tegwick
50c68575f1
Assistant: codex Assistant-Model: gpt-6-astra Assistant-Session: 01a0cffe-206f-7323-89d3-b20ca4cf4bf9
1.8 KiB
| title | status | maturity | families | scales | admission | ||||
|---|---|---|---|---|---|---|---|---|---|
| Islands of Interactivity | draft | hypothesis |
|
|
research-candidate-only |
Islands of Interactivity
Context: a document contains a few interactive regions surrounded by readable content. Client execution cost competes with interaction readiness and coordination.
Proposed solution: deliver useful document content, then activate selected regions with separately scoped client behavior and an explicit loading policy. Declare what works before activation and how regions exchange shared state, if needed.
Result: the execution surface can shrink, but delayed activation, shared dependencies and cross-region state introduce complexity. A region is not a security sandbox or proof of fault containment. Loading several runtimes can erase the intended savings. A dense application with tightly coupled interactions may benefit from one client root instead; measure before claiming faster delivery.
Evidence: B04 locates Astro's client-island mechanism. Server islands are a separate mechanism and are not generalized here. No application benchmark or independent full-pattern recurrence is established by this documentation.
Proposed relations: can complement Progressive Enhancement but does not guarantee a no-script task; Failure Boundary requires its own contract.
Next test: fail one region's script download and delay another; verify document reading, pre-activation controls, focus continuity and cross-region consistency.
Evidence locators: breadth review. All relations are research proposals; no canonical identity or admitted graph edges. The next test is planned, not executed. Known uses of the complete proposed contract have not been verified in this investigation.