VR and women: a tale of two problems
May 11, 2017 10:22 AM   Subscribe

The rise of modern virtual reality brought with it a nasty rash of discrimination and harassment issues, both inside the headset and out. Problem #1 is a tale as old as time: a high-profile lawsuit between Magic Leap and former employee Tannen Campbell, who sued the company for sex discrimination after originally being hired to help create a more female-friendly product. The second problem? A growing list of women who have experienced physical/sexual harassment from other users while inside multiplayer VR experiences, where being touched inappropriately can have the same emotional and psychological impacts as if they were harassed in real life. (CW: descriptions of physical/sexual harassment in the links)

A select quote from the sex discrimination lawsuit filing against Magic Leap, a prominent virtual/augmented reality company valued at over $4 billion:

"One woman asked Thompson a question in front of the group and Thompson responded, 'Yeah, women always have trouble with computers.' The women in the group, in apparent disbelief, asked Thompson to repeat what he said and Thompson replied, 'In IT we have a saying; stay away from the Three Os: Orientals, Old People and Ovaries.'"

(There are plenty more where that came from -- read the full document here.)

On May 8th, it was announced that Magic Leap is settling with Campbell out of court for an undisclosed amount.

Aside from the workplace, discrimination and harassment experienced in VR is shockingly similar to how we experience the two in reality. Because it’s now possible to reach high levels of physical "believability" in VR, what happens to your VR body in virtual space can feel like it actually happened to you -- not just the physical sensations, but also the emotional experience. The phenomenon is widely enough accepted that even Elle reported on it in the context of VR harassment in June of last year:

"The impact of negative female stereotypes will be worse in VR," says Jesse Fox, a researcher at Ohio State University whose studies on gender and virtual reality back up her concern. "People will actually be interacting with the stereotypes, so it will affirm gender assumptions one level deeper than traditional media might."

Just a few short months after the Elle article came out, Jordan Belamire published her story of harassment she experienced in VR -- her experience is a perfect example of the emotional consequences of feeling physically present in a social VR experience. “The psychological trauma was about equal,” Jordan said in an interview with the Guardian. “The shock I felt from real life groping versus that first virtual groping was about the same.”

If you ask the VR community what they think, some of the results are depressingly predictable (CW: shitty internet dudes), while others are more... compassionate? While one could say some people are encouraging harassment, albeit aimed toward virtual women instead of actual people, many other developers of VR content are actively working toward better solutions for a harassment-free VR experience.

The creators of the game Belamire was playing when she was harassed (QuiVR) quickly released a statement after news of her harassment spread across social media. They took her experience to heart, expressed a profound sense of regret that it had happened, and then announced they would be adding new design features so users can avoid/prevent harassment.

Since Belamire spoke out, social VR experiences like Rec Room have continued to struggle with harassment issues of their own, despite efforts to design in ways for users to protect themselves from bad actors: muting/blocking, kicking users and reporting their behavior, and personal space bubbles similar to the one Altspace VR implemented last year in an effort to counteract their own community's issues with harassment.

Unfortunately, high-profile reports of harassment continue to come out. Last month, Cy Wise, Studio Director of VR game studio Owlchemy Labs, spoke out about her experience being harassed in VR and how troubling it was. "I found myself grasping for any means to protect myself. Tools were in place but they were rudimentary and ineffective."

Meanwhile, earlier this month Facebook announced and released Facebook Spaces, bringing social VR to a much, much wider audience:

"Booth says that because social VR is restricted to goofy and experimental interactions, it’s easy to exclude more serious expressions like furrowed brows or grimaces. Nobody is getting fired from their job in VR, he says, or delivering bad news to a loved one, at least not yet. So Booth’s theory is that the software can stay restricted to the fun stuff for now."

If Facebook doesn't seem to be concerned about negative interactions, how often is harassment in VR really happening? Data on virtual harassment is practically nonexistent, being an incredibly new issue, but an ethnographic study of harassment in social VR spaces did come out earlier this year with some eye-opening statistics: two out of seven women and 21 out of 99 men self-reported experiencing harassment in VR, with 42% of users saying they had "witnessed someone else being harassed."

Assuming that the Sisyphean task of designing harassment out of VR will never end, it seems we'll be dealing with the legal implications of VR assault and harassment soon enough. But does being harassed in VR actually fall under the legal definition of harassment? Can you be prosecuted for virtually harassing someone? The jury is still out on that one. It may take a lawsuit -- the first of its kind -- to find out whether the courts are willing to prosecute virtual assault, and what kind of culpability can or should be placed on the companies who create the social VR experience where the harassment happens.
posted by Snacks (61 comments total) 49 users marked this as a favorite
 
Good lord, it's for real. Thank you for discovering and posting about this problem, Snacks.

In some ways, it's worse than the IRL stuff because you feel gaslighted and can't figure out if you should be feeling traumatized, abused, and mentally raped or not.

Plus, a couple/three years ago, nobody would have taken you seriously if you spoke out. Talk about feeling silenced by technological advances and prototypes.
posted by infini at 10:32 AM on May 11, 2017 [1 favorite]


VR appears to be another BoyZone that women are best off avoiding.
posted by Carol Anne at 10:34 AM on May 11, 2017 [2 favorites]


I want to cry, from the sheer relief of this long FPP's existence, and all the links. I'll be back later.
posted by infini at 10:34 AM on May 11, 2017 [3 favorites]


VR appears to be another BoyZone that women are best off avoiding.

