spacialized virtual KVM #56

New Issue

utopiah · 2022-10-12T07:07:13Z

utopiah commented

2022-10-12 07:07:13 +00:00

Generalize triggerbox, see e.g #28 , based on position of other devices, in particular their screen.

Related past explorations

remote keyboard #3
outside of VR https://twitter.com/utopiah/status/1325206376742866944
VR to eink https://twitter.com/utopiah/status/1250464627739279360
screensharing (and potentially controlled even though not binded) https://twitter.com/utopiah/status/1567868019698810882 via noVNC

See also the import/export label for issues https://git.benetou.fr/utopiah/text-code-xr-engine/issues?labels=25

Generalize triggerbox, see e.g https://git.benetou.fr/utopiah/text-code-xr-engine/issues/28 , based on position of other devices, in particular their screen. Related past explorations * remote keyboard https://git.benetou.fr/utopiah/text-code-xr-engine/issues/3 * outside of VR https://twitter.com/utopiah/status/1325206376742866944 * VR to eink https://twitter.com/utopiah/status/1250464627739279360 * screensharing (and potentially controlled even though not binded) https://twitter.com/utopiah/status/1567868019698810882 via noVNC See also the import/export label for issues https://git.benetou.fr/utopiah/text-code-xr-engine/issues?labels=25

utopiah added the import/export core interactions labels 2022-10-12 07:27:51 +00:00

utopiah commented

2022-10-12 09:12:51 +00:00

See also previous work on dragging iframes e.g https://fabien.benetou.fr/Wiki/MetaWhiteboard?action=whiteboard

Note that iframe as a solution is problematic as it does not work in WebXR and also means the state will not be preserved, unless done via URL parameters.

See also previous work on dragging iframes e.g https://fabien.benetou.fr/Wiki/MetaWhiteboard?action=whiteboard Note that iframe as a solution is problematic as it does not work in WebXR and also means the state will not be preserved, unless done via URL parameters.

utopiah referenced this issue

2022-10-12 14:28:14 +00:00

AR in VR mode proper #57

utopiah commented

2022-10-13 09:00:53 +00:00

In this context provenance #20 is particularly important.

For example if one were to drag an item from the screenstack genator then drop it on another display, it could display that very item. Rather what it should do is display what it represents, namely in this context the URL of a wiki page, thus the page itself. If the target device of that display is able to do so it should open the page instead.

In this context provenance https://git.benetou.fr/utopiah/text-code-xr-engine/issues/20 is particularly important. For example if one were to drag an item from the `screenstack` genator then drop it on another display, it could display that very item. Rather what it should do is display what it represents, namely in this context the URL of a wiki page, thus the page itself. If the target device of that display is able to do so it should open the page instead.

utopiah referenced this issue

2022-10-13 12:44:26 +00:00

clarify the federation aspect #47

utopiah referenced this issue

2022-10-13 13:19:10 +00:00

Screenshot as instant perspective #24

utopiah commented

2022-10-13 13:48:59 +00:00

Consider also the Web Share API https://developer.mozilla.org/en-US/docs/Web/API/Web_Share_API and how a mobile is mostly a place of sharing text, including links, and thus how it translates there.

utopiah commented

2022-11-01 14:57:23 +00:00

Consider extending glTF for that.

It would allow to have the 3D representation of an object but also possibly a text description, HTTP interface, SWAGGER description of that API, etc.

This could faciliate discovery then composability.

See also current implementation of manipulation of containers, e.g https://twitter.com/utopiah/status/1533064468292419585 to faciliate the usage of externals resources (e.g CPU, RAM, GPU, disk, etc).

See also past IoT explorations via WebThings.

Note that from https://fabien.benetou.fr/Content/OnManagingInterfaces one would stick to HTTP and only fallback to other levels if need be, e.g WebDAV over sshfs.

Consider extending glTF for that. It would allow to have the 3D representation of an object but also possibly a text description, HTTP interface, SWAGGER description of that API, etc. This could faciliate discovery then composability. See also current implementation of manipulation of containers, e.g https://twitter.com/utopiah/status/1533064468292419585 to faciliate the usage of externals resources (e.g CPU, RAM, GPU, disk, etc). See also past IoT explorations via WebThings. Note that from https://fabien.benetou.fr/Content/OnManagingInterfaces one would stick to HTTP and only fallback to other levels if need be, e.g `WebDAV` over `sshfs`.

utopiah commented

2022-11-01 15:00:04 +00:00

See also Designing for serendipity: supporting end-user configuration of ubiquitous computing environments https://dl.acm.org/doi/abs/10.1145/778712.778736 on the value of going beyond "just" KVM.

Own perspective on interfaces, including https://twitter.com/utopiah/status/1568912887334174727 , Future of Text presentation, subsequent https://fabien.benetou.fr/Content/OnManagingInterfaces

See also Designing for serendipity: supporting end-user configuration of ubiquitous computing environments https://dl.acm.org/doi/abs/10.1145/778712.778736 on the value of going beyond "just" KVM. Own perspective on interfaces, including https://twitter.com/utopiah/status/1568912887334174727 , Future of Text presentation, subsequent https://fabien.benetou.fr/Content/OnManagingInterfaces

utopiah commented

2022-12-05 07:16:05 +00:00

Offline version based on history, e.g ~/.bashrc shortcuts on last reMarkable pages or NAS photos or browsing history (cf tweet), could already be very useful.