Move the naming-scheme spec from notes/dev/specs/ into docs/guides/ so it
publishes to docs.makerfloss.eu, add it under the Hardware nav, and link it
from the editing guide. Repoint the stale references in CLAUDE.md and the
migration plan to the new path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Mirror the auto-indexed per-host pattern for a new docs/services/
category, seeded with the six things currently deployed on or around
makerfloss.eu: docs, slides, forgejo, gandi-dns, marp, mermaid.
Generator/hook generalisation:
- scripts/gen_overview.py: replace the hardcoded `hostname` check
with a configurable `key_field` (default: hostname). Add a generic
`key-link` column kind (replaces the old `hostname-link`) and a
`url-link` kind that renders the value as a clickable link.
- scripts/overview_config.yml: declare hardware's key_field, then add
a `services` block (key_field=name, its own kind/status enums,
grouped by kind for the index table).
- scripts/mkdocs_hooks.py: route by `page.file.src_uri` so each
hardware/* page gets a "Specs" table and each services/* page gets
a "Service" table; both share the helpers in gen_overview.
Wiring:
- Makefile: docs-index and docs-check now regenerate and drift-check
both indices.
- .forgejo/workflows/docs.yml: same on the CI runner.
- mkdocs.yml: add Services to nav.
- README.md, CLAUDE.md: list services/ in the repo-layout block.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adopt one filename convention everywhere except hardware host pages
(those are pinned to their hostnames by the index validator):
- ASCII lowercase
- kebab-case word separator
- dated docs: YYYY-MM-DD-label.md (hyphen between date and label)
- English regardless of content language
Renames:
docs/makerFLOSS_house_rules.md -> docs/house-rules.md
docs/infrastruktur/ -> docs/infrastructure/
docs/infrastructure/vps-og-dns.md -> docs/infrastructure/vps-and-dns.md
docs/presentations/2026-05-11_messaging.md -> docs/presentations/2026-05-11-messaging.md
docs/presentations/2026-05-11_SoMe-taxonomi.md -> docs/presentations/2026-05-11-social-media-taxonomy.md
notes/todo/2026-04-14_todo.md -> notes/todo/2026-04-14-todo.md
notes/todo/OM_services.md -> notes/todo/om-services.md
notes/todo/indkøbsliste.md -> notes/todo/wishlist.md
notes/communications/2026-03-16_facebookpost.md -> notes/communications/2026-03-16-facebook-post.md
Other touches:
- Document the convention in CLAUDE.md under Working Norms.
- Refresh the stale Repository Structure block in CLAUDE.md to reflect
the docs/ vs notes/ split introduced in f6d589e.
- Update the House rules link in docs/index.md and the nav entry in
mkdocs.yml to the new house-rules.md path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The docs/ tree previously conflated published-site content (3 pages
were in mkdocs.yml nav) with working notes (~18 files that just sat
in the repo). Restructure so each tree means one thing:
- docs/ everything here is built and shipped to docs.makerfloss.eu.
Adds docs/presentations/ for the two Marp decks
previously living under docs/møder/.
- notes/ repo-only working material, not built. Contains
meetings/, todo/, dev/ (was docs/superpowers/), and
communications/ (the launch Facebook post).
- sandbox/ test-mermaid.md, the Marp/Mermaid pipeline sandbox.
Other touches:
- Drop "_noter" suffix on meeting filenames; drop "_presentation"
from the messaging deck's basename for symmetry with SoMe-taxonomi.
- Update CLAUDE.md and docs/index.md path references.
- Drop the now-redundant --exclude-dir=superpowers from
build-slides.sh since superpowers/ is no longer under docs/.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Delete two stale single-meeting todos: 2026-03-16_todo.md and 2026-05-05.md.
- Rename `2026-04-14 TODO.md` -> `2026-04-14_todo.md` to match the
underscore convention used by the other dated files, and update the
CLAUDE.md reference.
- Remove the two Marp/CSS fit-test sandboxes (labdesign-fit-test.md
and 2026-05-11_messaging-presentation-fit-test.md); the responsive
experiments were never folded back into the canonical decks.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
