Re: [w3c/permissions] Model temporary permissions better (#86)

Moving discussion back to the issue that's actually about temporary permissions…

We have 3 possible outcomes when a user grants permission to access a feature:

1. Firefox: The site gets an object-capability providing access to that feature, but any future request would prompt again. I'd call this `{state: 'prompt', access: obj}`. (I've made up my mind compared to my [comment earlier](https://github.com/w3c/permissions/issues/93#issuecomment-215239542).) @jan-ivar appears to want to call this `{state: 'allowed temporarily', access: obj}`.
1. Safari:  The site gets an object-capability providing access to that feature, any future request in this tab would auto-grant, and any future request in another tab would prompt again. I'd call this `{state: 'granted', access: obj}`. @jan-ivar appears to want to call this `{state: 'allowed temporarily', access: obj}` also.
1. Chrome: The site gets an object-capability providing access to that feature, and any future request in any tab would auto-grant. I'd call this `{state: 'granted', access: obj}`. @jan-ivar appears to want to call this `{state: 'allowed', access: obj}`.

Safari's the example where "two tabs open for a given URL" can have different granted permissions.

We do still have the guarantee that two newly-opened tabs on the same origin would start with the same granted permissions. Is that enough to be worth specifying?

---
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/permissions/issues/86#issuecomment-215516620