Gamepad multitouch API

[bradleyneedham]

Request for position on an emerging web specification

Information about the spec

• Spec Title: GamepadTouch Interface
• Spec URL: https://knyg.github.io/gamepad/extensions.html#gamepadtouch-interface
• GitHub repository: https://github.com/w3c/gamepad/
• Issue Tracker (if not the repository's issue tracker): https://chromestatus.com/feature/4782975812108288
• Explainer (if not README.md in the repository): https://github.com/bradleyneedham/gamepad/blob/multitouch-explainer/multitouch_explainer.md

Design reviews and vendor positions

• TAG Design Review: Gamepad Extensions API touch input w3ctag/design-reviews#799 (comment)
• Mozilla standards-positions issue: Request for Mozilla Position: Gamepad multitouch API mozilla/standards-positions#724

Bugs tracking this feature

• Chromium Bugs: https://bugs.chromium.org/p/chromium/issues/detail?id=1143942

Brief summary of the issue or the subject of the spec.
Addition of new web interfaces that provide web developers with access to the touchpad inputs from modern Gamepads.

[marcoscaceres]

@bradleyneedham, I don't think we are opposed to this, but we should really move the out of the extension spec into the spec proper (and use that to gauge implementer interest and commitment). I see there has been some movement already on the Chrome bug - what's the status there?

From a spec perspective, it's missing all the details about how/when the events are fired. The proposal also needs some wording around ids to make they remain unique per origin to prevent fingerprinting.

There probably needs to be some explanation about the overlap with touch events too. I see Rick Byers argued in the TAG issue that they are "different", but it would be great to go into the details as to why they are different.

[bradleyneedham]

@marcoscaceres thank you for the comment. The windows implementation on chrome is almost done, I am currently trying to get it committed.

Touch events like other Gamepad data are pulled rather than fired to a callback at the application layer. I will see if I can make this clearer in the spec as well as clarify the uniqueness per origin of the ids.