IdeaWell

Project ideas from Hacker News discussions.

Show HN: Honker – Postgres NOTIFY/LISTEN Semantics for SQLite

Original Article

Hacker News Discussion

📝 Discussion Summary (Click to expand)

1. SQLite‑native push notifications via WAL stat – Honker adds NOTIFY/LISTEN‑style events by watching the WAL file with stat(2), delivering single‑digit‑millisecond latency without a daemon and keeping everything inside the same SQLite file.

"Hey HN, I built this. Honker adds cross‑process NOTIFY/LISTEN to SQLite. You get push‑style event delivery with single‑digit millisecond latency without a damon/broker, using your existing SQLite file." — russellthehippo

2. Performance of frequent stat polling – Community members note that stat(2) is far cheaper than feared (sub‑microsecond per call) and that syscalls, while slower than pure userspace functions, are still affordable for this use case.

"Small correction on ambiguous wording - syscalls do not evict all your stuff from CPU caches." — vlovich123

3. Alignment with existing queue/workflow patterns and atomic commits – The project mirrors familiar Postgres‑style primitives (pub/sub, durable queues, offsets) and the key selling point is that notifications and business data share the same transaction, avoiding the “sent‑but‑rolled‑back” problem of external IPC.

"atomic commit with the business data is the selling point over separate IPC." — tuo-lei

🚀 Project Ideas

Generating project ideas…

SQLite Event Stream Hub

Summary

  • Turn any SQLite database into a real‑time event bus with pub/sub, durable queues, and per‑consumer offsets, removing the need for separate brokers.
  • Provides a managed API/UI that handles WAL‑based notifications, replication across regions, and offset persistence to survive checkpoints.

Details

Key Value
Target Audience Developers of lightweight, single‑writer apps who want reliable async notifications without running their own message broker.
Core Feature Event streams triggered by SQLite WAL changes, delivered as ordered, replayable messages with semantics similar to LISTEN/NOTIFY but language‑agnostic.
Tech Stack Backend: Go + SQLite + NATS JetStream; Frontend: React; Deploy: Kubernetes; Persistence: SQLite (embed) + PostgreSQL for metadata.
Difficulty Medium
Monetization Revenue-ready: Tiered subscription (Free 10k events/mo, Pro $19/mo, Enterprise custom)

Notes

  • Directly answers HN comments like “how to use stat(2) for notifications” and “need durable work queues without external IPC”.
  • Mitigates WAL checkpoint loss by storing consumer offsets in a separate SQLite metadata table that survives file truncation.
  • Enables integration with Litestream read‑only replicas, allowing event streaming from remote replicas for multi‑region use cases.

Honker‑Lite SDK

Summary

  • A lightweight, cross‑language library that exposes honker‑style WAL‑polling primitives as a reusable SDK for pub/sub, work queues, and event streams.
  • Lets polyglot microservices add durable, atomic notifications to any SQLite‑backed app without building custom brokers.

Details

Key Value
Target Audience Engineers using Python, JavaScript/TypeScript, Ruby, or other languages that only have process‑level concurrency and need reliable event delivery from SQLite.
Core Feature API to register listeners on a SQLite file, receive fired events with payloads, and manage queues/retries entirely within the same file.
Tech Stack Language‑specific wrappers (Python/JS/Ruby) built on top of SQLite C API; uses stat() on the WAL file; optional Go core for performance‑critical path.
Difficulty Low
Monetization Hobby

Notes

  • Aligns with HN feedback that “syscalls are cheap enough” and that “cross‑language notification is the missing piece”.
  • Provides a minimalistic wrapper that handles atomic commit of business data and queue entries, eliminating the “notification sent but transaction rolled back” problem.
  • Can be packaged as a zero‑config npm/PyPI/RubyGem, encouraging adoption in small services and hobby projects.

SQLite Workflow Bridge

Summary

  • A self‑hosted engine that packages durable queues, dead‑letter handling, and event streams into a single SQLite file with atomic transactional guarantees.
  • Designed for developers who want a Postgres‑like workflow engine but prefer SQLite’s simplicity and zero‑ops deployment.

Details

Key Value
Target Audience Teams building internal tools, schedulers, or micro‑batch processors that rely on reliable task queues and need strict ACID guarantees.
Core Feature Integrated queue tables that atomically commit task enqueue/dequeue with business data; supports per‑consumer offsets and retry policies.
Tech Stack Rust core for high‑performance WAL monitoring; SQLite for storage; optional Docker image for easy deployment; client libraries in Go and JavaScript.
Difficulty High
Monetization Revenue-ready: Revenue‑share license for hosted instances, or “Pay‑what‑you‑want” for support contracts.

Notes

  • Directly addresses HN concerns about “atomic commit with the business data” and “handling WAL checkpoint loss”.
  • Includes automatic checkpoint‑aware offset tracking to prevent event loss when SQLite truncates the WAL. - Can be packaged as a standalone CLI tool and library, offering both open‑source core and commercial support options for enterprise use.

Read Later