Neal Stephenson Joins Magic Leap
magicleap.com
magicleap.com
--- Here’s where you’re probably expecting the sales pitch about how mind-blowingly awesome the demo was. But it’s a little more interesting than that. Yes, I saw something on that optical table I had never seen before--something that only Magic Leap, as far as I know, is capable of doing. And it was pretty cool. But what fascinated me wasn’t what Magic Leap had done but rather what it was about to start doing.
Magic Leap is mustering an arsenal of techniques--some tried and true, others unbelievably advanced--to produce a synthesized light field that falls upon the retina in the same way as light reflected from real objects in your environment. Depth perception, in this system, isn’t just a trick played on the brain by showing it two slightly different images.
Most of the work to be done is in applied physics, with a sizable dollop of biology--for there’s no way to make this happen without an intimate understanding of how the eye sees, and the brain assembles a three-dimensional model of reality. I’m fascinated by the science, but not qualified to work on it. Where I hope I can be of use is in thinking about what to do with this tech once it is available to the general public. "Chief Futurist" runs the risk of being a disembodied brain on a stick. I took the job on the understanding that I would have the opportunity to get a few things done. ---
I think that there's a presumption (that I've shared) that Magic Leap is trying to make AR goggles or glasses. But, as has been extensively commented upon in the past, that's just crazy. We're just barely at the point of doing semi-decent VR (Oculus Rift) and just barely at the point of doing semi-decent wearable heads-up displays (Google Glass). The idea that Magic Leap could in any foreseeable timeframe create a device that has all the virtues of the Rift + Glass + A huge dose of additional technology on top of both is just laughable.
But if they're trying for something much heavier-weight, like the ability to create convincing illusions not in the form-factor of "some goggles," but rather, "a specially prepared room and table," then that's maybe a little more realistic -- and of course less obviously revolutionary.
An "optical table" is to optical technologies as a solderless breadboard is to electronics.
Basically, it's a big, stable platform with lots of threaded holes of a standard size and pitch for attaching lasers, mirrors, etc. Most have some kind of pneumatic isolation or damping to keep vibrations from being transmitted from the floor. Things like interference phenomena are sensitive to displacements of a few nanometers, so you really don't want things like passing trucks to ruin your experiments!
Examples here: http://www.newport.com/Optical-Table-Selection-Guide/140219/...
But I mention the "they're trying to build like the equivalent of the early table-based Microsoft Surface (before that meant a tablet), but pseudo-holographic instead" idea just because it's so unbelievable to me that stand-alone goggles can possibly deliver what they're claiming.
If you read through the depths of Magic Leap's site combined with their patent applications, it becomes clear that they are trying to develop a set of goggles which combines some form of projection onto the retina [0] with some form of selective blocking [1] (to give contrast, and prevent the projected images from appearing as hazy mirages over the light otherwise reaching the retina).
Especially the blocking would be impossible to achieve with anything but goggles.
[0]: https://www.google.com/patents/US20140071539 [1]: https://www.google.com/patents/US20130128230
Still unclear. They do seem to imply they want it to be an unconstrained mobile AR device, but that is indeed ambitious enough to warrant skepticism. Walk-around tracking for home/anywhere is still an unsolved problem for Oculus, and overlay AR is at least several times more demanding.
That's actually how (kinda) Stephenson's VR system he describes in Snow Crash works.
This kind of gets away from many of the issues you have with the Oculus Rift where you have such a tiny window of time (20ms or so) to react to how the person is moving his head. You still have to change the image based on the user's position, but not as much on head rotation which is the really hard part. Just each eye's location in space matters.
Multiple users would likely require multiple projectors.
probably because they's explicitly what they've claimed they're working on in their recent funding announcement. The CEO described their product as a "lightweight wearable".
I think they're mostly in patent hell.
It would be interesting if this was progressing because some major patent roadblocks where about to expire.
