The biggest issue for VR is similar to that of Second Life: outside of gaming, it needs to seamlessly augment reality.
If the pervasive influence of social media in our times has taught us anything, it is that a platform that does interface with our everyday existence will never be more than a cool tech demo.
Off the top of my head, for success VR needs to provide:
* Virtual Conferences
* Remote Home Sale Walk-throughs
* Credit for remote class attendence.
* Remotely Shared Sports Viewing with Friends
Until the VR experience can supplant some of the activities above, I doubt it will take off.
I hear "Virtual Conferences" as a big VR application. But how about we solve videoconferences first? Skype/Hangouts etc. hasn't improved noticeably in the last ten years. Videocalls still suffer from latency and poor image quality.
really? I frequently video conference to places will 11 hour time differences, and I am constantly amazed how well it works. I am pretty sure 10 years ago this wasn't practical.
It definitely wasn't. I worked at a web conferencing company in 2007 and my sense was that video conferencing was limited to big entities with deep pockets and dedicated setups. Now anybody can do it and it works pretty well. It would be nice to get from "pretty well" to "so close to perfect that everyone forgets this was ever a hard thing", and it does seem like progress has been a bit slow on that front, but it seems like we're at the point where it will continuously improve constantly but almost imperceptibly.
I remember skyping my girlfriend over shitty satellite internet from an island off Brazil in 2009, on an Atom-powered netbook running Linux. It worked (barely), but on regular internet it was not worse than the low end quality of today.
I did a Hangout a few weeks ago from Europe to a startup in TX. At least on my end I was on a 1Gbps fat academic ISP; can't speak for the other guy, but he was in his office. Still, voice and video quality was poor and we had to restart everything once.
Also have had many poor quality video conferences on Lync/Skype for Business where all participants have been on the same fat academic ISP.
Yeah it happens. I still have both Skype and hangouts and swap depending on who is having a bad day. Still, it used to be much worse, but it's nowhere near perfect yet
I recently listened to an interesting discussion[1] about VR on a friend's recommendation. One of the guests had a great test for quality VR: can it perform as a hiking simulator? Can it mimic all the qualities of a good hike -- visuals, body movement, sounds?
It seems to me we're much closer to a great AR experience than truly great and immersive VR.
>> Can it mimic all the qualities of a good hike -- visuals, body movement, sounds?
I've got some VR hiking goggles to sell you. They are cheap, lightweight and totally wireless, but only work in the woods. They were featured on a recent SouthPark episode.
Too many discussions focus on VR being a drop-in replacement for reality. Nobody should expect them to mimic reality any more than they expect computer screens to mimic sunlight. I don't much care if VR helms don't allow me to jog up Everest. I just want to be free to look around as if I were standing on the top.
I think many people who are physically unable to take a hike (disabled, elderly, ill, etc.) would appreciate a VR system that can closely mimic a real hike.
Also, if a VR system can perform well as a hiking simulator it would be able to realistically simulate things that don't exist, places that are too difficult/expensive for an average person to visit, or experiences that are too dangerous to attempt in reality.
EDIT: For example, I'd love to visit the moon or Mars but it's highly unlikely I'll ever be able to physically go there. A highly immersive VR simulation would be very appealing to me!
But if they are disabled, then they in fact want the VR system not to emulate a hike. They cannot walk and so do not want a VR system that makes them walk. The OP was discussing a VR scheme that would mimic body movements akin to the physicality of actually hiking (ie VR on a treadmill).
Agreed. While I do think there are some interesting applications for building perfect reconstructions of reality, I think the most interesting experiences won't be in that direction.
Just because something is in 3D doesn't make it automatically better -- often times it's worse as you now have artificial constraints that you didn't have before. For example, it's far easier to scroll through Instagram with your fingers (a UI win) and have every photo at 100% scale than to glide through a 3d representation where much more work has to be done. Certainly you wouldn't want the world to scroll you in VR -- you'd throw up.
This was true before with SL, and it'll be true with VR in the future. It'll be great for certain kinds of games, therapy, housing walk throughs, and other contexts for which a synthetic 3D world is useful, but it won't be the panacea that replaces all other forms of UI. Immersion also has a cost -- it's very intense and you don't want it for hours and hours and hours. Just like you'd never want to be on a rollercoaster for 8h, VR for 8h will also be too much (unless the world itself is not an intense game, you don't have a screen 2" from your eyeballs, etc).
