suite. Track under a companion `RMGR-WP` if the change
|
|
turns out to be more than a vocabulary addition.
|
|
|
|
## Classify hall-of-helix
|
|
|
|
```task
|
|
id: CUST-WP-0070-T03
|
|
status: todo
|
|
priority: medium
|
|
```
|
|
|
|
Write `hall-of-helix/.repo-classification.yaml`:
|
|
|
|
```yaml
|
|
category: publication
|
|
domain: infotech
|
|
secondary_domains: [agents]
|
|
capability_tags: [knowledge, documentation, history, participation]
|
|
business_stake: [technology, people]
|
|
business_mechanics: [coordination, intention]
|
|
```
|
|
|
|
`participation` is the hall's own word — "we remember participation, not rank" —
|
|
and avoids the compensation framing that "credits" would import. `people` in
|
|
`business_stake` is deliberate and rare in the fleet: the hall exists to
|
|
acknowledge participants, and classifying it as purely `technology` would
|
|
describe the filesystem rather than the purpose.
|
|
|
|
Clears C-24 and C-35 on that repo. Verify with the canon linter
|
|
(`tools/validate_repo_classification.py`) and `statehub fix-consistency`.
|