Re: [w3c/FileAPI] Blob URL store partitioning (#153)

Sure, but the first post explaining why you want to do agent clustering isolation uses 3rd party partitioning as its primary and only motivation.  An explainer laying out the additional motivations would be useful for clarifying why we are attempting a bigger hammer than needed for 3rd party partitioning.

> (To explain the scenario above: if A embeds B and B mints a blob URL, you'd want the user to be able to copy that URL and navigate to it. It's not clear to me how that would work if lookup for blob URLs would use the storage key.)

(Just to answer this, but not suggesting we need to do this if agent cluster isolation is preferred for other reasons, but:)

A blob URL has a uuid in it, correct?  That can be used to look up the storage key associated with the blob and apply it to the context created by navigation.  This seems just like how a browser has to use the uuid to lookup the origin of the blob based on the URL today.

-- 
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/FileAPI/issues/153#issuecomment-883445710