Correct. However on the otherhand, walking around a 3D art gallery where each wall is dedicated to one of the accounts you are following would be a benecicial use of the technology. You could easily scan the small framed images for those you like, and click on them to visualise them at full size in front of you.
Have you tried this? I've done a couple of 3d museums in the oculus & cardboard and it's a terrible experience. Scanning small images is not an improved experience over existing feeds on mobile phones, especially for something where you end up only looking at a photo you like for 0.4-2 seconds anyway.
If the content is 2d, I imagine a 2d interface is the right way to view it. Real paintings aren't 2d, though - Even modern fully-computerized paintings are actually 3d planar constructs, because artists use many layers in their construction, though that's often not exposed.
Real paintings have brushstrokes, bumps, texture and depth. Even the most high-res scans of paintings utterly fail to capture that. The Met has a fantastic collection of art scans that you can browse online[1]. Some of those works are stunning, even in thumbnails. But they still pale in comparison to how they are when you're up close, when you can see the grain of the canvas, the brushstrokes and dabs of paint, the way that it all pieces together. I always view a painting from multiple angles, looking from the left and right, tilting my head. Texture matters, a lot. VR can provide that in a way a 2d (or even flat 3d) monitor never could.
The real promise is an endless, densely packed 3d museum, where you can single out an exhibit, open it in it's own room, and enlarge/move it as you see fit. You walk down the aisle, viewing many pieces in an infinite array, and zoom in when you like one. You get many constantly changing angles, which is what our brains were built to deal with.
It's not there yet. Tilt Brush gives us a taste. More will come.
I feel like it's missing a lot of the point of museums. It's like asking why go to Japan when you can just virtual walk around Tokyo? Experience is far more than just tricking yourself visually, and it's more than just being able to look at brush strokes up close in 3D.
Instead of reproducing existing things poorly that you seldom want to do anyway, VR should be about doing things that are either impossible in real life or are compelling enough even if done in a non-realistic form.
I agree with the key point here: VR will not replace all other forms of UI.
It is a mistake to look at VR UIs as strictly 3D. There is no reason you could not scroll through your photos on a 2D screen with your fingers in VR. You can even have the added benefit of spawning multiple giant 2D screens, then grabbing photos off them for a scrap book or a virtual art gallery.
I've shared a great deal of locomotion experiences with other Vive owners. VR sickness seems a lot rarer than it was hyped up to be, and definitely a lot easier to acclimate to. I wonder how many vomiting journalists would be fine if they eased into it over a day or two. I can fly around in Windlands without flinching these days. Don't get me wrong, nausea is still a real design constraint, but we might be a lot more adaptive to VR and weird locomotion.
On the topic of session lengths, I firmly believe this is just an issue with content and ergonomics. I've done a few 6 hours sessions in multiplayer games, the problem is barely any exist. I haven't heard anyone experience "immersion fatigue", but some games will cause actual physical fatigue. Thankfully I already stand all day and it isn't an issue. I think if you already can play a game for 8 hours, VR will be there to provide the same in the future.
I think this is the underlying cause for many of the failed projects from the 'Second Life bubble'. Many human activities, especially those related to information intake and processing have a long history of being 2D. From clay tablets to paper books, from the first photograph to 4K television, virtually all of the information designed for human consumption has been 2D. Only an incredibly small percentage is in 3D and even less is designed to be consumed in 3D (e.g. video games are often 3D internally but are designed for 2D consumption). This might or might not change in the future but if it does it will be changing neither soon nor rapidly.
Yes, one can use 2D interfaces inside 3D worlds but any 3D system short of the matrix will have a noticeable quality degradation (e.g. visual fidelity, haptic feedback, fatigue, etc) compared to directly using a 2D interface in real life. A few edge cases might become feasible much sooner, e.g. using VR to emulate large screens but that's hardly 'Virtual Reality'. Until we come up with meaningful ways to represent more kinds of information in 3D and/or create a matrix-style experience, VR will most likely be limited to applications that are inherently 3D.
This obviously includes entertainment (VR games, 'movies', 3D art) and CAD/CAM/3D modelling which are already very much 3D in nature and internal representation, even if they are currently usually projected onto 2D screens. Outside of that, the main 'educational' usage currently consists of building 3D environments (labs, buildings, factories, complex machinery) that emulate real-life environments in order to familiarize trainees with the workflows in those places. This can be surprisingly effective, especially considering how much cheaper VR mockups are compared to e.g. NASA's real-life 1:1 ISS training replica.
