Re: [w3c/ServiceWorker] TPAC 2026 Status Report (Issue #1845)

yoshisatoyanagisawa left a comment (w3c/ServiceWorker#1845)

Thanks @LJWatson for the reminder!

Thanks @monica-ch for reviewing the report before the post!

Here is a summary of the ServiceWorker specification's progress over the last 12 months, blockers, testing status, and next steps.

### 1. Progress - What has your spec achieved in the last 12 months?

In the past year, our primary focus has been on improving "Spec Health" rather than introducing new features. We have spent significant effort on continuous issue triaging and extensive editorial maintenance to make the specification cleaner and more robust.

#### Key Algorithmic & Normative Fixes:

* **Request Soft Update**: Added support for requesting soft updates (#1847) to allow cross-spec integration (e.g., for FedCM).
* **Header & Request Handling**: Redesigned the Service-Worker-Allowed header extraction (#1830), and ensured NavigationPreloadManager.setHeaderValue rejects invalid header values (#1827).  
* **Module Scripts**: Clarified that module service worker scripts are fetched with same-origin credentials (#1829).  
* **Event Origins**: Updated ExtendableMessageEvent to work properly with origins (#1821).  
* **SW static routing API**: Fixed the Lookup Race Response algorithm (e.g., #1811, #1807).  

#### Editorial & Infrastructure:

* **Task-Queuing Steps**: Clarified and fixed numerous task-queuing steps across various algorithms (e.g., Client/Clients API, NavigationPreloadManager, ServiceWorkerGlobalScope.skipWaiting) to resolve ambiguities (#1834, #1835, #1836, #1839, #1844).  
* **Cache Request Handling**: Removed invalid request initiator and destination assignments from Cache.addAll(), aligning the algorithm with Fetch (#1822).  
* **API Consistency & Model Alignment**: Aligned ancestor-origin references with HTML's Document model (#1812), corrected documented messaging event interfaces on ServiceWorkerContainer and ServiceWorkerGlobalScope (#1825, #1814), and clarified FetchEvent client-ID initialization, including null handling and default values (#1832).  
* **Publishing & Infrastructure**: Set up automated publishing for the spec (#1804) and added an implementation report (#1805).

### 2. Blockers - What blockers, if any, are preventing you from progressing your spec to CR or Req?

* There are no critical blockers preventing general progress, as the specification is already quite mature.  
* Our main ongoing work is maintaining interoperability and resolving edge-case behaviors across different browser implementations.

### 3. Testing - Is there a testing plan for your spec?

* We continue to rely heavily on Web Platform Tests (WPT). Tests are actively added or updated whenever edge cases are resolved or algorithmic behaviors are clarified.

### 4. Next Steps - What are the next steps for your spec and do you need any help?

* **Continuous Maintenance**: We will continue our regular issue triage process to keep the backlog manageable and ensure the specification remains healthy.  
* **Address External Integration Requests**: Implement necessary spec changes to export SW primitives for JIT registration, as requested by FedCM and Payment Handler (#1846).  
* **Resolve Parallel execution issues**: Continue to address missing tasks in parallel steps and refine dedicated parallel queues for APIs (e.g., #1740, #1840, #1841).  
* **TPAC 2026 Discussions**: We plan to discuss the open issues labeled with TPAC2026 to reach consensus on remaining architectural edge cases. Specifically, we would like to go over the following issues:  
  * #1500
  * #1313 
  * #1310 
  * #1254
  * #1201  
  * #1141  
  * #777

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/ServiceWorker/issues/1845#issuecomment-5902714890
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/ServiceWorker/issues/1845/5902714890@github.com>