Chatto is now open source

Original Article ↗ Hacker News Discussion ↗

Positioning & Goals

  • Framed as an alternative to Slack/Teams/Mattermost, with a Discord-like feel.
  • Aims to be pleasant to use, self-hostable, and not “open-core” crippled; all features available when self-hosted.
  • Focuses on single-tenant, per-community servers rather than large federated mega-instances.

UX, Performance & Design

  • Multiple testers praise it as very fast and snappy compared to Slack, Matrix clients, and some others.
  • Design and UX are seen as unusually polished for an open-source chat app.
  • Some criticize it as visually similar to Slack/Discord and question whether it solves noise/complexity problems.

Architecture & Deployment

  • Backend in Go, frontend in Svelte; shipped as a compact self-contained binary.
  • Uses NATS with JetStream for messaging and persistence; some devs share positive experiences with NATS at scale.
  • Uses LiveKit for voice/video.
  • Self-hosting is reported as straightforward; examples for S3-compatible storage and other components are provided.
  • Resource usage: tens of MB for a fresh instance; ~10MB RAM per additional connected user reported, with room for optimization.

Licensing & Business Model

  • Backend is AGPL; frontend is Apache 2.0.
  • Rationale: AGPL discourages proprietary hosted clones while keeping UI customisable and enterprise-friendly.
  • Commercial “Chatto Cloud” hosting is planned, but self-hosting is unrestricted.

Clients: PWA, Desktop, Mobile

  • First-class PWA experience is the current focus, including push notifications and voice/video on mobile.
  • No official native apps yet; community Tauri wrapper exists, seen as desktop-ready but not mobile-ready.
  • iOS push via PWA/Web Push for now; native apps with APNS/FCM are planned but not imminent.

Security, Privacy & Encryption

  • Chats are encrypted at rest; core messaging is not end-to-end encrypted.
  • Some see lack of E2EE as acceptable for self-hosted, controlled environments; others want E2EE for group chat.
  • Per-user keys and account deletion semantics raise questions about legal retention and soft delete for business use.

Interoperability & Network Effects

  • No federation or Matrix/XMPP compatibility; deliberate choice to keep system simple and business-friendly.
  • Some argue this harms adoption due to strong Slack/Discord network effects; others say private communities don’t need federation.
  • Lightweight identity federation across Chatto servers is being explored.

AI-Assisted Development Debate

  • Project reportedly used “agentic coding”/AI assistance.
  • Some are impressed by solo dev productivity; others strongly object on ethical and environmental grounds and say they’ll avoid the project.
  • Counterarguments note open-weight models, historical IP reuse, and personal productivity gains; disagreement remains unresolved.

Comparisons & Alternatives

  • Frequently compared to Mattermost, Matrix, Zulip, Rocket.Chat, Fluxer, and Discord.
  • Advantages mentioned: simpler setup than Mattermost, more polished and faster than many Matrix clients, clear licensing, included SSO without enterprise upsell.
  • Disadvantages/unclear: no E2EE for chats; no migration/interop beyond a Slack-import tool; mobile-native gap; no clear feature comparison tables.

Adoption Concerns & Wishlist

  • Requests for: native mobile/desktop apps, Slack migration and ongoing sync, public read-only rooms, import/export, drop-in voice rooms, auto-updates, resource sizing guidance, and better new-server onboarding.
  • Some users are ready to move off Slack/Mattermost once native clients and smoother onboarding are available.