These applications are worthwhile in their own right and they will probably be able to sustain a decent market for VR hardware and development. However, I'm personally much more interested in seeing what kind of concepts that aren't already inherently 3D can be enhanced or revolutionized by adding another dimension.
Do these planned top-down approaches ever work? Especially in education? It seems like when technocrats have questionable ideas they try to shove it down the throat of kids instead of having a more decentralized approach with more choice. Negroponte's failure with the OLPC project comes to mind. A mix of cheap netbooks, tablets, and phones made it quickly irrelevant.
If we see social VR in education it'll be impromptu study groups in Altspace or screensharing in Bigscreen. It'll happen dynamically as kids decide what platform they prefer. Teachers will probably make similiar decisions on a per teacher/classroom way. I've already spent a bit of time in Altspace and Bigscreen and its very impressive from many aspects like usability, discoverability, presence, features, etc. I can't imagine using a SL-like system. Its too messy, ugly, and complex. It feels very Web 1.0.
Also, my god, the uncanny valley in SL is inexcusable. Either do photorealism or use cartoony avatars, there's really no room for middle ground here.
> Negroponte's failure with the OLPC project comes to mind. A mix of cheap netbooks, tablets, and phones made it quickly irrelevant.
To be fair to OLPC, it just predated those other devices. It [the hardware] was developed precisely because cheap laptops didn't yet exist, and then the form and market came into existence and obsoleted a large part of the project's efforts (the physical hardware development, not its educational goals).
Not really, but the time they were actually shipping the olpc in non-trivial deployments, the netbook revolution was already here. Intel also had the education aimed Classmate netbook as well. 2007 article on the two competing platforms:
>nd then the form and market came into existence and obsoleted a large part of the project's efforts
It was also a terrible, terrible product. The crappy screen, low ram, low performance, lack of software, etc. When I used one I considered it something for toddlers or preschoolers. The Classmate by comparison is something I could use right now if I had to and still be mildly productive with.
The market moves fast and OLPC wasn't ready for it. They built their house on not just last gen technology but last junky gen technology (milquetoast AMD designs, shovelware linux distro, etc). They got leapfrogged trivially. The fact that they couldn't see that is inexcusable. Moores law reigned supreme back then.
While the OLPC was ambitious, it pretty much is the poster boy for incompetent central planning by out of touch technocrats who thought they were smarter than the market, educators, and policy makers.
N.B.: I agree with pretty much the entirety of your post. My point was, and I still stand by it, merely that OLPC (hardware) was developed because nothing else in the price and performance (particularly battery performance) existed at the time. The relatively poor hardware was, unfortunately, necessary to achieve the 20 hour battery life that it could pull off. It's important to recall that one of the major targets for the device was areas with poor power infrastructure. The 4 hour life of the Classmate PC would not have been conducive to that environment.
However, I agree with the rest of your post. I was much more interested in the project from when I first heard about it (2005 or 2006) and lost a lot of interest once netbooks (and similar) came about, at reasonable costs, a couple years later. OLPC would have been much better off if they had focused on their software and curricula components, particularly being OS agnostic as much as possible (I recall a lot of teeth-gnashing about the idea of distributing Windows to 3rd world countries, particularly from FOSS zealots of the time). Let the hardware and OS vendors compete, provide a compelling suite of software, tools, and digital library that they (vendors) can all ship on their own systems instead.
I seem to remember around 2007 there were several netbook manufactures who were just getting started calling out the OLPC as the direct inspiration for creating the market segment in the first place. Meanwhile, that article describes Intel's device as having a better CPU, but half the screen rez, 1/5 the battery life and costing hopefully only a bit more than twice as much two years later.
The OLPC hardware was very low-spec. But, it was a brilliant design for being so low spec. It's UI was considered alien at the time. But, today the app-based, fileless approach has been proven out by the tablet market to be much better for the mainstream, non-hardcore user.
It's unfortunate they insisted on running so much in Python on such a tiny CPU. I understand that they really wanted view/modify source as standard for the kids. But, the only language I could see that working out well at the time would have been Lua --which doesn't have nearly as much clout as Python even today.
In my outside observation, OLPC was mostly tanked by politicians declaring that "If our kids can't use it to learn Microsoft Office, I don't want it even for free." There were a multitude of issues. But, that one really cut off the investment that could have overcome them.
