JCRT #

Developed by Adam DJ Brett

Working locally #

• npm run dev serves a quick build (5 items per collection) on http://localhost:8080; npm run dev:full serves everything.
• npm run build is the production build: tests → nanoids:check → check.mjs pre → Eleventy. CSS purge/minify, Pagefind and the post-build checks run inside Eleventy's eleventy.after hook (eleventy.config.js), production builds only; a failing check fails the build.
• npm test runs the script tests; the build runs them too.

Sitemap + IndexNow #

• The OAI-PMH feed, its records index, the DOAJ feed and the citation sitemaps are built, not tracked: _data/oaiFeeds.js computes them and content/sitemaps/oai-feeds.11ty.js writes them byte-exact under _site/sitemaps/.
• Build-time sitemap integrity check: npm run sitemaps:check; OAI-PMH protocol checks over the built feed: npm run oai:validate (both also run inside the build).

Standard.site / AT Protocol #

• npm run sequoia:publish is what the workflow runs; only new or edited documents are written (staged files carry their existing AT-URI, so the CLI skips the rest). npm run sequoia:publish:dry shows the plan without touching the PDS. npm run sequoia:sync is the repair path: it pulls AT-URIs back from the PDS into _data/standardSiteRecords.yaml when the workflow's commit step failed — do not run it until the orphaned records are pruned (npm run sequoia:prune lists them; node scripts/sequoia.mjs prune --max 4000 deletes, by hand, within the PDS rate budget).
• The atproto: line in content front matter is a frozen legacy copy; nothing reads it. _data/standardSiteRecords.yaml is the only source of <link rel="site.standard.document">.

Credits #

Full list with licenses in CREDITS.md. In short:

• Icons are inline SVG from Phosphor Icons (MIT) and, for the ORCID iD glyph, Academicons (SIL OFL 1.1). No icon font is loaded; Font Awesome has been retired.
• Layout descends from HTML5 UP's Editorial (CC BY 3.0); system font stacks follow Modern Font Stacks.
• Build performance optimization audit and recommendations by Brennan Kenneth Brown
• Credit to 11tybundle.dev for the build-speed pattern that inspired the faster local JCRT workflow. Their latest-issue / cache-first approach helped reduce local iteration time by roughly 60-70% in practice.

Next Steps #

5. idea: use RT for book reviews
6. add pagination to 24.2 and 25.1 pdfs
7. Apply the ORCID iDs. Review output/orcid-candidates.yaml, set confirm: true on accepted rows, run node scripts/orcid-lookup.mjs --apply (the script was removed from the tree on 2026-09-23 as unused; restore it with git checkout 8757b76d2 -- scripts/orcid-lookup.mjs when resuming). Suggested order: the 29 rows whose reason says ORCID record lists a JCRT work (near-certain), then the rest of the high rows, then the 17 medium by hand. Known-soft: andrew-w-metcalfe, james-c-james-craig-livingston, mark-murphy are probably wrong; simon-clark is borderline (shares only "london").
8. Sweep author affiliations for typos. Auditing 109 rows found three. Where a high ORCID match's institution disagrees with ours, one of the two is wrong.
9. Look for more duplicate author files. The four found were surfaced by accident (two slugs sharing one ORCID iD). A deliberate pass would likely find others.
10. Clean up orphaned AT Protocol records for the four deleted author pages — local keys are gone so the build passes, but nothing removed them from the PDS.

EBSCO #

contact ebsco have link to pages or pdfs

increase incoming links #

CHANGELOG #

1. check author pages especially mine prove they are working

2026-08-30 — ORCID tooling, author dedup, OAI-PMH fix #

Full notes: ~/github/personal/jcrt-author-merge-and-orcid-notes.md

Added scripts/orcid-lookup.mjs (removed from the tree 2026-09-23; in git history at 8757b76d2) — finds ORCID iDs for the 638 authors with no orcid: in front matter. Writes a reviewable report to output/orcid-candidates.yaml; it never edits content unless you pass --apply. Full pass: 105 high, 17 medium, 319 low, 197 none.

node scripts/orcid-lookup.mjs --deep     # regenerate the report (instant when cached)
node scripts/orcid-lookup.mjs --apply    # write rows you marked `confirm: true`

Fixed /oai identifying the journal as a person. repositoryName was "Victor Taylor", a hardcoded fallback in scripts/generate-local-sitemaps.mjs and scripts/lib/oai-pmh.mjs with OAI_REPOSITORY_NAME never set anywhere. Now "Journal for Cultural and Religious Theory". The rest of the endpoint (all six verbs, resumption tokens across 1194 records) was checked and is healthy — /oai with no verb returning badVerb is correct behavior, not a bug. Needs a deploy to reach the live endpoint.