Not possible if you're part of the original design prototype development.
posted by infini at 10:36 AM on May 11, 2017 [8 favorites]


This is one hell of a first post. Absolutely great work, Snacks. Will dive in and report back.
posted by FirstMateKate at 10:37 AM on May 11, 2017 [16 favorites]


"how often is harassment in VR really happening?"

Well, VR is in its infancy, but if MMO's are any indication of the future, it will prevalent unless effort is done to stop this. Without accountability, such mediums will attract the worst offenders.
posted by el io at 10:38 AM on May 11, 2017 [17 favorites]


Maybe there could be a ranking, you start out as a normal person and if you get downvotes, your gameplay gets restricted. A downvote from 1 player puts the offending player at a remove (space bubble), more downvotes and they get pulled from the game. Next time they want to play, they can't be in a space where they got a downvote. Men. Men, you have a PR problem.
posted by amanda at 10:50 AM on May 11, 2017


This is really an excellent first post- actually this is an excellent post in general. Thank you Snacks.

And I agree with el io that unless this is taken seriously and dealt with at the beginning, it will be an endemic problem. From my perspective, a big part of the problem is that a lot of the techbros behind VR are part of the same community that doesn't see a problem with harrassment in tech in general.
posted by happyroach at 10:54 AM on May 11, 2017 [8 favorites]


I liked the idea near the bottom of the QuiVR Devs Fight Harassment article about sending offending players off with a flick.

When harassment does happen—and I see no way to prevent it entirely so long as multiplayer experiences exists—we need to also offer the tools to re-empower the player as it happens. For example, what if a player had tools on hand to change the outcome of the encounter before it ended in a negative way . . . Would the author’s experience have been any different if she could have reached out with a finger, and with a little flick, sent that player flying off the screen like an ant?

If we can actually implement this kind of creative thinking without getting mired in #notallmen issues, maybe it has a chance.
posted by amanda at 10:55 AM on May 11, 2017 [7 favorites]


Maybe there could be a ranking, you start out as a normal person and if you get downvotes, your gameplay gets restricted.

Let's be real, women would just be harassed and then downvoted once they speak out. In the "growing list of women" link, she outright says "the user base has to change", and I could not agree more. It's a harsh viewpoint that I don't see brought up too much, a lot of people talk about how to implement tools to stop the users from doing what they're doing, or trying to change their behavior through compassion or whatever. But honestly I think its about time we said "we might be better off if we just got rid of the assholes".
posted by FirstMateKate at 11:01 AM on May 11, 2017 [26 favorites]


Wow. The more technology advances, the more the neanderthals need to reassert their place in the system. Reminds me of this early precursor: A Rape in Cyberspace (1993)
posted by Mchelly at 11:23 AM on May 11, 2017 [4 favorites]


As a woman in tech & engineering, with also the experience of different generations, and cultures (having worked in HP back in the mid 1990s in India in new product introductions), I would like to caution against mashing up the two issues as the OP has done in this post.

There is the ongoing issue of harassment of women in tech that is a IRL problem that is researched and documented and also part of a larger problem with women in STEM, academia, and whatnot. For this there are still laws in place, and the HR department, and systems and structures for support.

And, wholly new and emergent, is the issue of virtual assault - as the tag says - which really needs to be looked at as an issue in its own right.

"The impact of negative female stereotypes will be worse in VR," says Jesse Fox, a researcher at Ohio State University whose studies on gender and virtual reality back up her concern. "People will actually be interacting with the stereotypes, so it will affirm gender assumptions one level deeper than traditional media might."

I am a woman of colour, albeit a global nomad in my own head, on the outside I can only be perceived as a stereotype of the Other and a woman. The virtual stereotype is SO ugly that there are layers of abuse taking place simultaneously that make it an experience of it's own. It's the equivalent of being shoved back in time to the colonial era where you were less than nothing.

This experience needs more study, and that apart from what happens in the meatspace.

Further, this OP is only about VR - AR has it's own flavour of abuse and harassment. And if FB and Tesla and Google et al are successful, then the BCIs will allow for the same goddamn bullshit to take place directly inserted into your brain without even teh power to pull the plug.

So, the crap that happens IRL actually becomes much worse once it's boosted into the virtual realm of "experience" and because it's all so new, its raw and harsh and there are no law or structures of support. There is often not even a way to explain to someone - a nurse, say - who doesn't go online much less know what VR, AR, et al are, what you have just been through. Not like the bruises you might have from an IRL assault.

There's "nobody" there to point fingers at, and no way to stop them. Read this, and weep. This goaded him on, and even when I turned away from him, he chased me around, making grabbing and pinching motions near my chest. Emboldened, he even shoved his hand toward my virtual crotch and began rubbing.

There I was, being virtually groped in a snowy fortress with my brother-in-law and husband watching.

