Re: [spec-reviews] Presentation API Request for Feedback (#61)

Review went well today. The Second Screen WG had some followup from the discussion that I would like to bring back to the TAG:
* The SSWG would like to reference the TAG's [Private Mode](http://w3ctag.github.io/private-mode/) 
spec under the expectation that Private Mode would define how to start up a browsing context in a "clean state" (e.g., empty database, local storage, cookies, etc.). If the document would be able to define that (and be ready in a reasonable time), then it may be reasonable for the Second Screen spec to reference it. Otherwise (and perhaps more likely?), the Second Screen spec would need to be explicit in what states it expects the various features it cares about to be in. Either way, we need to reply back with our suggestion after discussing this.
* We identified a somewhat interesting issue that is manifest when you start launching browsing contexts on screens where there is either no facility for processing user input, or no user nearby (i.e., a second screen connection to a "smart" projector). In these cases, some parts of the web platform are not workable when no user is available. ```window.alert()``` is one such example--a blocking API that requires a user to dismiss the UI. Another example is the ```input.type=file``` when it is ```click()```-ed. (Other things that require user interaction or user consent are generally designed for async behavior and can tolerate not being responded to.) We felt this was an opportunity for collaboration between the Web Platform WG and the SSWG. As the TAG, let's make this happen.


---
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/spec-reviews/issues/61#issuecomment-152439245