Simplify Readme

Author: martinstabe

README.md

@@ -1,281 +1,11 @@
 # martinstabe.github.io
 
-This repository now contains a parallel Eleventy implementation of `martinstabe.com`, migrated from the original Jekyll site while preserving the historical content archive and legacy permalink structure.
+A new Eleventy implementation of `martinstabe.com`, migrated from the original Jekyll site while preserving the historical content archive and legacy permalink structure.
 
-The original Jekyll source is still present in the repository:
+Some of the original Jekyll source is still present in the repository:
 
 - posts remain in `_posts/`
 - Sass sources remain in `css/` and `_sass/`
-- legacy tag metadata remains in `tags/`
-- the original Jekyll layouts and includes remain in `_layouts/` and `_includes/`
 
 The active Eleventy source lives in `src/`, and the generated deployable output is written to `dist/`.
 
-## Build And Run
-
-Install dependencies:
-
-```bash
-npm install
-```
-
-Build the site:
-
-```bash
-npm run build
-```
-
-Sync the FT RSS feed into the local stories YAML source:
-
-```bash
-npm run sync:stories
-```
-
-Run a local Eleventy dev server:
-
-```bash
-npm run dev
-```
-
-The current build target is `dist/`. The build script cleans `dist/` before each build so removed routes do not leave stale files behind.
-
-## URL Parity Check
-
-The repository includes a repeatable parity check for the deployment contract defined in the migration spec: any internal `martinstabe.com` URL reachable by recursively following links from `https://www.martinstabe.com/blog/` must also exist in the new build output.
-
-Build the site first:
-
-```bash
-npm run build
-```
-
-Run the full parity workflow:
-
-```bash
-npm run parity:blog
-```
-
-This writes:
-
-- `data/url_parity_live_blog.json`: crawl inventory from the live site
-- `data/url_parity_dist.json`: local route inventory from `dist/`
-- `data/url_parity_report.json`: parity diff report
-
-The full parity command exits non-zero if the live crawl finds internal paths that do not exist in `dist/`.
-
-You can also run the stages individually:
-
-```bash
-npm run parity:blog:live
-npm run parity:blog:dist
-npm run parity:blog:check
-```
-
-For bounded smoke tests while iterating on the crawler, you can cap the live crawl:
-
-```bash
-node scripts/crawl_live_blog.js --limit 200
-```
-
-## Deployment
-
-This implementation is intended for Cloudflare Pages.
-
-- build command: `npm run build`
-- output directory: `dist`
-
-The repository also includes GitHub Pages safeguards:
-
-- [`.nojekyll`](.nojekyll) at the repo root to bypass branch-based Jekyll builds
-- [`.github/workflows/pages.yml`](.github/workflows/pages.yml) to build with Node and deploy `dist/` directly via GitHub Actions
-
-If GitHub Pages is still enabled for the repository, it should be switched to GitHub Actions rather than branch-based publishing.
-
-## Implementation Structure
-
-### Eleventy App
-
-- [eleventy.config.js](eleventy.config.js)
-- [package.json](package.json)
-- [src](src)
-
-The Eleventy config:
-
-- uses `src/` as the input directory
-- uses Nunjucks for layouts and templates
-- outputs to `dist/`
-- passthrough-copies `img/`, `slides/`, `CNAME`, and the Google verification file
-- adds custom filters for relative URLs, human-readable dates, RFC 822 dates, and XML escaping
-
-### Layouts And Partials
-
-Shared templates are in:
-
-- [src/_includes/layouts/default.njk](src/_includes/layouts/default.njk)
-- [src/_includes/layouts/page.njk](src/_includes/layouts/page.njk)
-- [src/_includes/partials](src/_includes/partials)
-
-These replicate the old Jekyll page shell, header, footer, analytics, social, comments, and head metadata in Eleventy form.
-
-### Content Sources
-
-The Eleventy build does not duplicate the archive into `src/`. Instead it imports the existing source content:
-
-- blog posts from [_posts](_posts)
-- tag metadata from [tags](tags)
-- links archive data from [_data/links.yml](_data/links.yml)
-
-## Additional Features Added During Migration
-
-The Eleventy implementation includes behavior that did not previously exist as a single coherent layer in the Jekyll site.
-
-### Stories Section From FT RSS
-
-Stories sourced from the FT RSS feed can now be synced into [_data/stories.yml](_data/stories.yml) with [scripts/sync_stories.js](scripts/sync_stories.js).
-
-Each synced entry is stored as YAML and preserves manual overrides on subsequent syncs, including the `include: false` switch to exclude an item from the generated site.
-
-Included stories are rendered into the new `/stories/` section and linked from the main navigation.
-
-#### Updating Stories
-
-Source:
-
-- RSS feed: `https://www.ft.com/martin-stabe?format=rss`
-- local cache: [_data/stories.yml](_data/stories.yml)
-- sync script: [scripts/sync_stories.js](scripts/sync_stories.js)
-
-To refresh the Stories section:
-
-```bash
-npm run sync:stories
-npm run build
-```
-
-Notes:
-
-- the sync merges by feed `guid`
-- new feed items default to `include: true`
-- setting `include: false` on any item suppresses it from `/stories/`
