In my last post I touched on the hardware limitations / requirements of reality transfer. High quality capture devices are on the horizon but are probably still a few years off before everyone can have one in their lappy-top without setting fire to their Levis. Next up... disillusionment.

You might be thinking that the visual effects that we see on TV and in movies are a good barometer for judging what we can do in an augmented / virtual application... kind of... but be careful not to carry your expectations too far. Everything in movies and TV these days is 'baked' with a gonk-load of hardware / software and for thousands of man-hours in post-production time. What polish we see... is achieved (purchased) with innumerable CPU cycles and pure herculean effort. Something that an average every-day developer won't have access to.

For an immersive real-time experience... you should probably dial in at a fraction of that to keep things manageable (it's a possibility that the hardware requirement will still be significant).

Now, a word from our sponsor... disillusionment. What causes it? We need a good example... hmmm... what about that expensive security camera system installed in San Francisco? That will do!

It was hoped that installing video cameras in various high-risk locations in the Bay Area would help reduce crime rates but in the end resulted in frustration.

"Some cars and bicycles going through the intersection show up on just a single frame."

Public officials weren't happy...

"If this is a representation of the system, we're throwing money away," said Theresa Sparks, president of the San Francisco Police Commission, after being shown the footage.

What's the problem?

But they acknowledged that most of the city's cameras achieve only 80 percent of the resolution they are capable of, and that they generate, at best, two to four frames per second because the city lacks the data storage space to accommodate more footage.

Oh yeah... it gets worse.

Officials with San Francisco's telecommunications office said they were doing their best with limited funds. They said they needed hundreds of thousands of dollars' worth of additional data storage space for the footage, which is kept for a week before being erased. The officials also said they were nearly $200,000 short on cash needed for camera maintenance and would have to tap into their own budget to cover the costs.

As I said before, I love to see a good demo and like to keep in touch with emerging trends in technology... but the injustice... and disillusionment is going to happen if we don't get some face time with people that understand how regulations, acts, and auditing requirements can turn a very cool concept with incredible potential into something that you probably won't want to include in your resume.

Would you be surprised at this point if I said that AR solutions are going to be expensive? Or... that most of the people that are going to want it don't really need it? How about... developers who hope to enter the field sucessfully will need to become well versed in GPS, epipolar geometry, fiduciary markers, image registration, RFID / barcodes, eigenfaces, Tums, etc (the more you know about what the world around you is constructed of - the more compelling your AR solutions will become).

The good news is that there is no requirement that any technology has to pass through the Trough of Disillusionment... the path you take is up to you.

Here are a few ideas that might help you keep it real...

Buy a GPS
If you don't have one... get one! Nah... buy two! Your perception of the world will change overnight and the knowledge you gain from dealing with accuracy gremlins will be invaluable.

Buy a Camera
A nice one... like a D50 or similar and put the photos on Flickr (make sure they are geotagged). In the process you'll be introduced to people that know tons more about image processing than you ever will.

Live in 2D
If your phone can run Scanbuy - get it... if it can't - throw it away and get one that can. Start using the app - you'll save some money and gain important knowledge about product markers.

Play in 3D
Try Geo Caching... or plan a monthly activity somewhere you haven't been. What kind of services can you depend on? Or you could spend a day in the field with your client... anything that you can do to see the world at a different angle can be useful.

Own the Demo
Never show anyone anything that is rendered optimally... 1/4 of what you expect the final product to be is a good start. Having internal throttling mechanisms in place at the start will make you look smarter in the end!

Own Performance
Make your customers define operational limits - worst to best... and get used to running all your tests at the bare minimum and absolute maximum and never in the middle. If you do this, road blocks will tend to surface sooner than later... giving you more time to either work around the obstacle, negotiate more time, or revisit requirements.