MoQ wants to unify ingest and delivery—at sub‑second scale
Wowza outlines Media over QUIC (MoQ), an in-development IETF publish/subscribe streaming protocol built on QUIC/WebTransport that targets sub-second live video delivery at large scale, positioning it as a potential single protocol spanning ingest and distribution. The article compares MoQ with RTMP, HLS/DASH (including low-latency variants), and WebRTC, and highlights MoQ’s relay-based fan-out, tracks/groups/objects data model, and browser support via WebTransport/WebCodecs. It also notes MoQ’s current limitations, including evolving drafts, interoperability/tooling gaps, limited hardware encoder support, and the need for hybrid fallback delivery, and states that Wowza has joined the OpenMOQ consortium to help advance implementations and testing.
Key Takeaways
- MoQ (MOQT) is being standardized at the IETF and aims for sub-second live delivery at “CDN scale” via relays and pub/sub subscriptions.
- Architecturally, MoQ targets the gap between WebRTC latency and HLS/DASH scalability—without requiring SFU-style complexity for one-to-many.
- Browser delivery hinges on WebTransport + WebCodecs/MSE, positioning MoQ as a first-class web protocol rather than a proprietary player stack.
- Reality check: drafts are evolving, interoperability/tooling is immature, and hardware encoder support is limited—plan for RTMP/SRT/HLS/WebRTC coexistence.
- Wowza joining OpenMOQ signals vendors are investing early to shape the stack (relays, players, testing) before “winner” status is decided.
Why It Matters
MoQ is the industry’s latest attempt at the “one protocol to rule live” narrative: keep WebRTC-like latency, keep CDN-like scale, and delete the operational tax of protocol handoffs. If it lands, MoQ could shift value from packaging/origin complexity toward relay networks, QUIC-native observability, and track-level control—areas where CDNs, cloud edges, and platform vendors can differentiate. For product teams, the near-term play isn’t a rip-and-replace; it’s piloting MoQ for high-value sub-second use cases (sports, auctions, surveillance) while maintaining HLS/WebRTC fallbacks until the standard and tooling harden.
Read full article at wowza.com