Re: [w3c/ServiceWorker] Should a service worker be allowed to register new service workers? (#1117)

yoshisatoyanagisawa left a comment (w3c/ServiceWorker#1117)

Thanks for the detailed update, @asutherland. It's very interesting to see how Firefox managed to mitigate the nesting/abuse concerns using UA-internal lifetime constraints.

Some contributors on the Chromium side tried to expose `navigator.serviceWorker` in dedicated workers last year (crbug.com/40364838), but we found it extremely tedious and complex to implement correctly due to underlying threading and context-binding issues in Chromium. As a result, that effort has stalled, and we don't have active plans to ship it in the near future.

As you can see, there were discussions about potentially restricting `register()` or removing it from the spec due to these complexities and the potential for abuse. However, Firefox has shown a way to mitigate the risk without such restrictions. Now, it could be a choice to allow `register()` as-is while expecting UAs to implement proper lifetime restrictions.

Anyway, let me leave this open for further discussion.



-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/ServiceWorker/issues/1117#issuecomment-4889482910
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/ServiceWorker/issues/1117/4889482910@github.com>