No, there was plenty of existing hardware and software that could have been used, but the whole thing was full of extraordinary "not invented here"isms that made learning the OLPC a learning dead end of nontransferable skills.
In 2005, what was an existing, portable, long battery-life device that cost under $200 that would have satisfied the goal of getting a PC into students' hands?
I once loved Second Life. So many science demonstrations (NASA, NOAA, and many museums setup virtual campuses there that are still amazing), gatherings, museums, and educational experiences that only a virtual world can provide. Then it became a ghost town (with some mostly-NSFW exceptions).
The problem is that SL is closed source. I once tried to run a virtual field trip with 10 kids to show them around, but Linden Labs blocked us for having too many users access SL from the same IP address. You have to lease property on Linden Labs servers for a presence there. Why am I going to invest the time and energy into building something amazing on rented space that will vanish the moment I stop paying for it?
If SL were to be released as a open-protocol, like WWW, where anyone can pop a server onto the Internet and host a community, I think SL would thrive again. I would love to see that.
I found a nine-year-old blogpost I wrote on the science attractions in SL. It really was an amazing place. I would really like to see something like it again, but open. I think there's a huge opportunity there for someone with the right technical capabilities.
The problem I've run into is that it runs worse on my brand-new computer than Second Life ever did, and that's even on the mostly-empty demo areas they have.
Did OS/Hypergrid ever solve the avatar/inventory issues? I haven't followed SL/OS for the last few years so my information is probably out-of-date but when I last used it, it was still a hit-and-miss experience.
I'm skeptical that closed source is what's actually holding SL back. I think splintering it into lots of shards would actually hurt something like this, which is closer to a social network. Once you shard it out, you get even further from critical mass, lots of desolate servers with nothing going on and inconsistent rules/user experiences
Rather than sharding, consider federation. Anyone should be able to run a server and allow "doorways" or other access points to other servers, hyperlinks.
I recall, circa 2000, a project that I can't name at the moment that I toyed with in college that had this concept. You could host your own area and connect it to other areas hosted by other people. The name "Metaverse" is in my head, but googling it (quick search, probably missed something) was dominated by Stephenson's concept from Snow Crash and not a software project by that name.
> Rather than sharding, consider federation. Anyone should be able to run a server and allow "doorways" or other access points to other servers, hyperlinks.
> You could host your own area and connect it to other areas hosted by other people.
Sounds like Croquet/OpenCobalt/Qwaq/Teleplace/Terf (hooray for forks!). Different spaces can be hosted by anyone, like Web pages, and the hyperlink equivalent is a 'portal' which you can step through to go from one "world" to another.
As far as I understand it, the client processes the world via a stream of events, which a p2p protocol tries to keep consistent between all of the clients (e.g. user avatars) inhabiting a world. Going through a 'portal' switches your client to the another "world", and hence another event stream, so there's no central authority.
Edit: It seems to be p2p, rather than server-based, according to:
> The Open Cobalt application leverages peer-based messaging to eliminate the need for virtual world servers/commercial services and makes it very simple for end-users to create and securely share deeply collaborative virtual worlds that run on all major software operating systems.
You realize the content is more produced for the other users, right? You maintain your copyright, and literally fortunes have been made in SL by selling virtual goods.
...inside the captive environment maintained exclusively by Linden Labs. They may make noise about copyright, but the creator of the data does not have the same level of control over it as I might have over a book or music recording that I produce myself.
Linden Labs could, at any time, change the terms of service such that you must license all "your" assets in a particular way to keep them live on the servers.
The fortunes are few and far between, and largely the result of a negative-sum game with other users. You cannot take more out than other users pay in, minus the Linden tax. The house always wins, yet only about 5000 users have even profited as much as $10.
The creator of the data does not have the same level of control over it as I might have over a book or music recording that I produce myself.
How do you figure? If you take a music recording and want to put it out there on iTunes or Spotify, you have to license it to the company so they can do the thing you're asking them to do!
If you want to sell 3d models, textures, and code, you can do that on your own website right now. If you want to put that content up on Second Life (or literally any other third party), you're going to have to give them the rights to do so.
I'm also curious about the source on that 5K/$10 number. I'm a mediocre scripter, and made about $100 there on a small gadget that translates text based on a popular webcomic series.
It looks like I was relying on old information. It was from an article about Anshe Chung, the Second Life millionaire character of Ailin Graef. Other (more recent) articles indicate that in 2007, About 50k accounts were better than break-even profitable on a monthly basis.
