search

immersive-web / capture

Capture composited content, layers and real world through a privacy preserving high level API
https://immersive-web.github.io/capture/
Other
7 stars 4 forks source link

Port explainer from initial proposal #2

Open alcooper91 opened 1 year ago

alcooper91 commented 1 year ago

@cabanier PTAL, this is just porting my initial proposal into a more formal location; I think if we want to spend more time iterating on the shape, etc. we can do that in a future change.

cabanier commented 1 year ago

Thanks for working on this @alcooper91 . I'm still unconvinced if this API is needed. For the use cases you list, it would be more convenient to the user to just use the system's screen recording functionality. Do you need this API because screen recording is too hard to reach on handheld devices?

Looking at the 2D version of the screen capture API, it seems more targeted in sharing your screen with others. Is that a better use case for this proposal?

alcooper91 commented 1 year ago

Hey Rik, My primary motivation for doing this move here was to help address https://github.com/immersive-web/webxr/issues/1200 (and the further discussion in https://github.com/immersive-web/proposals/issues/68) per the discussion at TPAC where I was asked to port this to a more public place.

My initial step here was to just take my old explainer from my personal repo to here as a jumping off point for us to refine the shape of things; though it's not something that we are actively working on at the moment.

Per my read, there is a desire for Web Developers to trigger this system capture from within their experience (e.g. without forcing you to leave it and initiate the system recording) and then be able to do something with it. An acceptable implementation of this API does then become simply triggering that system recording/prompt/ui and optionally providing a mechanism to directly put that recording into the WebShare API.

Looking at the 2D version of the screen capture API, it seems more targeted in sharing your screen with others. Is that a better use case for this proposal?

Is this referring to e.g. getDisplayMedia ? I put a section on why I felt it wasn't a viable path in the explainer; but essentially the intention with the API I'm proposing is to be lower friction and to be extensible for other recording types in the future if we want that.

AdaRoseCannon commented 1 year ago

Thank you!

alcooper91 commented 1 year ago

It's also worth noting that while the System functionality may be ideal; it seems as though (from discussion on the WebXR issue) developers are simply compositing their own capture (and requesting raw camera access where needed to do so for immersive-ar).

AdaRoseCannon commented 1 year ago

/agenda to bring to attention of the group