5 ms·
What you’re posting here is “common knowledge”, but it is actually in conflict with social presence research. VR beats video conferencing even without any facia
by gfodor 5y ago
What you’re posting here is “common knowledge”, but it is actually in conflict with social presence research. VR beats video conferencing even without any facial tracking, and the abject failure of video conferencing in the pandemic to replace most of the feeling of face to face communication should be enough to convince people there is merit to the studies that have been done showing it scores poorly.
In VR you multitask by bringing the computing environment into the virtual environment. So there is no deficit there. I think you haven’t tried these things before so you are just speculating.
- teucris 5y agoInteresting. Links to research?
- gfodor 5y agoHere is a talk I gave a few years ago - I haven’t kept up tho so there may be better sources now. https://m.youtube.com/watch?v=_5w8xbeCc2Q https://m.youtube.com/watch?v=_5w8xbeCc2Q
- sweetdreamerit 5y agoThe literature I know defines social presence as something that has only partially to do with technology (and Virtual Reality) [0]. The advantage of VR is that it can enhance the sense of immersivity in some circumstances [1]. [0] https://www.tandfonline.com/doi/abs/10.1080/01587919.2017.1324727 https://www.tandfonline.com/doi/abs/10.1080/01587919.2017.13... [1] https://scholarspace.manoa.hawaii.edu/handle/10125/50061 https://scholarspace.manoa.hawaii.edu/handle/10125/50061
- jonnycomputer 5y agoAnd the actual benefit of the feeling of face to face communication is?
- gfodor 5y agoWe spend a sizable % of US energy moving physical bodies around, so apparently people do in fact like face to face communication.
- johnny53169 5y ago> In VR you multitask by bringing the computing environment into the virtual environment. So there is no deficit there. Writing/typing would be really bad in VR though, and that's a major need for meetings, I don't see a good solution for that.
- gfodor 5y agoThe solution is you use a keyboard and mouse. No need to reinvent the wheel. It can be brought in via passthrough camera tracking.
- dathinab 5y agoThe ting is the "felling together" part is less dependent on the medium (real workd, audio, video, VR) but on the way the meeting is held. Video conferencing didn't fail because video conferencing is fundamentally bad but because: - meetings are often done in a fundamental bad way, where people don't have any "felling together" even is they sit physically in the same room. - people having bad microphones, the video part in a video conference is the least important the audio part is what makes it work or fail (because meetings are all about speaking, except if you are mute/deaf, in which case VR might help). - technical difficulties all over the place, there is no reason for this to get better with VR - peoples homes being fundamental unsuited for conferences (i.e. a lot of background noise), again nothing which will change with VR - also VR headsets and mimic don't work well together, proper gestic is possible but often requires full body tracking which put more requirements to the environment you use it in. - VR headsets are much more straining to use then audio-only conferences, and somewhat more straining then video conferences. Which can be a major no-go for anyone doing many conferences, like the management deciding weather or not to buy into it. - Companies want virtual meeting rooms not a meta-verse, something which already exists, and looks better then the honestly crappy looking thinks Facebook presented.
- Tagbert 5y agoIf people use microphones with noise cancellation, it can remove most of the background noise. When I was still in the office I used a Jabra headset with a noise suppressing microphone, I could speak on a call with people talking nearby and their conversation was not noticeable on the call. I would hope that a VR/AR headset for this purpose would have a similar microphone setup.
- gfodor 5y agoI wrote “feeling together in unmediated communication” on purpose. The unmediated part is the most important part of the definition.
- somewhereoutth 5y agoI'd be curious to know how VR beats video conferencing - by what metrics etc. Video conferencing has certainly been wildly successful in actually getting (collaborative) work done - though in fact it is screen sharing that is the secret ingredient here. Would screen sharing in VR be any different? I guess 3D models or images could be shared usefully. It is notable the difference between tele conferencing and video conferencing - and that difference is faces (and screensharing). Until VR can show faces, it is basically a tele conference with a puppet show. My point about switching is that in VR you are completely immersed, so are limited to what the VR provides, whereas with a vid conference I can e.g. go make some coffee while still participating. I'm limited mainly by social convention, not the technology itself. The nature of modern work is we are at the centre of various tools and technologies (half of which are a bit broken) that we choreograph together to get useful stuff done. No VR system could hope to replicate it all. Actually I'm currently working on a VR project (amongst other things), so these concerns are real and pertinent to me. (As ever with engineering there is a tendency to focus on the what and the how, instead of the who and the why. Furthermore, people will use technology as they see fit, which might be quite different from how it's inventors intended or envisioned. To develop on this point, even the most 'non-technical' human is an accomplished tool user, and they won't be reading the manual anymore than they absolutely have to.)
- mbesto 5y ago> VR beats video conferencing even without any facial tracking In what terms does it "beat it"? I still have Zoom meetings where people just dial-in, have crappy video, etc. In other words, video conferencing is still being adopted decades after it was introduced. > In VR you multitask by bringing the computing environment into the virtual environment. Multi-tasking isn't a real thing. The overhead from context switching is real and no amount of technology will improve this human deficiency.