You can put your music recording on both iTunes and Spotify, or neither. Or BandCamp and SoundCloud and your own CDs. If you don't like the terms of service, you can distribute your stuff by other channels.
When you're producing content in SL, for SL, it's like you're writing sheet music for the Gronxophone, which is a trademarked, patented instrument that only one person on Earth can play. If you don't like the terms, tough.
I know this is anecdote and a single data point, but I find it interesting nonetheless.
My mom, 60+ year old, decided she wanted to learn proper English. Being an aircraft controller, she already knew enough to get by on the job (communication is pretty much standardized). But nowhere near enough to get by in the US.
After a few conventional approaches (such as physical English classes), she somehow found out about Second Life and enrolled in a (paid!) course.
Fast forward a couple of years. She can now hold conversations in English and go out by herself without a family member translating. As an observer, the progress was outstanding.
One thing that I have noticed is that it can be way more interactive than "normal" classes. For instance, there was a lesson that took place at an airport. So, instead of just an opening dialog in a book, and then some discussion, their avatars were at a functioning airport. So they could go all the way from arriving at the airport to boarding the plane, having dialogue all the way. Not sure if the airplane did or could take off though.
At another one that I witnessed, they were at a clothing store. So they could actually interact with a human shopkeeper, buy the article of clothing they wanted, then put on their avatar.
Having attended "standard" classes all my life, I was blown away by how much the virtual environment would foster dialogue. Usually, a teacher has a hard time coaxing this kind of interaction in a classroom setting.
Also of importance is that, since you are hidden behind an avatar, shy people seem more comfortable. Things such as race and, to an extent, age, are completely hidden.
Not sure if VR would help anything, except for the "cool" factor, which can't be discounted.
The ability to cheaply simulate environments is at the core of many VR education/training concepts. I hadn't heard of it being applied to language education, that's really interesting!
Second Life specifically excels at being a platform for roleplaying (in the educational sense as well as the D&D sense). The built-in creation tools allow people to very quickly create environments that have enough fidelity to be a significant step up from 'this table is now an airplane' and thanks to the marketplace one can often quickly and cheaply acquire ready-made props for virtually any scenario. Using the more advanced creation tools one can even create video-game-level fidelity however that is usually not necessary for educational environments.
Second Life is/was actually great especially in its open content and customization system and economy.
But there's kind of a weird thing about Second Life and VR and video games. Video games have really filled the virtual reality space without being called virtual reality.
For example, the detailed simulation of Los Angeles in GTA 5 or the details and immersive story in The Witcher 3. It isn't usually called virtual reality and doesn't usually involve HMDs, but these types of games are largely what people were thinking of when they started talking about VR many years ago.
Now there are things like GMod (Garry's Mod), Minecraft, Space Engineers, Scrap Mechanic etc. that allow you to satisfying your virtual reality engineering itch with component-based development i.e. snapping things together, which is just much more practical, efficient, and just easier for people who are trying to entertain themselves than doing a bunch of scripting in OpenSim/SL/OpenCroquet/High Fidelity/etc.
Look at the first person perspective in the Mirror's Edge series. Its a very cyperbunk-type vision along the lines of early VR concepts.
Anyway long story short, VR is a massive business, they just call it "gaming" and we don't bother wearing HMDs.
48 comments
[ 3.6 ms ] story [ 105 ms ] threadIf the pervasive influence of social media in our times has taught us anything, it is that a platform that does interface with our everyday existence will never be more than a cool tech demo.
Off the top of my head, for success VR needs to provide:
Until the VR experience can supplant some of the activities above, I doubt it will take off.Also have had many poor quality video conferences on Lync/Skype for Business where all participants have been on the same fat academic ISP.
It seems to me we're much closer to a great AR experience than truly great and immersive VR.
[1]: https://www.relay.fm/rocket/20
I've got some VR hiking goggles to sell you. They are cheap, lightweight and totally wireless, but only work in the woods. They were featured on a recent SouthPark episode.
Too many discussions focus on VR being a drop-in replacement for reality. Nobody should expect them to mimic reality any more than they expect computer screens to mimic sunlight. I don't much care if VR helms don't allow me to jog up Everest. I just want to be free to look around as if I were standing on the top.
Also, if a VR system can perform well as a hiking simulator it would be able to realistically simulate things that don't exist, places that are too difficult/expensive for an average person to visit, or experiences that are too dangerous to attempt in reality.
