[paco.org Content Model]

paco.org Content Model

Sprint: EI-3.2 — Taxonomy & Content Model
Date: 2026-08-08

Purpose

This document defines the editorial content model for paco.org after the EI-1 inventory and EI-2 media architecture work. The model is intentionally small. It describes what a piece of content is, what it is about, and how it participates in editorial navigation without reproducing the accumulated WordPress taxonomy.

The governing principle remains: Preserve what is worth remembering. Curate everything else.

Primary content types

Writing

Authored essays, observations, professional reflections, and historical pieces. Writing is the principal editorial corpus. Blog is not a subject and must not be used as a catch-all category.

Briefing

PTIR editions. PTIR is a publication/series, not a subject category and not part of the author’s general Writing stream.

Historical collection item

Material whose identity depends on membership in a preserved collection. The Loudmouth issues are the current example. Collection membership is independent of subject taxonomy.

Portfolio

Curated evidence of historical professional work. The retained journalism/design/video pages should be consolidated into a coherent historical professional portfolio rather than restored as a set of disconnected legacy pages.

Archive record

A historical record retained for chronology/provenance without republication as a first-class article. EI-3.5 will implement the 72 ARCHIVE classifications using this boundary.

Canonical subject taxonomy

The retained corpus supports four durable top-level subjects:

Technology — computing, software, systems, IT practice, programming, the web, and technology culture.
Teaching — instruction, learning, pedagogy, curriculum, and the practice of teaching technology.
Computing History — historical computing, early personal technology, Internet history, period technology culture, and autobiographical encounters with older systems.
Journalism & Media — newspaper work, editorial/design work, multimedia production, publishing, and related professional history.

A fifth miscellaneous or Personal category is deliberately not created. Personal voice is a characteristic of much of the site, not a useful browsing subject. A new top-level subject should be added only when a meaningful body of retained/current material cannot fit the four subjects above.

Collections and publications are not subjects

These values describe editorial relationships and must not be mixed into the subject vocabulary:

The Loudmouth — historical collection.
PTIR — recurring publication/briefing series.
Featured — editorial promotion state.
Archive — discovery/preservation state.
Writing — corpus/landing surface.
History — discovery surface centered on Computing History and historical collections.

Front matter

New or normalized authored posts should use the smallest applicable set of fields:

---
layout: post
title: "Title"
date: YYYY-MM-DD
subject: Technology
featured: false
importance: 0
---

Optional fields:

content_type: writing       # writing | briefing | collection | portfolio
collection: "The Loudmouth"
series: PTIR
edition: Morning
original_date: YYYY-MM-DD
revised_date: YYYY-MM-DD
historical: true

Rules

subject contains one canonical subject. Use a second subject only when it materially improves discovery; do not tag every incidental topic.
collection expresses membership in a preserved historical collection. It does not replace subject.
series expresses recurring publication membership such as PTIR.
featured is a Boolean editorial choice, never a category.
importance controls relative prominence among Featured items. Square-bracket importance indicators remain the presentation convention where importance is displayed.
original_date records the first known publication date when a substantially revised version is published later.
revised_date records substantive modernization. Cosmetic repairs do not require it.
Historical publication dates must be preserved when known. A rewrite must not silently replace historical provenance with the rewrite date.
Existing categories may remain temporarily for Jekyll compatibility during migration, but EI-3 should progressively replace their semantic role with the fields above.
Blog must not be assigned as a category or content type going forward.

Retained-corpus mapping

EI-3.1’s 31 retained items map as follows at the architectural level:

The 16 PUBLISH Loudmouth issues: content_type: collection, collection: The Loudmouth; preserve their historical dates and collection identity.
Technology/history essays such as The Age of Marvels, A Programmer, Again, The Intercom, Twenty-thirteen, Narcopapa (2005), and Random videos from the 1980s: Writing, with final subject selection between Technology and Computing History during EI-3.4.
Why IT support is hated: Writing / Technology.
The retained La Vida, Features Design, Editorial Pages, Express VitalStim, Multimedia, My Work, News Design, and Editorial/Marketing Video material: consolidate as Portfolio / Journalism & Media rather than restoring eight disconnected legacy navigation destinations.

Discovery surfaces

Writing

The primary entry point for authored material. It should expose Featured writing, recent Writing, and subject browsing. It should not present Blog as a peer taxonomy concept.

Featured

A small, metadata-driven editorial selection. Featured remains intentionally scarce.

History

A curated discovery surface for Computing History plus historical collections and relevant professional history. It is not synonymous with the chronological Archive.

PTIR

A dedicated publication landing page and archive. PTIR should not be selected by title-string matching once EI-3.6 normalizes metadata; use series: PTIR/content_type: briefing.

The Loudmouth

A preserved historical collection with its own landing page and issue sequence. Modernization is limited to presentation, navigation, media repair, and contextual notes.

Privacy and production boundary

The EI-1.5 and EI-3.1 production gates remain authoritative. DELETE material is not a hidden content type. It is excluded. The 90 Ben/Sophie records and their associated production-ineligible media must never be migrated, republished, restored, or used as production sources.

Media follows docs/media-manifest.yml and docs/MEDIA_ARCHITECTURE.md: Google Drive is archival/master storage; GitHub contains only production media required by paco.org.

Migration sequence

EI-3.3 normalizes the 16 PUBLISH items without rewriting their period voice.
EI-3.4 applies this model to the 15 REWRITE items and consolidates the historical journalism portfolio.
EI-3.5 implements the archive-record boundary for the 72 ARCHIVE items.
EI-3.6 updates navigation and discovery surfaces to use the resulting metadata consistently.
EI-3.7 validates the entire 228-item disposition ledger against production.