As it progressed, my joking comments toward BigBro442 turned angrier, and were peppered with frustrated obscenities. At first, my brother-in-law and husband laughed along with me— all they could see was the flat computer screen version of the groping. Outside the total immersion of the QuiVr world, this must have looked pretty funny, and definitely not real.

posted by infini at 11:38 AM on May 11, 2017 [6 favorites]


Not possible if you're part of the original design prototype development.

Or if you want to take part in the fields where this tech will become increasingly important. Or just if you want to, you know, take part in the fucking future.

This reminds me of an article I read and cannot find at the moment, about how the actual design of VR has sexism built in. Apparently there are two mechanisms the human brain uses to compensate for movement when perceiving visual stimuli. Men rely more on one, women on the other. Guess which one they designed for.

So, women are also more likely to get physically nauseous when trying to use the tech because of the fucking design.

I swear to God, this is one of those days when I want to burn everything to the ground.
posted by schadenfrau at 11:40 AM on May 11, 2017 [18 favorites]


*hands schadenfrau a pretty matchbox from a cool downtown bar & bistro*
posted by infini at 11:45 AM on May 11, 2017 [2 favorites]


What's also chilling is this is what these people do when they know there are no consequences. This is how they behave when they think they can get away with it. We are never going to get closer to an actual social experiment than this, and it turns out men really are fucking terrible.

What do we do with that? I almost just typed out something about how maybe this faux-experimental data will finally convince men to believe women about the kinds of things women are subject to when no one's watching, and then I just laughed bitterly and deleted it.

(Also, infini, I'm sorry. Goddammit. I'm sorry that happened, and I want someone to burn for it.)
posted by schadenfrau at 11:55 AM on May 11, 2017 [14 favorites]


" I almost just typed out something about how maybe this faux-experimental data will finally convince men to believe women about the kinds of things women are subject to when no one's watching, and then I just laughed bitterly and deleted it."

Well, that crossed my mind as well... And any man that is genuinely curious about such things can relatively easily get empathy for women in a VR space by assuming a female avatar. This is true in MMOs as well.

I remember reading 'a rape in cyberspace' when it came out... Essentially virtual sexual harassment came about the moment there were virtual environments.

I think one thing that can help mitigate this is having smaller community based virtual environments; complete with cultural norms. This ended up happening with MUDs/MOOs (for better and worse), and can also happens in communities like metafilter.

Things get harder if large megacorps create huge worlds that are largely poorly moderated (think Facebook - harassment is okay, but breastfeeding is not). Granted, you'll have smaller communities filled with awfulness (think VR stormfront), but you can also have places where moderation and community norms ensure behavior that is free of harassment (think VR metafilter).
posted by el io at 12:23 PM on May 11, 2017 [4 favorites]


I just wanted to jump in here and give some credit to folks who are doing VR in a way that reduces harassment.

I've been enjoying AltspaceVR. It's like second life for VR. They have women on their dev and design teams and have put some thought into the problem and have come up with some nice solutions. For instance when you point at a person to see their name a mute button also pops up, so you can silence people (to your ears) easily. You can also turn on an invisible personal space bubble. If it's on and someone enters or reaches into your bubble that person's avatar will disappear -- to both you and everyone else around -- robbing them of their power in the interaction. They've been careful to keep the bubble around your avatar's face and torso only so that you can still give high fives and shake hands while the bubble is activated.

So there are ways of preventing harassment if game devs are willing to use them. And I get the impression that there (relatively) a lot of women in VR so I'm holding out hope that this particular segment of gaming might turn out to be a nice place for women.

And not all platforms make women sick. I do get sick when I use the Oculus, but the HTC Vive has never made me feel ill. It's designed to track movement using emitters placed at opposite ends of the room and does an amazing job.
posted by antinomia at 12:40 PM on May 11, 2017 [7 favorites]


this tweet from Cy Wise captures it all

The dominant problem was the complete lack of agency. They accosted my body, and I had no comparable physical recourse available to me.

FB will be bad. F8 will be a nightmare.
posted by infini at 12:47 PM on May 11, 2017 [1 favorite]


Schadenfrau, I remember that article, I agree that the consequences are unfortunate for women if true, but how else would you do VR? The vibe approach is a bit better since you're instigating the motion most of the time but otherwise it's pretty much the same idea.

I'm also appalled at the fact that it took such short time for her to experience harassment!! WTF! Still don't know why she didn't log off and try another game with somebody else instead of running away, but the fault is clearly on him not in her.
posted by coust at 12:49 PM on May 11, 2017


Also for those who are sceptical about the weird way what you see in VR connects to your brain... the first time I tried the Vive set of demos, when I got to the tabletop figurines part I had a mini freak-out because your start the demo in the middle of the table (which was waist-level for me), and it's JUST NOT RIGHT, solid objects aren't supposed to intersect in the middle of your body and I was super uncomfortable. I also had a REALLY HARD time walking across the castle wall to see what was inside for the same reason. So I have no difficulty seeing how this felt real and uncomfortable for her.

I feel that as with multiplayer gaming the problem is not VR, but assholes (but those assholes are usually guys). So game developpers will have to put some energy/thought into mitigating those issues as we start to understand them better.
posted by coust at 1:05 PM on May 11, 2017


Still don't know why she didn't log off and try another game with somebody else instead of running away,

Maybe its a horse you can't climb back on immediately. Besides her HUSBAND was playing in that game.
posted by infini at 1:08 PM on May 11, 2017 [1 favorite]


Maybe its a horse you can't climb back on immediately. Besides her HUSBAND was playing in that game.

