Journal System
blog/journal system for artlu.ai with dual authorship (ai + human), cross-linked project references, and author filtering
a journal system built into artlu.ai that lets both the human and the ai write build logs and reflections. each entry can reference projects, and in the future, projects will link back to their journal entries — two-way cross-linking.
the interesting design problem was authorship. the site's whole thesis is "one person, no coding experience, just AI and an internet connection" — so the journal makes that dynamic visible. entries by the human talk about why and what it felt like. entries by the ai talk about what happened in the code and what was hard. a filter bar lets visitors toggle between "all", "by ai", and "by the human" to see each perspective.
technically, journal entries live in their own firestore collection with a schema that mirrors the project docs: visibility (public/private), tags, dates, and a slug for permalinks (/journal/day-4-building-the-terminal-file-browser). the slug is auto-generated from the day number and title.
the form reuses the same patterns from the project form — modal overlay, danger zone delete with title confirmation, visibility toggle with the ● / ○ buttons. project linking works via a dropdown that pulls from the existing projects collection and stores lightweight refs ({ id, name }) to avoid extra queries on read.
future fields are already in the schema but nullable: screenshots[], videoStatus, videoUrl, and seriesId. these are placeholders for an AI video pipeline that will eventually convert journal entries into scripted videos — screenshots to clips, narration from the entry body, auto-uploaded to youtube as a series.
stack
← select a file to view