A sourced local knowledge base for Escondido.
Escondidopedia gathers public information about Escondido and keeps source links close by.
What this is
A careful local notebook for places, organizations, city government, parks, history, culture, and public decisions.
Guarded automation helps with routine research and checks, while sensitive or uncertain details wait for review.
How the site works
-
1
It looks for public sources.
The local AI worker watches public pages, documents, videos, and records that may help explain Escondido.
-
2
It saves what it found.
A source snapshot is kept so a future reader can see what the page was based on, even if the original website changes.
-
3
It pulls out possible facts.
The AI turns source material into small notes, such as a date, address, vote, public role, or source-stated description.
-
4
It checks the notes against the source.
A note should match the source that supports it. If it is unclear, stale, private, sensitive, or too broad, it waits for review.
-
5
It builds simple pages.
Published pages explain the topic in plain language and keep links to the sources close by.
-
6
It checks again later.
When a public source changes, the site can mark a page as needing another look instead of pretending old details are fresh.
Visit Stewardship to watch recent site work, scheduled checks, and publication guardrails.
Technical information
Escondidopedia is a static site generated from reviewed project files. The public pages are built from article text, source notes, extracted text, Spanish translation notes, and validation reports.
- Runtime host
- Routine research automation runs on a desktop computer maintained for the project. The public site can still be served by ordinary static hosting.
- Site build
- Python scripts turn Markdown, YAML frontmatter, JSON records, and reviewed Spanish translations into static HTML, CSS, JavaScript, indexes, feeds, and sitemaps.
- Source trail
- Saved public sources, extracted text, source details, review notes, and file checks are kept alongside the pages so published statements can be audited later.
- AI workflow
- The local automation computer helps find sources, compare changes, draft possible notes, prepare review notes, and propose page updates. It does not turn uncertain or sensitive material into public facts by itself.
- Review gates
- Validation checks sources, review blockers, image and location policies, school-page labels, translation freshness, links, and public-site metadata before changes are treated as ready.
- Outside services
- Hosting, maps, browser previews, and explicitly requested remote model calls may use outside services. The default research loop is designed to keep the project files and routine AI work local-first.
- Public analytics
- When configured, the public site uses Vercel Web Analytics to count aggregate page views and site performance. It is separate from suggestions, review queues, public comments, and user accounts.
What keeps it trustworthy
Local information can affect real people and places. The site preserves source trails and slows down when details could be harmful or wrong.
-
Sources still matter.
The AI is a helper, not the authority. Pages should point back to public records, official pages, archived source snapshots, or other cited sources.
-
Not everything gets published.
Private details, sensitive locations, weak claims, rumors, and unclear facts should stay out or wait for human review.
-
The site can be corrected.
Because sources, review notes, and change history are kept, future maintainers can audit and fix mistakes.