- From: Mark Foltz via GitHub <sysbot+gh@w3.org>
- Date: Fri, 23 Oct 2015 20:41:12 +0000
- To: public-secondscreen@w3.org
mfoltzgoogle has just created a new issue for https://github.com/w3c/presentation-api: == Add a "starting" state to PresentationConnection == When calling PresentationRequest.start(), it may take some time for the presentation to be fully loaded and ready for connection on the presentation screen. In the meantime, the controlling page is unaware of the initialization progress of the presentation, since we only resolve the Promise with a connected PresentationConnection. Activities (i.e., media playback) will continue in the meantime, while the user has already indicated they want the content to be presented remotely, which creates a jarring user experience. If the controlling page were aware that presentation was starting, it could pause the media and update the UI to show the user what was happening. @avayvod has proposed adding a starting state to the PresentationConnection, with a new start algorithm that would roughly go as follows: 1. If the user selects a screen, then resolve the Promise right away with a "starting" PresentationConnection. 2. If the user cancels presentation, reject the Promise. 3. Once the receiving context is fully loaded, transition the PresentationConnection to connected. 4. If there is an error loading the receiving context then transition to disconnected (or perhaps a new error state). Messaging would be allowed during the "starting" state in case the presentation wants to convey metadata about the chosen screen, or show some progress information. See https://github.com/w3c/presentation-api/issues/214
Received on Friday, 23 October 2015 20:41:14 UTC