Hacker News Reader

Comment by radiorental | Hacker News Reader

We've been researching solutions in this space for quite a while.

We have 6x 80" smartboards and have tried everything from Lync to Realtimeboard.

While some solutions work better than others, no one has solved a fundamental problem... When you try to collaborate in real time with a remote person - all the human communication cues are lost.

Collaboration is a dynamic and fluid process. We subconsciously read each other to gauge the reaction to a proposal, we interject and riff off of others. We try to pull introverts into the conversation and tame the extroverts.

All of this is practically impossible when a colleague is elsewhere on the network.

View on HN

Two anecdotes; not sure if they help at all, but they're related and I like sharing :)

1. I work in a loud, open plan office where I wind up taking many phone discussions with 5+ participants. Since our teleconferencing software doesn't have a global push to talk, we all have to manually toggle our mic mutes, which has become an ad-hoc indicator of a desire to speak (along with the background noise that is a cue that 'someone' is thinking).

2. There's this let's play series (whose name eludes me at the moment) that doesn't have a face cam. The actors, then, are forced to constantly narrate their actions: they hmm and haw more often than normal, fill empty airspace with idle, easily uninterruptible chatter, etc. In other words, it seems like their discourse adjusts to the limitations of the medium.

Anyhow, I hope you can solve that problem before AR/VR takes over the entire space! :D

I make a VR app for this use case: http://hyper-room.io

Website doesn't show it now but there's whiteboards as well:

http://i.imgur.com/TzXrRPc.jpg

Currently the only nonverbal cues I can really get across are hand/head movements(with cube avatars) but as far as achieving presence goes I find that to go a long way. Even without facial expressions or complicated hand gestures having that "physical" presence seems to make each session feel really concrete, and less like you're talking to someone through the ether. To make up for the loss of facial expressions I've found myself and others becoming more gestural when using the app, which may not be ideal, but the fact that you can adapt at all(and that it feels natural) sets it apart from 2D interfaces.

That's really neat! The facial cues can be done today in real-time to a surprisingly high-quality degree (I do virtual filmmaking as a hobby), and with cheap mocap hardware, you could get the full upper body and what not. I'd expect when combined with the HTC Vive that you'd get a pretty decent VR experience.

Is there any solution to this other than in-depth VR?

Teleportation? Synthetic local proxy android?

I'm not trying to be sarcastic. This is definitely a problem faced by any company with remote teams, including mine. I'm not saying any of the solutions I posited are feasible, but one day, far in the future, I'm hopeful that some kind of solution will be developed. If anything, because there is a huge market in facilitating communication between remote teams.

There's a few emerging attempts. From all that I've experienced though, nobody has completely 'solved' remote collaboration.

For some tech nerdery: Oblong Mezzanine (http://www.oblong.com/), Prysm Visual Workplace (http://www.prysm.com/prysm-visual-workplace), MultiTaction MT-Canvus (https://www.multitaction.com/products/mt-canvus side note: check out their touch tech, it's f'ing cool), DisplayNote Montage (https://www.displaynote.com/wireless-presentation-system/), as well as MS Surface hub, all the SMART products, and the multitude of other 'meeting room IWB / touch solutions' out there.

All of these centre around the concept of a large display surface, some digital whiteboarding, flexible content layout then some VC 'tacked on' so that remote paricipants can join in.

What I find waaaay more interesting is some more of the combination acoustic + CV based camera tracking that's starting to make it's way to market. In addition to tracking people, it can also track objects of interest — things like auto-framing a whiteboard when someone approaches it. Cisco is currently doing this on their collaboration endpoints with SpeakerTrack and PresterTrack and other vendors have similar tech too. There's also a solutions that are starting to use camera arrays or super wide angle sensors to capture a larger scene and extract areas of interest (https://www.huddly.com/, http://www.owllabs.com/).

Have you looked at The Surface Hub?

Other than tighter integration with skype for business, how does it solve some of the issues I laid out above?

I don't think our IT team got these specific devices in to test. The 20K for the 80" is a dealbreaker compared to the smartboards. Curious if there's something special about them compared to a smartboard on a win10 box?

The Surface Hub is the successor to Perceptive Pixel [0] which has some very smart people working on what makes multitouch work. Hardware-wise it's a capacitive screen vs. almost all other big touchable screens that use other technologies that are inferior.

It will do nothing to address your concerns about the absence of body language, gaze direction, facial expressions, etc. It's a trade-off, you give up some of the bandwidth of meeting face-to-face for the benefit of including people not physically present (or expense of bringing remote people local). The technology used to bring in remote people can also add value by capturing/saving work more automatically and seamlessly than, say, someone stepping back and snapping a photo of a whiteboard.

Some of the bandwidth loss can be compensated for by examining your processes, identifying what you value, and formalizing the valued communication in another form. Often the other form is saying things out loud; instead of looking for head nods, you ask people to say "yes" or "no." The technology can aid in this by providing other channels like up and down votes or "question" votes in place of confused looks (that are all to easy to overlook when you're in person anyway).

[0] https://en.wikipedia.org/wiki/Perceptive_Pixel

Reply on news.ycombinator.com