- From: guidou via GitHub <sysbot+gh@w3.org>
- Date: Mon, 07 Aug 2017 19:00:40 +0000
- To: public-media-capture-logs@w3.org
> I think the question to ask is: What can developers not express with this? A constraint would allow using ideal to express a preference that is not a hard requirement. For example, an application might prefer native modes, but be willing to accept non-native ones. This could be easily expressed with constraints, but probably not with the keyword approach, unless you make another exception. Constraints would give users fine control over multiple processing techniques. For example, frameRate adjustment, resolution downscaling and even new effects that users have requested such as video mirroring would be easily controlled separately. It is not clear to me that the keyword approach allows controlling frame-rate adjustment separately from rescaling, which an application might want to do. Native camera modes are normally triplets (width, height, frameRate), so if a camera has 640x480x60Hz as native mode, and the user requests max 30Hz, then it should throw OverconstrainedError since 640x480x30Hz is not a native mode, right? > Remember the current spec says nothing about this, and I'm only advocating for a SHOULD. I still prefer UAs in charge of this decision, so they can handle concurrent access well. > A new constraint would undermine that, and do nothing to solve the implementation discrepancy developers are facing today, because constraints don't have defaults. What would new processing constraints exactly undermine? IMO, they would do a lot to solve implementation discrepancy in a straightforward and explicit way, for those who care about the discrepancy. Users who don't care about the discrepancy are fine without using the constraints, even if browsers keep their current behavior as default. I suspect these are the majority since no one has filed bugs in Chrome about ideal not preferring native modes (for comparison, there were lots of popular bugs asking for support for ideal before Chrome 59). Do you have evidence that this discrepancy (in which all browsers correctly implement the spec) is negatively affecting users? -- GitHub Notification of comment by guidou Please view or discuss this issue at https://github.com/w3c/mediacapture-main/issues/472#issuecomment-320751249 using your GitHub account
Received on Monday, 7 August 2017 19:00:47 UTC