They were watching on screen, but not playing (I assume not fully understanding how she felt). My thoughts were that this seemed like a pretty cusual game without much at longterm positionning/ranking at stake so not really any consequences if you quit, so why tolerate this any longer than you have to (online idiots never stop especially not if you ask them to). But maybe she's not used to playing online so that sad truth hasn't been made apparent to her yet. Anyway, whatever were her reasons to stay in the game, she's the victim so it's better to start thinking about how to prevent than how she should have reacted.
posted by coust at 1:26 PM on May 11, 2017


I really appreciate the conversation that's being had so far. I've been working in VR for three years now in one way or another, so I have a vested interest in where this is all headed. In the course of putting this post together, it was very difficult for me to find links to sources that I don't know personally.

Anecdotally, practically every single woman I know that works in VR professionally has been harassed at some point in VR, whether it's by a stranger, a member of the press (?!) or a colleague (!!!). Few of them are willing to hang their name and reputation on speaking out.

A running theme from the incidences that I've heard about: the people who don't think it's a big deal -- or who don't seem to experience the sensation of personal space violation -- are oftentimes the ones doing the harassing, blissfully ignorant to what it feels like for those on the receiving end.

Responding to harassment with body language and emotional expression is severely limited due to the rudimentary ways that bodies and faces are portrayed in VR. There are little to no consequences for harassing others, and no way for those being harassed to escape (unless that kind of functionality is built in).
posted by Snacks at 1:31 PM on May 11, 2017 [9 favorites]


I work at Altspace. I'm interested in any ideas people have about how to improve this kind of situation. My personal opinion (based largely on communities like Metafilter) is that high-quality moderation is the most useful tool for shaping the culture, and we're kind of trying to do that one.

I think it can be tough for early VR companies to focus on this stuff, because there are so many things to build that it's fun to work on all kinds of crazy new features 24/7. It's tempting to think that you can just give users tools and let them work the culture out for themselves, because that frees your hands to do other work.

It's nothing new if you read the other things he writes, but I enjoyed this recent talk by Raph Koster at GDC which was addressed to VR community-builders.
posted by value of information at 1:54 PM on May 11, 2017 [4 favorites]


Reader, I read the comments. Don't read the comments.
posted by The Underpants Monster at 2:00 PM on May 11, 2017 [4 favorites]


I like the idea of being able to "flick" the offender away and/or publically humiliate them in front of their buddies - usually there are buddies involved in front of whom this crap is acted out.
posted by infini at 2:03 PM on May 11, 2017 [3 favorites]


I think it can be tough for early VR companies to focus on this stuff, because there are so many things to build that it's fun to work on all kinds of crazy new features 24/7.
I agree and understand, but this same approach - "We'll sort out the icky social stuff later… right now, shiny new tech!" - that makes it incredibly hard to acknowledge or deal with problems once that technology has scaled up. Twitter is a classic example: they've been so focussed on growth that trying to provide decent responsive systems for harassment and abuse have been forever placed on the back burner, and now become incredibly difficult to implement.

As difficult as it is to face, every new technology company must seriously consider all the ways their product could be abused before being introduced to the market. But the vast majority of them don't, because the people building the tech are often young privileged white males who have rarely, if ever, experienced harassment, disability or bigotry.

Yes, this extra work slows down rollout and extends the runway. But due diligence on these issues also means that the technology will ultimately be more inclusive… which means that it's also likely to be more successful.
posted by Bora Horza Gobuchul at 2:21 PM on May 11, 2017 [5 favorites]


What's even more indefensible is that harassment is trivially verifiable in a virtual space. There's no he said / she said; there's a fucking record. Read the complaint, check the flagged interactions, issue a lifetime ban. Which would also work for something like Facebook. So it's fairly obvious that this is just not a priority.
posted by schadenfrau at 2:39 PM on May 11, 2017 [14 favorites]


FFS, why can't personal space bubbles of some sort be on by default? Not something that would make the other avatar disappear, but something that would keep any two avatars from getting closer than a certain distance (which users could change for themselves as an option). By default. The intruding user could feel some generalized-over-that-whole-side-of-body gentle push resistance, and the programming could ensure that the avatar being intruded upon doesn't feel anything. Presumably the system already distinguishes between touches on different regions of the avatar-body, so users should be able to turn off the personal space bubble for certain areas only, eg. enable hand shaking but keep the personal space bubble up on the rest of their avatar's body to avoid butt groping and the like.

It seems to me that this technology has the potential to really promote excellent consent practices, if you have to communicate with someone before your avatars virtually-physically interact. It wouldn't solve the harassment issue, but there is absolutely no technical reason why VR environments can't make virtual assaults near-impossible, as far as I can see.