Wikipedia suggests that he's a sci-fi author, is there a particular reason for his fame, could you recommend some of his works?
The Baroque Cycle for a semi-fictional view of the beginnings of science - the Newton and Hooke era.
Anathem for an interesting take on the philosophy of science disguised as a sci-fi epic.
IMO it is one of his best works. Nanotech/Networks/Crypto for the masses to understand. I read, loved and was caught up in the VR fever of the 90's via Snow Crash, and love his other books, but TDA is the one I'll never get rid of.
* cult sex scenes
* a forced-participation theater that humiliates you using Occulus Rift-technology
* the Kill Bill-esque ending
* ... and, most of all, the idea of continous education using an immersive Minecraft-like book/world that expands in complexity as your education grows.
Too bad we (the hackers) never completed even a crude version of the Primer (the book in the last bullet point) for the young geeks out there.
I like your point; it compliments itself nicely with my previous one. Soon enough, computers may be able to do many things very well -- however, hackers are not catching up to corporations.
What I mean: if I remember correctly, in the DA world TV and big companies have as big as a grip on general populace as they have to day, but hackers are able to create alternatives, like the mentioned Primer for children's education.
In the real world, "hackers" (or those with the technical know-how to be one) love Apple and Google as much as the rest of the populace does, and leave the big things (OS, main APIs, maps, voice assistant, their personal data, ebook stores, videos available to young children) to them with very little opposition.
Damn, I'm going to have to read it again now. Still, it's about time for a refresh...
There was a time when I thought Snow Crash was the best. There was a time when Cryptonomicon was a lot of fun (still is). Nowadays I incline slightly more towards the 'philosophical opus' type of vibe that Anathem gives off.
Anyway, all his books are pretty good representatives of one sub-genre or another. He's a very good author, and he wrote in a lot of different keys through his career so far.
The Baroque Cycle is a massive piece of work spanning 3 volumes, comprised of 8 nominally independent books. If it seems intimidating, just try the first one and see if you're not hooked. I'd love it if there was twice as much material.
Anathem is by far my favorite. Its hooks take longer to set, but for me they set much deeper. There is a lot going on in this book, and it will truly blow your mind if you let it.
The beginning is actually really tough going as a completely new reader today. It's just so ridiculous. I can see where he was coming from, as I grew up in that era, but it's actually pretty bizarre now given the reality is nation states, religion and banks turned out to be so much more powerful than corporations.
Which is one of the perils of predictions in ageing sci-fi.
I've been on a sci-fi kick recently of all the classics I never read (William Gibson, Ender's Game, The Mars Trilogy, Forever War, Starship Troopers, A Canticle For Leibowitz, Philip K. Dick, Hyperion Cantos, Ringworld) and re-reading some I've not read for a long time (Foundation Series).
I personally found that Snow Crash is by far the most dated book. Even Ringworld and the foundation series were better.
Some of the space opera, on the other hand, was so earnest and certain that we'd be flying around at light-speed by now you feel almost disappointed for the authors.
All cyberpunk is like that, for reasons that are pretty obvious.
> Even Ringworld and the foundation series were better.
Of course. Physical reality changes much more slowly.
And when I refer to Foundation & Ringworld I meant that they are from the 60s and so have some weird cultural ideals as well as some (unintentional) misogyny & racism in the foundation series.
I've a major in Physics and it took me a couple readings to figure out all (well, most of) the connections therein. My favorite game to play while reading Anathem was figuring out where are the borders between historical fact (translated into the fictional world of Arbre, of course), current hypotheses within present-day science, and just downright fiction. Quick quiz: is "geometrodynamics" Stephenson's invention, or a term used in the real world? You get puzzles like that at every step, some easier, some harder.
A few examples that stand out:
Actual history of science - well, Thelenes, Adrakhones, Saunt Tredegarh, Saunt Muncoster, etc. (again, real people disguised under the mask of Arbran characters)
Current hypotheses - the whole Multiverse thing, the Fraa Paphlagon / Hugh Everett parallel.
