- From: boblund via GitHub <sysbot+gh@w3.org>
- Date: Thu, 14 Jun 2018 21:40:41 +0000
- To: public-web-and-tv@w3.org
If you use the VttCue text attribute, DataCue data attribute or some other generic mechanism, then there would need to be a specification to denote how the user agent would encode the attribute contents so that JavaScript could differentiate an emsg contained in the attribute from some other data type. This specification could be done in [1]. For what it's worth, there was some resistance when writing the HTML5 recommendation to a generic DataCue used in this way. The argument was that Cues should have an explicit type to avoid misuse, either by error or deliberate. An EmsgEventCue with uri, presentationTime, duration, id and emsgData attributes would be a more explicit way to expose emsgs that addresses this concern. The user agent work should be roughly the same in all cases since the UA still needs to detect the emsg and format it for whatever API. [1] https://dev.w3.org/html5/html-sourcing-inband-tracks/ -- GitHub Notification of comment by boblund Please view or discuss this issue at https://github.com/w3c/me-media-timed-events/issues/2#issuecomment-397447176 using your GitHub account
Received on Thursday, 14 June 2018 21:40:47 UTC