(I'd say that the fact that they don't is the stupidest thing I've heard all month, but that place of dishonor has been Trumped. Speaking of serial sexual assaulters.)
posted by eviemath at 3:23 PM on May 11, 2017 [8 favorites]


I believe this was the article schadenfrau was mentioning:
Is the Oculus Rift sexist?
posted by Harpocrates at 7:40 PM on May 11, 2017 [1 favorite]


eviemath: "FFS, why can't personal space bubbles of some sort be on by default? "

OMG. This is so blazingly, blisteringly obvious now that you say it out loud that I also can't believe that this isn't already a thing. I mean, why should any rando be allowed by default to have the ability to grope you? Why not encode affirmative consent by default instead? I would love to hear from any VR insiders about why this isn't standard.
posted by mhum at 7:43 PM on May 11, 2017 [7 favorites]


It is actually a pretty complicated thing to set up a personal space bubble for room scale VR. What happens to a user when they move physically in the room but in the VR world, their avatar does not move? Disconnect between movement in your head in VR and your head in real live can cause motion sickness. Without quite elaborate haptics systems, there is not a trivial answer.

I'm not saying that it's not worth considering methods similar to this, but you have to remember that tools designed to prevent harassment might be used for harassment themselves. The right solutions to these problems could be difficult to come up with, and I don't think much effort is being put into finding them.
posted by demiurge at 8:08 PM on May 11, 2017


Since moderation is evidently too much t ask for, I'm thinking of a three-phase technological solution.
1. No anonymity: all users have to be verified by phone or the like.
2. A Slap File. All participants should have the ability to quickly and easily block someone- making it so neither can interact with each other at all. this should be a single button or click ability, and everyone should be told about it when getting an account.
3. Publicly available harassment blocker lists, similar to ggautoblocker , so that people won't have to even interact with potential harassers in the first place. these lists should be available to select as part of the opening of an account.

I mean, that's just a start, and it would probably need a lot of modification. Admittedly, it galls me to have to rely on a tech solution. It also really ignores the fundamental fact that the vast majority of guys behind social media see nothing wrong with harassment, and have designed their sites to facilitate harassment.
posted by happyroach at 8:40 PM on May 11, 2017 [1 favorite]


There's a transitive problem with silencing abusive users: what do you do when the silenced abuser gets third parties to act on their behalf, either without their knowledge or as a vigilante reaction to the abuser's account of events? Silencing can't be the only solution; it's one thing to see people occasionally reacting to an invisible players' actions, but if you silence every random stranger it would break the immersive quality of the world.

It seems to me that at some point you need a moderator who can make the decision to kick some abusive users and tell others to knock it off. Silencing may reduce the need for moderation, but it's not a solution for substantial and repeated abuse.
posted by Joe in Australia at 8:57 PM on May 11, 2017 [2 favorites]


And not all platforms make women sick. I do get sick when I use the Oculus, but the HTC Vive has never made me feel ill. It's designed to track movement using emitters placed at opposite ends of the room and does an amazing job.

You can do this with the Oculus Rift too. The tracking cameras can be set up to allow you to walk freely around your room.

My understanding is that a lot of sickness issues with current-gen VR are caused by how the software handles movement, rather than the hardware. Games that let the player physically walk around their play area, or instantly teleport between places inside the game world, are comfortable for the majority of players. Games that smoothly move the player around inside the game are uncomfortable for some. Games that try to map traditional PC/console game control schemes onto VR are very likely to make people feel sick (I’ve played the original 'Quake' in VR, and it was not good). Though obviously there is a lot more research to be done here.

To their credit, Oculus is actually doing more than a lot of other VR companies to make most of the VR software they fund accessible to a wide range of players. They push for games to be designed with player comfort in mind, even though this has pissed off certain segments of the userbase who want everything to be as hardcore as possible.
posted by fearthehat at 9:49 PM on May 11, 2017


It also really ignores the fundamental fact that the vast majority of guys behind social media see nothing wrong with harassment, and have designed their sites to facilitate harassment.

This is a problem. Especially when developer easter eggs and backdoors are then grey marketed to "horny young males" through the chans.

In the more formal economy, the enabling is evidenced by how Uber handled reports of harassment by their senior executives, but that, imho, is a much later in the stage gate process problem.

The giggling needs to be braked at the earliest testable stages.
posted by infini at 2:50 AM on May 12, 2017 [1 favorite]


What happens to a user when they move physically in the room but in the VR world, their avatar does not move?

VR developers have successfully encoded stuff like not being able to walk through walls, no? Does the wall object also feel or experience resistance when someone walks their avatar into it? When one person is not moving, or is moving away from the other, and the second person is attempting to impose on their personal space bubble, the personal space bubble limit can be treated like a wall for the imposing user, and the stationary or retreating user can be treated as if they were that wall - that is, there's no problem here if the user targetted for assault isn't also trying to move into the personal space of the user attempting assault.

What happens when two users (who are not in a physical room with each other) give each other a high five? Does each person feel resistance when their avatar stops moving, or do the avatars pass through each other like ghosts, or what? If both users are moving toward each other and imposing on each other's personal, they can each get a generalized resistance as if they were walking into a wall when they hit the other's personal space bubble. Or treat it like a normal two-avatar interruption, just extend the definition of each avatar's body out by a pre-set amount, if you want to be sloppy about it.

