[FoRK] Project Glass

Eugen Leitl eugen at leitl.org
Fri Apr 6 11:28:50 PDT 2012


On Fri, Apr 06, 2012 at 01:30:34PM -0400, Bill Kearney wrote:
>
>> I see no difference between VR and AR conceptually.
>
> You can't be serious.  They're, literally, worlds apart.  VR is great  

Yes, I can be serious, and practice doesn't use them differently.
VR/AR has been seen as one thing by pioneers like Sutherland and
all practical applications of VR did not include autistic HMD
operators. Stanislaw Lem treated the whole problem space pretty
thoroughly in 1970s, actually. It's too bad he's virtually unknown
in the US.

> but hardly holds enough appeal as it requires stopping participation in  
> the 'real world' for it's immersive experience.  Yes, for some this is a  

Yes, you already said that, but it's not true. VR isn't Jaron Lanier
with cyberglove and Polehemus HMD in front of an SGI SkyRise. 
That's so overspecified it's a caricature. 

> desired effect.  AR, on the other hand, plays along with the wearer in  
> the larger context of the real world and all of it's non-automated  
> experiences.

Advanced AR is a mirror world overlay over the real world. Eventually, 
the mirror worlds will become richer than reality. 

>> AR is just aligned VR overlay.
>
> No. With VR someone else has to set up the world, with AR you just play  

The overlay rendered is a synthetic world. It's an aligned VR overlay,
where the mirror world you see happens to fit the geography. 

> along with the existing one, without having to do all the ground work.  
> Dead spots with no data in VR pretty much kill the whole experience.  

Yes, sufficiently sparse datasets are useless. I agree. 

> Meanwhile the real world fills in the gaps, making for a lot less work  
> to have AR be perceived as useful.

That's an orthogonal issue to the concept.

> I'm sure it won't take long for someone to start overlaying VR onto the  
> real world, as the opposite has gone nowhere (pun intended).  Games  

The opposite is true, non-aligned VR has a continous vector towards
real AR, aka Artificial Reality. Augmented Reality is a dead end, however.
Augmented Reality always assumes a monkey in the loop. Artificial Reality
doesn't.

> played in the context of AR I'd daresay would hold a LOT more interest  
> to a MUCH larger audience than cosplay geeks masturbating in Second  
> Life.  Certainly a market with more monetizing potential.

Again, you're using different words for the same thing. Anchor a SL
world to physical terrain with a HUD and it's exactly AR. 

>> It is interesting how little fundamental novelty we see in
>> IT today. What would you consider genuine innovations of
>> the last two decades?
>
> A move toward always-available and inexpensive to access information is  
> certainly a leap in implementation, if built upon innovations beforee.  
> But it's hard to be dismissive of the implementation just because the  
> idea isn't entirely 'new'.  Lots of things were 'thought about' in a  
> technical sense but went nowhere without the infrastructure and other  
> actors covering the costs.

Implementations are just great. I love implementations of hitherto sterile
concepts.

However, implementations per se do not result in novelty. What I'm looking
for is concepts that will become implementations some 20, 30, 50 years hence.

The concepts I'm not yet aware of, I mean. 

What do you think does exist today that will become great in 50 years?


More information about the FoRK mailing list