-- manual edits in `_data/stories.yml` can be used to override display text or links before rebuilding
-- the generated Stories page links outward to the canonical FT URLs
-
-### Graphics Section From FT Asset Database
-
-Graphics sourced from the FT elections newsbox asset database can now be synced into [_data/graphics.yml](_data/graphics.yml) with [scripts/sync_graphics.js](scripts/sync_graphics.js) and [scripts/export_graphics.R](scripts/export_graphics.R).
-
-The sync filters records to `flourish_author == "Martin Stabe"`, transforms the fields required for the site, preserves manual `include: false` overrides, and renders included entries into the `/data-visualisation/` section as outbound links to FT.com.
-
-#### Updating Data Visualisation
-
-Source:
-
-- R data file: set `GRAPHICS_SOURCE_URL` in your environment before running the sync
-- local cache: [_data/graphics.yml](_data/graphics.yml)
-- sync scripts: [scripts/sync_graphics.js](scripts/sync_graphics.js) and [scripts/export_graphics.R](scripts/export_graphics.R)
-
-To refresh the Data Visualisation section:
-
-```bash
-npm run sync:graphics
-npm run build
-```
-
-The sync currently:
-
-- downloads the remote `.rds` file
-- filters rows to `flourish_author == "Martin Stabe"`
-- transforms the selected fields into YAML entries
-- stores the results in `_data/graphics.yml`
-- preserves manual `include: false` overrides on existing items
-
-### Tag Deduplication And Alias Merging
-
-Multiple tag tokens can now map to the same rendered tag page simply by pointing their legacy `tags/*.md` files at the same permalink.
-
-Example:
-
-- `ft`
-- `financialtimes`
-- `financial-times`
-
-This affects:
-
-- the generated tag index
-- individual tag pages
-- the tag footer on each post
-- inline legacy tag links embedded inside imported HTML posts
-
-### Canonical Post Footer Tags
-
-Post pages are generated through [src/posts.njk](src/posts.njk) using metadata prepared in [src/_data/posts.js](src/_data/posts.js).
-
-The footer tag list now uses canonical tag definitions rather than raw front matter values, so a post tagged `ft` can render as `Financial Times` and link to the canonical merged tag page.
-
-### Legacy Inline Tag Link Rewriting
-
-Some older imported posts, especially the `links-for-*` archive entries, contain literal HTML with embedded `/tags/.../` links in the body.
-
-Those links are now rewritten during the Eleventy import step so they resolve to the canonical tag permalink rather than an obsolete or duplicate path.
-
-### Special Handling For Delicious Archive Posts
-
-Many `links-for-*` posts are already HTML rather than Markdown. The importer in [src/_data/posts.js](src/_data/posts.js) detects those entries and bypasses Markdown rendering so they do not get mangled into escaped code blocks.
-
-### Relative Internal Links For Local Review
-
-The Eleventy build uses a custom `relativeUrl` filter so internal links are written relative to the current page. This makes manual inspection of the generated `dist/` output work correctly when served from a local static server.
-
-Canonical and feed URLs remain absolute where appropriate.
-
-### RSS Feed Recreation
-
-The RSS feed is generated by [src/feed.njk](src/feed.njk), using Eleventy data and custom XML/date filters to approximate the old Jekyll feed behavior.
-
-### Links Archive Rebuilt From Data
-
-The `/links/` page is now generated from [_data/links.yml](_data/links.yml) through [src/_data/links.js](src/_data/links.js) and [src/links/index.njk](src/links/index.njk), rather than relying on Jekyll template iteration.
-
-### Jekyll Sass Reused In Node Build
-
-The original Jekyll Sass has been retained and compiled via Node:
-
-- [scripts/build_sass.js](scripts/build_sass.js)
-- [css/main.scss](css/main.scss)
-- [_sass](_sass)
-
-This preserves the historical styling rather than replacing it with a new CSS layer.
-
-### Clean Build Output
-
-The build now removes `dist/` first using [scripts/clean_dist.js](scripts/clean_dist.js). This prevents stale generated pages from surviving route changes, which is especially important for merged tag aliases.
-
-### Footer Social Update
-
-The Eleventy footer partial at [src/_includes/partials/footer.njk](src/_includes/partials/footer.njk) has been updated to point to Bluesky instead of the legacy Twitter profile, including a new icon and layout adjustments to accommodate the longer handle.
-
-## Current Page Coverage
-
-The Eleventy build currently generates:
-
-- home page
-- blog index
-- individual post pages
-- tag index
-- tag detail pages
-- links archive
-- about page
-- RSS feed
-
-Static assets are passed through for:
-
-- `img/`
-- `slides/`
-- `CNAME`
-- `googlea1fed2510060dd96.html`
-
-## Notes
-
-- The original Jekyll site remains in the repository for reference during migration.
-- The generated `dist/` directory is a build artifact and should not be edited directly.
-- The Sass pipeline currently emits deprecation warnings from the legacy stylesheet source, but the build completes successfully.