Getting into the software design details,
this solution requires that avatar movements are coded in relation to a static universal frame of reference, so that you can tell whether a single user is imposing on another or whether two users are imposing on each other. I strongly suspect this is how the systems are coded, since passing data about movement of other users between independent and user-centered frame of reference instances of a multiplayer world would get clunky fast. But if not, consider switching to a universal frame of reference and single instance universe emulator program that all avatars inhabit as agents. Each agent is then an independent subroutine that checks in with the emulator at set (very small) intervals as it moves through the virtual world. Each agent then can keep track of whether it is imposing on any other object (user or inanimate) in the virtual world, and give appropriate feedback. My proposal simply adds a new case to the agent routine that says that if another object in the virtual world is imposing on you unilaterally (you are stationary or moving away), then your agent routine doesn't give you the haptic feedback. And because personal space bubbles would be some nonzero distance from visible surface of avatars if turned on, this would not contradict users' visual experiences. (Bonus: with this feature, also no group of users could gang up on a single user and act as a sci fi trash compactor room to assault the targetted user. The target in the center of the gang would effectively be immobilized, which would be annoying/harassing, but at least wouldn't feel anything physically.)

If your users can walk their avatars through solid objects in the virtual world, you could potentially add to the special case for non-imposing user that the imposing user becomes invisible to them while the personal space of the two avatars intersect (the rule would have to fix the imposing and non-imposing roles based on the start of the interaction, so that non-imposing user doesn't suddenly see and feel imposing user appear in the middle of them if they randomly make a move in the wrong direction). But then some users will get their kicks from the idea of virtually assaulting someone's avatar without that other user's knowledge; and for some of us, just knowing that is going on is squicky and harassing.

A second, both better and simpler option would be to make everyone ghosts to each other by default, unless they both turn that feature off in respect to each other. By this I mean that no haptic feedback is given when avatar objects in your program (the definition of which being extended slightly to include personal space bubble, say an inch out from visible avatar surface) occupy the same virtual world space, and each user's agent can render the portion of the other avatar that intersects your own virtual space invisible to you (so, this would only last while the two avatars occupy the same virtual space). This would not require any special cases to determine imposing versus non-imposing status. And it should be turned off only if both users consent.

Really, this should not be technically challenging beyond the basic challenge of creating a VR system that contains solid objects that don't exist in the physical world and doesn't make people motion sick.
posted by eviemath at 4:52 AM on May 12, 2017 [4 favorites]


I'm also available for hire as a consultant from academia, in either the US or Canada, if any VR companies want help developing these ideas further.
posted by eviemath at 5:02 AM on May 12, 2017 [2 favorites]


Why restrict your expertise by geography, now we're going deeper into virtual worlds? What you say makes perfect sense to me, a layperson, who only has the experiential knowledge of what keeps happening.
posted by infini at 5:04 AM on May 12, 2017


Eviemath you keep talking about feeling "resistance." Which makes me think you might be over-estimating the quality of the haptic feedback available at this time.

I am working on a multiplayer VR experience. We've avoided this sort of problem so far by people using it with people they already know, rather than random strangers. Like, it depends on sending and receiving invitations with people you've accepted as "friends."

I did work on one VR game with random matching, but in that one you were placed on opposite sides of a table and couldn't get into each other's space. We did include a "mute" option so you could turn off other people's voices, and I was one of the people pushing to include that. If it had been entirely up to me, I'd have dropped voice chat entirely and you'd have to communicate with gestures, but I was overruled.

Regarding walls, what I've mostly seen from VR devs is that they design the game so you aren't going near walls, or they ignore it. A fellow from Owlchemy Labs (they made Job Simulator and a Rick and Morty game) says they don't do anything to stop you, and just found people will get bored of sticking their heads through things. Though they focus on single-player games, where you can't ruin the experience for anyone but yourself.

The visual ghosting you mention I find plausible, though I don't look forward to adding a per-person option to the UI.
posted by RobotHero at 6:01 AM on May 12, 2017


VR developers have successfully encoded stuff like not being able to walk through walls, no?

No, not really. With the lack of haptic feedback, this is not a solved problem.

What happens when two users (who are not in a physical room with each other) give each other a high five? Does each person feel resistance when their avatar stops moving, or do the avatars pass through each other like ghosts, or what?

They pass through each other like ghosts. None of the commercial VR systems have any type of haptic devices that would provide any resistance. Technology that would provide such a thing is not even on the horizon.
posted by demiurge at 6:43 AM on May 12, 2017 [1 favorite]


demiurge, what would be the scoping in the context of the now released PR about the various brain computer interfaces the tech giants are allegedly working on?
posted by infini at 7:37 AM on May 12, 2017


Are you talking about Facebook doing stuff with reading brain signals? This isn't anything new and has limited relevance to what we're talking about here. It's basically replacing a hand-held controller with a head-held, less precise one. The "skin-hearing" is just an improved version of a technology that has been around for decades.

Now, when a tech giant is offering a product that does direct neural interface in the style of William Gibson, well, things are going to go completely off the rails, but that's outside the scope of this discussion.
posted by demiurge at 8:22 AM on May 12, 2017 [1 favorite]


That is why I'm scoping this discussion - is it about harassment in every new tech enabled environment? And the necessary structures - legal, social, organizational etc - that can provide support to the abused, once it's outside of IRL / meatspace?

