- From: Mathieu Hofman via GitHub <sysbot+gh@w3.org>
- Date: Fri, 22 Jan 2016 23:15:17 +0000
- To: public-media-capture-logs@w3.org
You're correct that this would still allow browsers to prompt for permissions on every confirmed change from the user, and that the user might reject the prompt. However you already have to deal with the device access potentially failing for other reasons. Furthermore, I don't fully agree with your assessment that this would result in an interruption to an ongoing call. The application can request the new devices and only switch away from the currently active device once the new device have been accessed successfully. If for some reason you don't get full access to the requested devices (rejected or other failure), you can notify the user and keep the call ongoing with the previously acquired devices. The exception might be on mobile devices where some platforms are not capable of having 2 webcams active at the same time. However I would expect such platforms to instead expose one camera with a `facingMode` constraint that can be switched between `user` and `environment` using `applyContraints()` on the existing stream. This use case would then not require a new gUM call to switch between front and back camera. -- GitHub Notification of comment by mhofman Please view or discuss this issue at https://github.com/w3c/mediacapture-main/issues/304#issuecomment-174083693 using your GitHub account
Received on Friday, 22 January 2016 23:15:20 UTC