Offline-First and Local-First Are Back
Two Show HN projects this week tackled sync and offline-first from scratch. Syncular is a TypeScript/Rust offline-first SQL sync layer built on CRDTs, and its creator noted this is the third time they have attempted this problem after earlier efforts in 2019 failed to reach production quality. The Bor project, a Linux desktop policy manager, also generated discussion around local state management and configuration drift, a related problem of keeping local state consistent without constant network round-trips.
The pattern here is that local-first is graduating from a niche philosophy into something builders are actually shipping. The Syncular thread had a commenter compare it to Lotus Notes in a positive way, which would have been an insult five years ago. The appetite for apps that work offline and sync intelligently is real, driven partly by reliability concerns and partly by the cost of always-on cloud architectures.
The main counterpoint in the threads was complexity: CRDTs and conflict resolution are genuinely hard, and several commenters pushed on edge cases around policy enforcement timing and drift detection. The problem space is well-understood but the engineering is still painful.
So what?
If you are building a mobile or desktop app and treating offline support as a stretch goal, you are leaving a real user experience gap. The tooling for offline-first is finally maturing, with Syncular and similar projects offering CRDT-backed sync without building it from scratch. Founders in productivity, field software, or any app used in low-connectivity environments should treat this as a foundation decision, not a later optimization.