21 May 2017 Cute-VR versus Real-VR.
There's a big debate going on at the moment about what we should call the spectrum of realities that we can now create on our devices. The common terms VR for Virtual Reality and AR for Augmented Reality and their combination VR/AR are being contested by MR for Mixed Reality (which covers all parts of the spectrum), xR or XR meaning "an agnostic symbol X + Reality". Then there's Immersive Computing and, presumably others.
This has all come to a head because Microsoft's Build conference insisted on defining everything in terms that fit with their vision of the future which is "Windows Mixed Reality" firmly tied to the "Universal Windows Platform" which is their way of trying to claim a single, unified way of doing things which just so happens to do things in a way that suits Microsoft. The talks from Build were really interesting and the stuff they are developing isn't bad, but their relentless on-message devotion to UWP undermined their "universality" because for obvious reasons they don't want to welcome Rift, Vive etc. into their vision. It is telling that Edge now "enables the WebVR standard" but for some strange reason the only "standard" device that can work on it is Hololens and (when they emerge) the new Windows Mixed Reality headsets which are more like a Vive/Rift. What's the point of conforming to a standard when no other standard devices (like Rift and Vive) can work on it?
A side effect of all the Microsoft stuff (and it won't have Microsoft quaking in their shoes) is that I'm going to seriously ignore Microsoft's vision for VR for the time being. For me, proper, universal WebVR is the only thing that interests me and I never want to be locked into a VR world controlled by Microsoft.
But that's not the point of this blog. Instead, I want to introduce yet another terminology: "Cute-VR" versus "Real-VR".
Google Daydream is an awesome technology and I bought an expensive smartphone with apps and features I'll never use because it was Daydream Ready. When my Daydream headset arrived I was keen to try out what those amazing guys at Google had done. With their awesome skills and vast potential budgets this was going to be amazing. Except that it wasn't. Daydream was and still is merely cute except for YouTube VR which I really enjoy [Though when YouTube VR is enabled for the Vive/Rift I'll probably abandon Daydream]. Not only is Daydream cute, it's crippled. The controller is fantastic technology that does amazing things via smart algorithms. But it's still pretty useless. It doesn't really know where your hand is so when you point a laser beam from it, it's coming from only an approximation of your hand. It's unnatural, it shifts over time and one small mistake (or accidentally touching part of your phone in the headset) sends you somewhere from which you can't recover.
I spent a month trying to get one of my rather simpler VR science apps to work convincingly. Admittedly a lot of that month was due to my incompetence with the Android debugging system and some poor bits of coding which worked fine for Rift and Vive but were unnecessarily restrictive for Daydream. But even when I had my own act together, Daydream is just cute. It can't do real stuff.
At Google's IO/2017 a talk proudly suggested that we went to WebVRExperiments.com to see what could be achieved. It really is rather sad. A few cute apps that you can tire of in seconds. In Daydream itself there are a few free things showing off the power without the limitations of WebVR. Again, one tires of them in seconds. They are cute, cute, cute. This is NOT a criticism of the smart folk who created them. 3DoF controllers are just so restrictive. Even Rift, for many months, was little more than a toy till Touch came out. That was game-changing. Once you ignored the official Rift advice and got long cable extensions and set up the sensors to be almost room scale, Rift started to be compelling.
It's probably my fault but so far I've not seen any AR that has interested me in the slightest. I'm aware that some corporations are using Hololens for seriously good stuff like helping tech service guys debug elevator mechanisms and I'm prepared to believe that there are great uses for a high-quality head-mounted AR (though Hololens is still in early days and "high quality" is a matter of opinion).
But having the ability for a cat to spin around on one's desk thanks to one's phone running AR is not something that appeals to me in the slightest. And although I can intellectually appreciate that seeing a new (virtual) chair in your real living room might make purchase of that chair a bit more compelling, I'm far happier relying on my wife who has a great visual memory and can "see" a chair in a room far better than I ever could with the aid of technology.
I still remember the first time I stepped into a Vive. What blew me away wasn't so much the VR (which was awesome) but that I was so immediately accepting of being in this new world. Being roomscale right from the start, Vive VR was just instantly compelling. I've not had a visitor, young or old, who hasn't felt totally at ease within complex worlds in just a few seconds - though admittedly that's rather easy to do when my default option is to but them into Google Earth VR. They (we) all stumble a bit with knowing which things to press on the controllers, but as soon as those things are clear, the new world is totally compelling.
And my own VR Science worlds are still a real pleasure to enter - especially when I've forgotten what they are and go in to remind myself. A couple of YouTube videos give a flavour of what I'm trying to do: Polymers in Solvents and Adhesion VR. Sure, there are plenty of issues and I'm definitely not a great VR programmer, but all the same the worlds aren't "cute" and there is some serious purpose to what's going on.
There's an important point here. My own VR worlds and even the simple demo worlds on the Three.js site aren't very sophisticated and are pretty limiting. They can be far less sophisticated than the more complex cute worlds. The difference between cute and real isn't "complexity" but, to use Kent Bye's favourite phrase, "embodied reality". In real-VR even though you "only" replicate your head, your XYZ coordinates in the room and your two hands, that's enough to convince yourself that you are there, even if "there" is only a room with a few spinning boxes. I've not (yet) tried the extra Vive controllers but I suspect that the extra information they provide will be really nice to have, but not necessary. Cute worlds never give you the impression that you're there. And I just don't want to be in such an obviously false environment.
So, real-VR is the only thing that now interests me. The cute stuff seems to me to tarnish the brand. I had at first worried that Daydream wasn't taking off very fast. I now worry that it might take off too fast, because if it does it will give lots more people the feeling that VR/AR/MR is just some tacky extra complexity they don't need in their lives.
And in case you think I might have a thing against Google, let me announce Abbott's law of VR: "If it doesn't run Google Earth VR then it's not VR." I remember the original blog from Google when Earth VR came out for the Vive. It said words to the effect of "Don't complain that there are a few issues, just marvel, as we do, that this stuff is possible at all." And what a marvel it is. I wrote a big adventure story for my 9-year-old granddaughter that had her flying heroic helicopter flights on Mt Everest. Her first training for the story (and in the story) was to get to know Everest backwards, forwards and sideways. After just a few sessions in Earth VR she could not only fly around Everest pointing out important details, she could get lost in the Himalayas and still find her way back to Everest. She could even fly from the world's most dangerous airport (Lukla) up the valley to Base Camp then up to the summit.
Real-VR is in another league
That is what real-VR is all about. It's about changing how we do things in a fundamental way. My little VR Science experiments are nothing in the grand scheme of things. But, golly, they show that science and VR are made for each other. My limits are partly my coding skills but mostly my imagination. I happened recently to see a little Twitter video of someone using VR to position a stop-motion figure. Suddenly I realised that I'd missed a whole area of Chemistry VR. Creating molecules on a 2D screen is really rather painful. Doing it in VR would be so much easier! I probably don't have the skills to implement it, but I guarantee that someone will do it and there will be no going back to the old ways.
It's been almost a year
Just under a year ago I put my first VR Science app onto the web. It is so embarrassing to think about it! The interface had to have two major upgrades (it, like all my apps, now uses dat.guiVR) and I had to remove lots of stupidities and inefficiencies. I've got about 8 online now and I've only been delayed in creating more by my attempt to make some of them work within Daydream. Now I've abandoned cute-VR I can start planning some more. But what really amazes me is that, as far as I can tell, there isn't a vast amount of science WebVR out there. Where are all the bright young scientists/coders who should be doing this stuff? Come on, scientists, real-VR is waiting for you. Just remember not to waste your time on the cute stuff.