<trevorfsmith> delete agendum 1

<trevorfsmith> Scribe: Leonard

Meeting started at 12:05

scribe: by Trevor

Trevor: 5 agenda items

#1: from Klaus

https://github.com/immersive-web/dom-overlays/pull/7

Spatial favicon next steps (Ravi)

Klaus's pull request for DOM overlay spec

Klaus: Looking for feedback. Near - handheld AR (phones)
... Next: migration path to headsets
... demoed at last F2F with Daydream.

<trevorfsmith> Rendered version: https://klausw.github.io/dom-overlays/index.html

Klaus: not intended to replace a more general layers approach where the application has full control over all content
... but may use same technology underneath.

Trevor: Assuming there is not full readership

<Rafael> q

Klaus: General idea is there is a new feature. Had used DOM overlays for capabilities
... developer provides root element of overlay in display. Application could check whether it worked
... if mode (screen or floating).
... How to deal with duplicate events. Should a screen touch generate a DOM event or a 3D event?
... proposal to introduce new event type <missed it -- beforexrselect?>.
... put an event listener on the DOM overlay and that determines whether to handle or pass through event.
... easier than handling DOM events and convert to XR events (or vice-versa)
... How to handle cross-origin events.
... Prevent information that is not passed through on regular 2D events.
... Consider this (cross-origin) somewhat optional

Rafael: Question about security. Prohibit cross-origin (XO) pose info

Klaus: Yes (XO pose info is static when outside of origin). May not apply to XO images
... only really applies to interactive iframes

Rafael: Concerned about how that would work. UX when "tracker" is XO

Klaus: The situation would be similar to 2D. Want to avoid situation where Origin and XO both want info

Rafael: How to specify non-interactive iframe?

Klaus: No way to currently do that.

Rafael: a full-screen transparent might be able to track everything

Klaus: Getting into areas that are more part of UA security and protection of user

Rafel & Klaus: Not run past security yet.

Rafel: Must specify at beginning of XR session -- seems strange

Klaus: With Android necssary to set up early. Don't want to force penality if not used.

Rafel: Is there performance penality for having it and not using it?

Klaus: Yes - composition

Rafel: Would be good to turn it on/off as needed

Klaus: Not sure of use case

Rafel: Initial orientation, then not. May need to turn on something for a short time later on.

Klaus: Not core use case. Sounds more like dynamic DOM layers

<someone -- didn't hear>

Klaus: Difficult to add (remove) feature at run-time

Rafel: Issue with HTML or Chrome

Klaus: Chrome. Not using DOM overlay, then compositing is a lot simplier. A lot of technical details of implementation

R&K: 3D element still visible behind DOM overlay

Trevor: Terminate resource used by overlay. Could be added to API later on
... Interaction bewteen RAF for DOM and RAF for 3D.

Klaus: Not addressed in document. Window RAF would continue (60fps). XR runs at its own rate
... not necessarily synchronized

Trevor: Developers would need to figure out how to manage

Klaus: Not strong guarantee

Trevor: May need addition wording for that

<trevorfsmith> Scribe: trevorfsmith

Leonard: Cross-origin events are important to control for security. Use cases like social media sites tracking details of a third-party XR session are problematic. So a vote for security.

<Leonard> Scribe: Leonard

<trevorfsmith> https://github.com/immersive-web/dom-overlays/issues/6

Next agenda: Feature Name

https://github.com/immersive-web/dom-overlays/issues/6#issuecomment-555692915

Dom-overlays feature name (Klaus)

Klaus: What should the name of this feature (see #1) be? The issue is headset that may not have a (an obvious) DOM.
... Are there any issues

-- Nothing reported --

<trevorfsmith> I think the 'dom-overlay' name is fine.

Klaus: Use DOM Overlay as prefix to items that may be shared to avoid confusion

Trevor: Email public list for opinions

Agenda #3: Input Profiles

https://github.com/immersive-web/webxr-input-profiles/issues/105

Manish: Push to WG. Not all needed people are on this call (Nell, Alex)

Leonard update on favicons

<trevorfsmith> Scribe: trevorfsmith

Leonard: We discussed how to spec size/quality of a spatial favicon. Leonard talked within Khronos and a working group has picked it up and is working it.
... Magic Leap is a member of Khronos so they're also in that loop.
... The 3D commerce working group in Khronos picked it up. They're focused on 3D in retail spaces.
... There's an XMP extension for metadata for glTF so fields in there might specify quality information that can be queried by browsers.

Piotr: There's an initial draft of hit-test spec sent out an hour ago.

<Leonard> Bialpio: Quick update on hit test. Initial draft of spec just distributed. Interested in feedback -- github or direct

<Leonard> ... Will issue a PR to main spec for some algorithms to have extenstion/hooks to cover extra capabilities

<Leonard> Trevor: No other items. Meeting ended.

This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/EMP/XMP/
Present: Trevor F. Smith Leonard mounir Albyhox alcooper cwilso kip
Found Scribe: Leonard
Inferring ScribeNick: Leonard
Found Scribe: trevorfsmith
Inferring ScribeNick: trevorfsmith
Found Scribe: Leonard
Inferring ScribeNick: Leonard
Found Scribe: trevorfsmith
Inferring ScribeNick: trevorfsmith
Scribes: Leonard, trevorfsmith
ScribeNicks: Leonard, trevorfsmith

WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)