EDIT: For example, I'd love to visit the moon or Mars but it's highly unlikely I'll ever be able to physically go there. A highly immersive VR simulation would be very appealing to me!
I mean right now, VR is very, very rough. We're basically wrapping the senses in hardware. (Analogy: Building a robot to press keyboard keys).
Eventually we'll get to the point of not needing that anymore. (Analogy: The robot is connected directly to the keyboard port)
..and the "drop-in replacement for reality" dream that's been around since the 90's or so can begin happening. The tech is in its infancy.
This was true before with SL, and it'll be true with VR in the future. It'll be great for certain kinds of games, therapy, housing walk throughs, and other contexts for which a synthetic 3D world is useful, but it won't be the panacea that replaces all other forms of UI. Immersion also has a cost -- it's very intense and you don't want it for hours and hours and hours. Just like you'd never want to be on a rollercoaster for 8h, VR for 8h will also be too much (unless the world itself is not an intense game, you don't have a screen 2" from your eyeballs, etc).
Real paintings have brushstrokes, bumps, texture and depth. Even the most high-res scans of paintings utterly fail to capture that. The Met has a fantastic collection of art scans that you can browse online[1]. Some of those works are stunning, even in thumbnails. But they still pale in comparison to how they are when you're up close, when you can see the grain of the canvas, the brushstrokes and dabs of paint, the way that it all pieces together. I always view a painting from multiple angles, looking from the left and right, tilting my head. Texture matters, a lot. VR can provide that in a way a 2d (or even flat 3d) monitor never could.
The real promise is an endless, densely packed 3d museum, where you can single out an exhibit, open it in it's own room, and enlarge/move it as you see fit. You walk down the aisle, viewing many pieces in an infinite array, and zoom in when you like one. You get many constantly changing angles, which is what our brains were built to deal with.
It's not there yet. Tilt Brush gives us a taste. More will come.
[1] http://www.metmuseum.org/art/collection
Instead of reproducing existing things poorly that you seldom want to do anyway, VR should be about doing things that are either impossible in real life or are compelling enough even if done in a non-realistic form.
It is a mistake to look at VR UIs as strictly 3D. There is no reason you could not scroll through your photos on a 2D screen with your fingers in VR. You can even have the added benefit of spawning multiple giant 2D screens, then grabbing photos off them for a scrap book or a virtual art gallery.
I've shared a great deal of locomotion experiences with other Vive owners. VR sickness seems a lot rarer than it was hyped up to be, and definitely a lot easier to acclimate to. I wonder how many vomiting journalists would be fine if they eased into it over a day or two. I can fly around in Windlands without flinching these days. Don't get me wrong, nausea is still a real design constraint, but we might be a lot more adaptive to VR and weird locomotion.
On the topic of session lengths, I firmly believe this is just an issue with content and ergonomics. I've done a few 6 hours sessions in multiplayer games, the problem is barely any exist. I haven't heard anyone experience "immersion fatigue", but some games will cause actual physical fatigue. Thankfully I already stand all day and it isn't an issue. I think if you already can play a game for 8 hours, VR will be there to provide the same in the future.
-- It's easy to make things hard to use, it's hard to make them easy to use. --
Yes, one can use 2D interfaces inside 3D worlds but any 3D system short of the matrix will have a noticeable quality degradation (e.g. visual fidelity, haptic feedback, fatigue, etc) compared to directly using a 2D interface in real life. A few edge cases might become feasible much sooner, e.g. using VR to emulate large screens but that's hardly 'Virtual Reality'. Until we come up with meaningful ways to represent more kinds of information in 3D and/or create a matrix-style experience, VR will most likely be limited to applications that are inherently 3D.
This obviously includes entertainment (VR games, 'movies', 3D art) and CAD/CAM/3D modelling which are already very much 3D in nature and internal representation, even if they are currently usually projected onto 2D screens. Outside of that, the main 'educational' usage currently consists of building 3D environments (labs, buildings, factories, complex machinery) that emulate real-life environments in order to familiarize trainees with the workflows in those places. This can be surprisingly effective, especially considering how much cheaper VR mockups are compared to e.g. NASA's real-life 1:1 ISS training replica.
These applications are worthwhile in their own right and they will probably be able to sustain a decent market for VR hardware and development. However, I'm personally much more interested in seeing what kind of concepts that aren't already inherently 3D can be enhanced or revolutionized by adding another dimension.