Merged four duplicate author records onto the byline with the middle initial: roger-green + rodger-k-green → roger-k-green (16 items now on one page), jason-alvis → jason-w-alvis, christopher-demuth-rodkey → christopher-d-rodkey. Also: 20 bylines normalized, 4 dead /authors/ links in issue bios repaired, 16 redirects added, affiliation typos fixed (Lindenwood Uniiversity, a stray leading :, Udger Hagedorn, and Richard Kearney who is at Boston College, not Boston University).

Author slugs: _config/authorSlug.js now has a CANONICAL_SLUGS map (the old one-off victor-e-taylor case, generalized). Author pages are keyed by authorSlug() of the article byline, not by the author file — see _data/tagIndex.js:127. If you delete or rename an author file, add its old slug to that map or existing bylines will link to a page that no longer builds.

Deleting any content file: remove its key from _data/standardSiteRecords.yaml too, or npm run standard:check fails and build:netlify aborts.

2026-08-30 (later) — author page/byline reconciliation #

Applied 28 ORCID iDs — every one corroborated by a JCRT publication already on that person's own ORCID record. 77 high and 17 medium rows still await review in output/orcid-candidates.yaml.

Swept every published byline against the author files and closed 17 orphan bylines — articles whose byline linked to an author page that did not exist, which also left the author's own page empty (and _includes/authors.njk hides bio, affiliation and ORCID on an article-less page, so those were invisible too).

• Renamed to canonical slugs: reginald-bell → reginald-bell-jr, carl-a-raschke → carl-raschke, senart-skof → lenart-skof (slug typo).
• Carl A. Raschke normalized — one page at /authors/carl-raschke/ now holding 49 items; all bylines and bio links show "Carl A. Raschke".
• Removed 4 dead author pages: michael-grimshaw, mike-sugimotor, dashan-datar (duplicates of pages that had the articles) and john_dhoe (placeholder).
• Created 7 author files that never existed: CJ Gordon, Jean Leclercq, Joyce Ann Konigsburg, Manuel Mejido Costoya, Manuel Vasquez, N.N. Trakikis, Nicholas Wolterstorff.
• 7 aliases for bylines that belong to an existing page (S.J. Cowan, Ben Stahlberg, Charles Winquist, Hent de Vries, Jonathan Scott Lee, Peter Heltzel, Carl A. Raschke).
• Byline typos fixed: "Stephen Benko" → "Steven A. Benko" (the Buffy article is the Meredith College Benko), "Jean Leclerq" → "Jean Leclercq" (matches leclercq.md).
• 24 redirects added; 8 files relinked off dead /authors/ URLs.

_data/tagIndex.js caches each file's computed author slugs, and CACHE_VERSION must be bumped whenever authorSlug() changes. Otherwise files whose content did not change keep serving their old slugs — and because netlify-plugin-cache persists .cache across deploys, a stale cache survives into production. This silently swallowed the Raschke merge (his page built with 2 articles instead of 49) until the version was bumped to 6.

Needs #

1. netlify integration

• after the domain name is activated impliement a DAM for pdfs and citations
• Find ways to decrease production build times
• add git-lfs CHANGELOG

Questions for Carl and Vic #

• TODO archives 22.1 files dont look right at all, bios.md missing, none of the articles match the live site
• TODO archives17.2 Jean Leclerq How to do things with words (of God)? Michel Henry’s Phenomenology of Religion - original site lists it but has a dead link, new build has no files on it whatsoever, would be sort id 04
• TODO archives 16.1, original lists "Review of Judith Butler’s Senses of the Subject Matt Waggoner", no files in original or new site exist, would be last item on page sort id 10
• TODO archives 3.3 "The City and the Stars: Politics and Alterity in Heidegger, Levinas and Blanchot. By Lars Iyer, University of Newcastle upon Tyne." original article missing entirely would be sort_id: 02

Need Help or Have Project ?? Contact Me #

• adamdjbrett.com
• info@adamdjbrett.com

Version additions #

v0.6 #

• autogenerate citations as *.rs and *.csl.json

v0.5 #

• Archives Post Layout
• Sidebar Nav Fixed

v0.4 #

• Integration with github API religioustheroy repo (Auto Update 1 Day Schemes)

Editorial Theme nicely coded examples #

• ghost
• last update 2 weeks ago jekyll option
• andrew - older but still good css