Out-and-out fiction - eh... this is harder. The Wick, maybe?
And then there's Fraa Jad, all alone in a category of his own. :) I daresay one of the most striking, memorable characters in all sci-fi - if you get the point of the whole book.
[1] http://anathem.wikia.com/wiki/Earth%E2%80%93Arbre_Correlatio...
The Mongoliad semi-fictional view of mid-thirteenth century Mongol invasion of Europe
Reamde MMO gold farming, social networking, criminal methods of the Russian mafia, Islamic terrorists
Reamde starts off with a lot of interesting ideas, and then morphs into quite possibly the worst watered-down, airport-paperback, fourth-rate-Tom-Clancy-triller nonsense I have ever read. Avoid it at all costs. Unbelievable plot and character motives. ick.
If you can suspend disbelief at the sheer ridiculousness of the situation, it's pretty awesome.
Writing aside I've found the plot pretty slow with not much interesting happening for most of the Zula portions (ie. middle half of the book). Maybe Stephenson's level of detail just isn't for me but it really wants for editing.
Anathem and Diamond Age were harder for me because of the depth of ideas. I had to slow down and think to get through them.
Baroque Cycle was harder because of the sheer number of characters with multiple and/or similar names, which is realistic but annoying. There's a list of characters in the back of the first book, which helps, but it's annoying to have to keep the first book handy when you're reading the others.
Stephenson does have a reputation for starting great books and not knowing how to finish them. But I think I've gotten more than my money's worth out of all of them.
http://www.cryptonomicon.com/beginning.html
Warning: written fifteen years ago, also really long.
I'd recommend his "Mother Earth Mother Board" article (more like a small book, be warned) as the definitive tome on the physical reality of how the Internet is held together across the world: http://archive.wired.com/wired/archive/4.12/ffglass.html
Google Earth was heavily influenced by ideas in his books.
The Kindle was codenamed "Fiona", after Fiona Hackworth, a character in one of his books who uses a super-duper e-book.
Gizmodo would beg to differ: http://gizmodo.com/how-magic-leap-is-secretly-creating-a-new...
I was recently pulled into the wormhole of present day speculations (science??) on this. And wow, sounds like what they are working on may yield real, new understanding on cognition and consciousness.
One of the entertaining reads: "Space, self, and the theater of consciousness" by Trehub http://people.umass.edu/trehub/YCCOG828%20copy.pdf
http://gizmodo.com/how-magic-leap-is-secretly-creating-a-new...
It was only until I clicked "Wizards Wanted" (yes, I get it. Magic.) that I sort of got an answer.
Edit: Actually, the Developers section has some more information. But my point still stands.
Two utilitarian sentences on what your thing is, and then all the literate prose you can produce. But gimme those two sentences.
If you're going to be stealthy, be stealthy (e.g. http://quanttus.com/). If you're going to attempt to market your company, then market it properly. This does neither.
My point was that magic leap uses a lot more words to tease, and as a result, the site borders on nonsensical.
And if the point of a site is to tell you what a company does or is working on, then that information shouldn't be buried in a random section.
But ultimately, it's not a big deal. Just the first thing that I thought of when I clicked on their site.
And so it is more practical for them to tease and build interest.
...not that it matters, seeing as they seem to have already built interest with the people that matter, investors.
Do you honestly not think that when they have something to sell they'll have a marketing strategy?
My feedback, and similar feedback from others, is not a complaint, it's feedback. Or, as someone else once said, "treasure your bugs."
Have a word with yourself!
I wonder how far they are away from going to market? 24 months minimum?
From their site:
>Imagine being able to generate images indistinguishable from real objects and then being able to place those images seamlessly into the real world.
Sounds like you generate a hologram anywhere. And project it into the real world. Think Star Trek holodeck, but everywhere, or anywhere from what they are selling. And the holograms they're projecting would be almost indistinguishable from other objects (I highly doubt that honestly, subsurface reflection takes A LOT of processor horse power with 1 view alone n).