If we see social VR in education it'll be impromptu study groups in Altspace or screensharing in Bigscreen. It'll happen dynamically as kids decide what platform they prefer. Teachers will probably make similiar decisions on a per teacher/classroom way. I've already spent a bit of time in Altspace and Bigscreen and its very impressive from many aspects like usability, discoverability, presence, features, etc. I can't imagine using a SL-like system. Its too messy, ugly, and complex. It feels very Web 1.0.
Also, my god, the uncanny valley in SL is inexcusable. Either do photorealism or use cartoony avatars, there's really no room for middle ground here.
To be fair to OLPC, it just predated those other devices. It [the hardware] was developed precisely because cheap laptops didn't yet exist, and then the form and market came into existence and obsoleted a large part of the project's efforts (the physical hardware development, not its educational goals).
http://arstechnica.com/uncategorized/2007/03/acomparison-of-...
>nd then the form and market came into existence and obsoleted a large part of the project's efforts
It was also a terrible, terrible product. The crappy screen, low ram, low performance, lack of software, etc. When I used one I considered it something for toddlers or preschoolers. The Classmate by comparison is something I could use right now if I had to and still be mildly productive with.
The market moves fast and OLPC wasn't ready for it. They built their house on not just last gen technology but last junky gen technology (milquetoast AMD designs, shovelware linux distro, etc). They got leapfrogged trivially. The fact that they couldn't see that is inexcusable. Moores law reigned supreme back then.
While the OLPC was ambitious, it pretty much is the poster boy for incompetent central planning by out of touch technocrats who thought they were smarter than the market, educators, and policy makers.
However, I agree with the rest of your post. I was much more interested in the project from when I first heard about it (2005 or 2006) and lost a lot of interest once netbooks (and similar) came about, at reasonable costs, a couple years later. OLPC would have been much better off if they had focused on their software and curricula components, particularly being OS agnostic as much as possible (I recall a lot of teeth-gnashing about the idea of distributing Windows to 3rd world countries, particularly from FOSS zealots of the time). Let the hardware and OS vendors compete, provide a compelling suite of software, tools, and digital library that they (vendors) can all ship on their own systems instead.
The OLPC hardware was very low-spec. But, it was a brilliant design for being so low spec. It's UI was considered alien at the time. But, today the app-based, fileless approach has been proven out by the tablet market to be much better for the mainstream, non-hardcore user.
It's unfortunate they insisted on running so much in Python on such a tiny CPU. I understand that they really wanted view/modify source as standard for the kids. But, the only language I could see that working out well at the time would have been Lua --which doesn't have nearly as much clout as Python even today.
In my outside observation, OLPC was mostly tanked by politicians declaring that "If our kids can't use it to learn Microsoft Office, I don't want it even for free." There were a multitude of issues. But, that one really cut off the investment that could have overcome them.
The problem is that SL is closed source. I once tried to run a virtual field trip with 10 kids to show them around, but Linden Labs blocked us for having too many users access SL from the same IP address. You have to lease property on Linden Labs servers for a presence there. Why am I going to invest the time and energy into building something amazing on rented space that will vanish the moment I stop paying for it?
If SL were to be released as a open-protocol, like WWW, where anyone can pop a server onto the Internet and host a community, I think SL would thrive again. I would love to see that.
I found a nine-year-old blogpost I wrote on the science attractions in SL. It really was an amazing place. I would really like to see something like it again, but open. I think there's a huge opportunity there for someone with the right technical capabilities.
http://ideonexus.com/2007/03/03/science-in-second-life/
The problem I've run into is that it runs worse on my brand-new computer than Second Life ever did, and that's even on the mostly-empty demo areas they have.
1. http://opensimulator.org 2. http://opensimworld.com 3. http://highfidelity.io
I recall, circa 2000, a project that I can't name at the moment that I toyed with in college that had this concept. You could host your own area and connect it to other areas hosted by other people. The name "Metaverse" is in my head, but googling it (quick search, probably missed something) was dominated by Stephenson's concept from Snow Crash and not a software project by that name.
> You could host your own area and connect it to other areas hosted by other people.
Sounds like Croquet/OpenCobalt/Qwaq/Teleplace/Terf (hooray for forks!). Different spaces can be hosted by anyone, like Web pages, and the hyperlink equivalent is a 'portal' which you can step through to go from one "world" to another.
As far as I understand it, the client processes the world via a stream of events, which a p2p protocol tries to keep consistent between all of the clients (e.g. user avatars) inhabiting a world. Going through a 'portal' switches your client to the another "world", and hence another event stream, so there's no central authority.
