Building a Real World, End User, Open Source Application in Clojure
summary
An open source document archive called Pepa document management system stores PDFs and blobs in Postgres, uses Postgres full‑text and advisory locks as a durable queue, and runs a processing pipeline: page extractor → renderer → OCR implemented as componentized background workers with threads and core.async for notifications. The web UI is ClojureScript + Om with a single app state, long‑polling push and Transit, routing kept in state, config as Clojure files, and the project is usable but still work in progress under AGPL.
AI-written summary · about
chapters (17)
- 0:00Intro
- 3:18Why open source
- 4:57Project overview
- 7:23Component library
- 9:22Postgres
- 10:53Schema
- 17:17Processing
- 20:49Logging
- 22:20Logging to it
- 24:33HTTP
- 25:28Frontend
- 26:02Data Model
- 27:14Documents
- 28:12Routing
- 30:12Processing Status
- 31:41Other Features
- 34:36GitHub
related talks
38mWhat we built with Open Source
27mWhy Your RAG is Failing , And How Open Source Can Fix It
40mGetting 50,000 Companies on Board with Clojure
33mOpen Sourcing: Real Talk
36mDocumenting Digital Infrastructure: Explaining Complex Open Source Tech for Everyone
36mArkisto: an open-source, standards-based framework for digital preservation
05mMoot
22mSupabase - powered by open source