Or is it just about harassment in VR? Which is, in a way, all of it i.e. "virtual" reality, as opposed to brick and mortar

And it's not just FB working on this, everyone else has their own approach & it's almost here else there wouldn't be so much hyping in the media.
posted by infini at 9:04 AM on May 12, 2017 [1 favorite]


The hyping, often times, is sidewise drumming of public support for public funds or favorable regulation. This explains google's wildly inaccurate estimation of the arrival of self driving cars. Original claims a few years back were for this year, no? Note they got out of laying fiber once the telcos showed they were better entrenched.

It would not surprise me at all if the brain implants are being funded by defense money.
posted by Strange_Robinson at 9:38 AM on May 12, 2017 [2 favorites]


Note they got out of laying fiber once the telcos showed they were better entrenched.

Go sit in the corner and think about what you've done.
posted by Etrigan at 9:45 AM on May 12, 2017 [5 favorites]


Did I pun? Newspaper and a stern no works better for this fool.
posted by Strange_Robinson at 9:51 AM on May 12, 2017 [2 favorites]


I don't think there's a difference between harassment online and real life. See this piece. Maybe laws have to be changed to make that fact explicit if prosecutors and judges don't share that belief. However, technological systems to prevent and punish harassment can (and should) still be worked on.
posted by demiurge at 10:26 AM on May 12, 2017 [2 favorites]


So, women are also more likely to get physically nauseous when trying to use the tech because of the fucking design.

I don't believe this is true. Women are more likely than men to get VR sickness for the same reason they are more likely than men to get car sickness. What that reason is hasn't been determined yet, but there is a definite connection in the literature.
posted by demiurge at 10:31 AM on May 12, 2017 [1 favorite]


Well, that crossed my mind as well... And any man that is genuinely curious about such things can relatively easily get empathy for women in a VR space by assuming a female avatar. This is true in MMOs as well.

That was certainly eye-opening for me in WoW, years and years ago.
posted by mordax at 10:33 AM on May 12, 2017


What's even more indefensible is that harassment is trivially verifiable in a virtual space. There's no he said / she said; there's a fucking record. Read the complaint, check the flagged interactions, issue a lifetime ban.

There could be a record. But I don't think they record hand/player position for every frame for every player. I mean it's not a crazy idea, and there are ways to make this not a crazy burden on storage (enable hands log only for players in touching distance of each other for example, keep low frequency position updates otherwise). But I doubt a small game developer put some effort in doing this, too many other things to handle at the same time. Should they record audio too? Once they do that... are we getting worried about privacy?

You could definitely code a virtual hands on virtual body part area detector, and trigger a full logging of the past events (and audio) to help with ban and other consequences. As the tech matures and our understanding of this space improves I can definitely see this as a must have. But this is all very immature/not really thought out/realized for now.

As for preventing getting into somebody else bubble, the solution probably lies in affecting the visible position of the intruding player for the intruded player, this way you avoid motion sickness in case of benign interactions and still keep the intruded player feeling safe.

Nome of this is technically super complicated, but this is a new space, with new kinds of interactions and new kinds of offending behaviors, and it's mostly pursued by small game devs with limited resources (since there's no market for big games in VR yet), so expect it to be a rough ride at the start but it'll get better.
posted by coust at 10:57 AM on May 12, 2017


Meanwhile, drop me a postcard occasionally to see how I'm doing
posted by infini at 11:23 AM on May 12, 2017 [1 favorite]


Also about the article about “motion parallax” and “shape-from-shading", it's a really old research paper. What you could do in 2000 in terms or real time shading VS what you can do now is vastly different (it also ignored stereoscopy which is one of the main cues for some reason), so maybe not so relevant now since we can incoporate a lot of the shape-from-shading cues. VR will probably always have simpler graphics since the frame rate has to be higher to reduce latency so there's less milliseconds available per frame, but still, what we can do in real time is quite impressive and will keep getting more impressive.
posted by coust at 12:24 PM on May 12, 2017 [1 favorite]


The danah boyd paper has a number of issues.

For example, the table on page 16 shows how often people picked a modified cylinder with mismatched depth cues vs a default cylinder: "in the center cell, both sides showed the default, thus either answer is correct". This value ought to be pretty close to 50%, since it's asking people to choose between two identical options, but the actual value in the cell is 27.8%, which is worrisome.

The paper itself is significantly more cautious about its conclusions than the article written about it:

In the threshold part of the experiment, individual variation was apparent, but no statistically significant differences were found based on either individual differences or sex differences. [...] When the cues were put into conflict, males chose shading 32% of the time while females opted for shading 55% of the time. The individual variation on this part of the study was extreme though not consistent.

Note that in Experiment 1 there are only five subjects of each sex (including the author!), so a difference of 20% between the sexes could be caused by a single person.

The second experiment doesn't mention sex differences at all, and the conclusion also doesn't mention any sex differences. The table for experiment 2 does show some small differences, but again, with only ten subjects per sex and huge individual variation, it's hard to draw conclusions.

And, as mentioned above, the experimental setup is not testing stereo vision at all, and the shading model used looks to be plain, untextured Lambertian (the shading model is not actually specified in the paper).