See https://en.wikipedia.org/wiki/Open_Cobalt for example.
Edit: It seems to be p2p, rather than server-based, according to:
> The Open Cobalt application leverages peer-based messaging to eliminate the need for virtual world servers/commercial services and makes it very simple for end-users to create and securely share deeply collaborative virtual worlds that run on all major software operating systems.
Linden Labs could, at any time, change the terms of service such that you must license all "your" assets in a particular way to keep them live on the servers.
The fortunes are few and far between, and largely the result of a negative-sum game with other users. You cannot take more out than other users pay in, minus the Linden tax. The house always wins, yet only about 5000 users have even profited as much as $10.
How do you figure? If you take a music recording and want to put it out there on iTunes or Spotify, you have to license it to the company so they can do the thing you're asking them to do!
If you want to sell 3d models, textures, and code, you can do that on your own website right now. If you want to put that content up on Second Life (or literally any other third party), you're going to have to give them the rights to do so.
I'm also curious about the source on that 5K/$10 number. I'm a mediocre scripter, and made about $100 there on a small gadget that translates text based on a popular webcomic series.
http://abcnews.go.com/Technology/story?id=3527537&page=1
https://en.wikipedia.org/wiki/Economy_of_Second_Life
https://www.technologyreview.com/s/408413/making-money-in-se...
http://www.washingtonpost.com/wp-dyn/content/article/2010/03...
You can put your music recording on both iTunes and Spotify, or neither. Or BandCamp and SoundCloud and your own CDs. If you don't like the terms of service, you can distribute your stuff by other channels.
When you're producing content in SL, for SL, it's like you're writing sheet music for the Gronxophone, which is a trademarked, patented instrument that only one person on Earth can play. If you don't like the terms, tough.
My mom, 60+ year old, decided she wanted to learn proper English. Being an aircraft controller, she already knew enough to get by on the job (communication is pretty much standardized). But nowhere near enough to get by in the US.
After a few conventional approaches (such as physical English classes), she somehow found out about Second Life and enrolled in a (paid!) course.
Fast forward a couple of years. She can now hold conversations in English and go out by herself without a family member translating. As an observer, the progress was outstanding.
One thing that I have noticed is that it can be way more interactive than "normal" classes. For instance, there was a lesson that took place at an airport. So, instead of just an opening dialog in a book, and then some discussion, their avatars were at a functioning airport. So they could go all the way from arriving at the airport to boarding the plane, having dialogue all the way. Not sure if the airplane did or could take off though.
At another one that I witnessed, they were at a clothing store. So they could actually interact with a human shopkeeper, buy the article of clothing they wanted, then put on their avatar.
Having attended "standard" classes all my life, I was blown away by how much the virtual environment would foster dialogue. Usually, a teacher has a hard time coaxing this kind of interaction in a classroom setting.
Also of importance is that, since you are hidden behind an avatar, shy people seem more comfortable. Things such as race and, to an extent, age, are completely hidden.
Not sure if VR would help anything, except for the "cool" factor, which can't be discounted.
Second Life specifically excels at being a platform for roleplaying (in the educational sense as well as the D&D sense). The built-in creation tools allow people to very quickly create environments that have enough fidelity to be a significant step up from 'this table is now an airplane' and thanks to the marketplace one can often quickly and cheaply acquire ready-made props for virtually any scenario. Using the more advanced creation tools one can even create video-game-level fidelity however that is usually not necessary for educational environments.
But there's kind of a weird thing about Second Life and VR and video games. Video games have really filled the virtual reality space without being called virtual reality.
For example, the detailed simulation of Los Angeles in GTA 5 or the details and immersive story in The Witcher 3. It isn't usually called virtual reality and doesn't usually involve HMDs, but these types of games are largely what people were thinking of when they started talking about VR many years ago.
Now there are things like GMod (Garry's Mod), Minecraft, Space Engineers, Scrap Mechanic etc. that allow you to satisfying your virtual reality engineering itch with component-based development i.e. snapping things together, which is just much more practical, efficient, and just easier for people who are trying to entertain themselves than doing a bunch of scripting in OpenSim/SL/OpenCroquet/High Fidelity/etc.
Look at the first person perspective in the Mirror's Edge series. Its a very cyperbunk-type vision along the lines of early VR concepts.
Anyway long story short, VR is a massive business, they just call it "gaming" and we don't bother wearing HMDs.