Re: [w3c/ServiceWorker] A way to immediately unregister a service worker (#614)

Unclaiming just means that the service worker no longer handles events for that client, correct?

If that is true, then I'm not sure we need a non-unclaiming CSD implementation. I would expect that CSD would unclaim. If we were to have multiple CSD directives for service workers, I would want delineation between stopping new events being handled by the service worker, vs going a step further and killing any in-flight events being handled by the service worker. In general, I would prefer to have the most extreme measure be the default MVP for a CSD service worker directive.

Stepping back, I see CSD as a safety mechanism for a serious problem, where collateral damage to the UX is OK. In these situations, an abrupt loss of any functionality provided by the service worker is going to preferable to a bad service worker continuing to operate unabated. Developers can always implement their own `unregister()` based mechanisms for less serious situations.

> Also, @mattto, didn't you say Chrome already 'unclaims' if things become corrupt? It seems reasonable to spec that somehow.
@jakearchibald How are we defining "corrupt"?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/ServiceWorker/issues/614#issuecomment-503252166