Re: [webidl] Add a [SecureContext] operator attribute (#65)

> now, I popup a new window when the user clicks the play button (or selects a movie) so that I work around the pop-up blocker, I manipulate focus to that the window popup is behind the main window (if it wasn't opened in a new tab), and I carry out my secure operation there.

If it's a short operation that doesn't require user input, then popping up a window, doing your powerful thing, and then closing window would be effective. Netflix in particular couldn't have done that because they required the WebCrypto API persistently in order to decode their magical DRM.

> Otherwise, I can put it in the background

Chrome's done a lot of work to make popunders difficult to impossible to create. I admit that I'm not up on the state of other browsers here, but aren't popunders something we should kill? :)

> Why then is it OK to have this loophole?

The thing I'm worried about is navigation. I see popups being used fairly often to move users from one page to another, and it seems strange to me to tie the new page's security state to the page from which it spawned. To make this at all effective, we'd more or less have to taint the whole window somehow, as a cooperative window could otherwise just navigate itself around until we considered it secure.

I guess I'd be willing to do that if Microsoft objects to/would have problems implementing the current definition. Perhaps you could start a thread in WebAppSec? Or extend the thread @rlbmoz started, I suppose: https://lists.w3.org/Archives/Public/public-webappsec/2015Oct/0073.html.

---
Reply to this email directly or view it on GitHub:
https://github.com/heycam/webidl/pull/65#issuecomment-150227443