It is entirely plausible that there are sex linked differences in depth perception, but this paper is far from conclusive evidence of that.
posted by Pyry at 1:23 PM on May 12, 2017 [1 favorite]


I am working on a multiplayer VR experience. We've avoided this sort of problem so far by people using it with people they already know, rather than random strangers. Like, it depends on sending and receiving invitations with people you've accepted as "friends."

Sadly, sexual harassment and assault are most often committed by exactly such known acquaintances, friends, family, dates, etc. in real life.
posted by eviemath at 7:51 AM on May 13, 2017 [1 favorite]


RobotHero and demiurge - fair enough! Consider my first suggestion as a future idea to apply when relevant, and the second suggestion as the more germane one at present.

This thread intersects interestingly with two other conversations I've had recently:

1. A different online discussion was recounted to me, where the argument was made that software engineers should not be called engineers (at present, at least), because other engineers have not only more significant professional certification standards, but also specific ethical and legal responsibilities for the output of their work, and can be de-certified (and thus effectively disallowed from working) if their work causes harms as a result of their failure to meet these standards. Whereas no such considerations apply to software engineers.

2. An in-person discussion of the human cognitive biases (that we all have) of not assigning costs or risks to the do-nothing option. It came up regarding vaccination policy, but my friend had a great example to illustrate the problem with not accounting for this bias: feeding your child comes with a certain amount of risk - food allergies, choking hazards, etc. - so why not avoid such risks by not making the active choice to feed your child?

I certainly believe that VR development is a huge and complicated task and that many VR developers may be small and under-resourced relative to the magnitude of the task, but choices are still being made: to develop VR in the first place, what direction to develop it in or applications to develop it for. Not centralizing affirmative consent in VR experiences is a choice. The companies having issues with this in their systems may not have seen it as a choice; they may have viewed it as the do-nothing option (which makes it obvious who has decision-making power in those companies, since that would not be the default/do-nothing choice for me, but as noted in the FPP that's a closely related yet distinct issue). But it's still a choice that has been made, and for which we may wish to hold developers accountable. Though social pressure on metafilter applied to folks on the end of the industry who are already looking into such issues may, admittedly, not be the most effective way to do so.
posted by eviemath at 8:15 AM on May 13, 2017 [1 favorite]


eviemath: "Sadly, sexual harassment and assault are most often committed by exactly such known acquaintances, friends, family, dates, etc. in real life."

I suppose, but it makes me feel like you'd still have social recourse if someone treats you poorly without me building tools for the purpose of recourse.* And to the extent you don't have social recourse, it's at least not a problem introduced by VR.

The ability to combine a feeling of a physical presence and distance is the unique element that's added by VR. So, it can make a visceral experience that people will be tempted to dismiss the importance of because it's "not real." And someone can enter your space without the corresponding physical ability to push them away. Those are the problems we're introducing, so they're the ones we absolutely should address.


* Tools which if they have any punitive element, I then have to consider how can they be exploited by bad actors. Like "sending offending players off with a flick" idea would probably mean the same people doing the groping might start flicking people left and right, to feed their same feeling of power over someone else. Whereas fading out people's hands as they approach can't really be exploited so that appeals to me as a solution.
posted by RobotHero at 9:00 AM on May 13, 2017 [1 favorite]


And to the extent you don't have social recourse, it's at least not a problem introduced by VR.

This sort of thing is what I was trying to link up to the costs or risks of the do-nothing choice, though. Yeah, this is a problem in real life. Given that, choosing to set up a VR system that enables the reproduction of that problem, even though some may view that as the do-nothing option, seems like an irresponsible design decision.

* Tools which if they have any punitive element, I then have to consider how can they be exploited by bad actors. Like "sending offending players off with a flick" idea would probably mean the same people doing the groping might start flicking people left and right, to feed their same feeling of power over someone else. Whereas fading out people's hands as they approach can't really be exploited so that appeals to me as a solution.

Yeah, that's exactly what I was worried about too.
posted by eviemath at 10:44 AM on May 13, 2017


Social VR, that is, group applications, ideally for teams of mixed gender shouldn't raise even more barriers for women and minorities than those which already exist.

MakeVR is a virtual reality tool targeted at opening up 3D, VR CAD design for everyone from professionals to the DIY and Maker community.

Seymourpowell demos VR software for collaboratively designing cars
posted by infini at 11:59 AM on May 13, 2017 [1 favorite]


Another day, another VR company sued for an outrageously hostile workplace. From TechCrunch, "UploadVR sued over ‘rampant’ sexual behavior in the workplace and wrongful termination":
The lawsuit alleges that Upload VR’s co-founders would discuss “how many girls they were going to have sex with” at the company’s parties. A room in the office with a bed was allegedly designated as the “kink room,” where employees would have sex.
And that's just the tip of the iceberg of allegations.
posted by mhum at 3:31 PM on May 16, 2017 [1 favorite]


It reads like a frat room or locker room. How much of this egregiousness has been spurred by the recent transfer of leadership at the very top of the food chain the country in which these offices are located?
posted by infini at 4:42 AM on May 17, 2017


« Older I promise to bring yall the hottest show EVER.   |   Falling Leaves Newer »


This thread has been archived and is closed to new comments