It will be interesting to see what they cook up.
Note that this effect is why 99% of the content in feature-length 3d movies appears at an apparent distance of more than 10 feet from you; the amount that accommodation falls off rapidly with distance, so it's a much less strong effect at that distance.
Guess it plans to sell hype.
http://www.csmonitor.com/Innovation/2013/1019/Inventions-tha...
(Source: if you read EVERY page on their mysterious web site, you can eventually suss it out)
Goes back to Carmack's idea presented in one of his Quakecon sessions (although obviously the idea has been in sci fi for a long time).
This is interesting, I have an idea of how it might work:
Producing a synthesized light field could be done with a huge array of tiny mirrors or refractive material (I know stuff like this has been done in the past). By simulating directional rays coming in, I'm guessing your eyes could be tricked into thinking there is depth since the movement of each eye would change the amount of light absorbed (i.e. "focusing" on different objects in the light field). I could be wrong about this, have not thought it through very well.
All that said, they've raised over $500 million from Google Ventures and others, so I'm guessing there is something here, as lacking as the website is.
I say this as someone that is excited about VR, AR and would love to see whatever Magic Leap is cooking up, and is a huge Neal Stephenson fan.
http://io9.com/how-neal-stephenson-is-helping-to-make-snow-c...
Well PCs powerful enough for proper amount of voxels would let us do interesting things. Dwarf Fortress with a voxel-based high-res graphics could be interesting.
https://www.youtube.com/watch?v=w8J5BWL8oJY
It starts off kinda weird, you gotta give it a few minutes.
I hope it doesn't end up like the whole "revolutionary CPU maker hires Linus" thing.
There, it ends up precipitating a gigantic MMO war.
Well, no, I'm thinking, who the hell are they. And I can't seem to figure it out from your website.
The Magic Leap technology involves light field projection - kind of like the opposite of light field photography if you remember the Lytro camera.
http://www.gamespot.com/articles/another-kickstarter-game-bi...
If Clang failed because of a lack of budget, then the failure was on the part of the project owners to set their funding threshold appropriately.
Of course that's on Stephenson, and he acknowledges it. He didn't understand the investment market. It probably wasn't a great plan in retrospect. But this seems like an odd place to complain when a risky investment doesn't pay off like you'd like.
The problem 100% was that Neal Stephenson was incapable of doing the project himself and was just hiring devs to do it and then ran out of money. It was poorly managed because they had lots of traction and risked basically no capital on development. If you give me half a million dollars I would have no problem getting that project to an early access state. Honestly I could do it with half that.
I can say that because I actually know how to make both the hardware and software side by myself if it came to that.
(Even if you are able to donate a year of developer time, valued in the six figures, which in fairness to your point is not something Stephenson brought to the table along with his relatively modest personal assets. Semi-famous authors aren't as rich as people think, either.)
I'm not saying I'm impressed with what they made, but I'm not too surprised either.
What they were trying to make was basically a better wiimote, and a wiimote is essentially an arduino with a single accelerometer hooked up to it and a bluetooth receiver. I think without the budget restrictions of a wiimote you could probably use better/more responsive parts and then basically just mimic a wiimote with the wiimotion plus(which is just a single gyroscope) and call it a day. Or you could add several of each of those components and then average their output or do other clever math with their separate outputs. I think this would take a little more than a year if the person who was doing it knew what they were doing.
Then you need to make the game/demo which you would do in parallel with a team of 3-4 additional people.
"Even if you are able to donate a year of developer time, valued in the six figures"
See this is where misunderstandings with kickstarter begin. We assumed almost no risk in making Road Redemption. In addition to salary we get 100% of revenue less distribution. So not only did we get to draw salary during development, now we get all the sales revenue and we own the IP. It is a crazy good deal, nobody needs to donate anything.
