- From: Jan-Ivar Bruaroey via GitHub <sysbot+gh@w3.org>
- Date: Fri, 21 Dec 2018 16:04:58 +0000
- To: public-webrtc@w3.org
jan-ivar has just created a new issue for https://github.com/w3c/mediacapture-screen-share:
== Address corner-cases where `getDisplayMedia` prompts only to later fail. ==
Basically, it's a bug in the spec that there are corner cases where calling `getDisplayMedia` will prompt the end user needlessly, only to fail. For instance:
```js
await navigator.mediaDevices.getDisplayMedia({video: {width: {max: 0}}});
```
Fixing this is tricky as it needs to balance 3 things:
1. **Fingerprinting** - mustn't reveal any device info through re-prodding on failure.
2. **Interop** - Different UAs may have different downscaling limits (inherently unspec'ed by constraints)
3. and needless prompting (this issue).
I think we should just leave room in the spec for UAs to optimize away the prompt in these edge-cases, with strict language not to take user configuration into account.
Please view or discuss this issue at https://github.com/w3c/mediacapture-screen-share/issues/96 using your GitHub account
Received on Friday, 21 December 2018 16:05:00 UTC