It is basically like you get a bunch of VC money and also the VC's have no equity interest.
The problem comes in when someone who is unqualified tries to middleman the operation and takes a bunch of money but is totally incapable of doing what they said they would do. Now they both don't have enough money to actually get the shit done at cost, and don't want to give away all the equity. The biggest problem though is that they are unqualified to discern who is capable of doing the task because they don't know anything about how to actually do things. So they fritter away the money and then the project falls apart.
It's a garbage-in, garbage-out scenario that is really, really hard to solve in software. And come on, why am I explaining this to you? Why are there zero compelling or realistic Wii/U sword games?
It also doesn't sound like you've put any thought into the challenge of simulating realistic swordplay as a generic engine that supports different historically accurate fighting styles, with all the nitty physics, kinematics, and for that matter historical issues to be dealt with.
But I don't know why I'm having to defend the concept that software costs money. I think I'll just quote Neal's apology:
"Members of the team made large personal contributions of time and money to the project before, during, and after the Kickstarter phase. Some members, when all is said and done, absorbed significant financial losses. I am one of them; that has been my way of taking responsibility for this. The team had considerable incentives--emotional and financial--to see CLANG move on to the next round of funding. They showed intense dedication and dogged focus that I think most of our backers would find moving if the whole story were told. I will forever be grateful to them. In the end, however, additional fundraising efforts failed and forced the team to cut their losses and disband in search of steady work." https://www.kickstarter.com/projects/260688528/clang/posts/9...
These were intelligent people, working with passion, on something hard. I'm glad your Kickstarter was a success, but maybe a little humility is called for.
[0] As in accelerometer+gyro, in general. It's fundamentally flawed, and you can't buy your way out of it with better parts.
"It also doesn't sound like you've put any thought into the challenge of simulating realistic swordplay as a generic engine that supports different historically accurate fighting styles, with all the nitty physics, kinematics, and for that matter historical issues to be dealt with."
You are controlling the sword so I'm not really sure what you mean. You need to mocap a bunch of moves from someone fighting in a historic style for the AI and then have the AI randomly jump between a half dozen attack sequences. Before our kickstarter we had a great animator and a bunch of playstation eyes and were able to get good motorcycle fighting animations for a price in the thousands.
I'm sorry but this project was obviously mismanaged. The extreme detail that went into designing the arenas seemed to consume all the resources of the project. Look at the videos of what they had. It is like 100% artistic detail and 0% gameplay.
Congratulations on shipping a game. I mean it, that's real. I haven't done that. But stay humble.
I have 2 more ideas if that doesn't work well. Though I would probably end up combining all the ideas in another weighted ensemble on top.
The game part of this however is not really that complex and could be done on a regular schedule.
Let me guess, every arena had a bike shed off to one side?
"Let me guess, every arena had a bike shed off to one side?"
I'm guessing that was a shot at Road Redemption, and I do wish the art was better, but having a huge art team to make 50+ miles of track from scratch would cost a lot more than 150k, and leave no money for making the actual game. We used a lot of premade assets from turbosquid and the unity asset store. We used EZRoads to make all the tracks except the rooftop levels. Most of the animations we had to mocap/animate ourselves though because there just are not a lot of solid on motorcycle animations available.
The point is that we actually made a game and it is pretty fun(90% user review score), and we did it with a lot less than 500k. Art does not make a video game. That is why I am saying the clang project was mismanaged. The art they made is beautiful, and looks like it is all custom. They could've easily gotten 90% of that from the asset store and it would be worse, but it would also cost less than $1000. Then they have at least 400k left to make the game. The problem they had to solve was difficult, but with 400k they should've been able to make something awesome.
Infact, that is specifically why I funded that kickstarter.
I mean, I hear you, but this is more like Arthur C. Clarke joining a spaceflight organization. Yes, he's coming at it from a different direction, but plainly he's got some valid ideas because other people keep implementing them, directly or otherwise.