Informative comment, thank you.

True, if AR is not Artificial Intelligent, discrepancy will remain between the human and machine.

If I’m correct, there have already been some legal issues with regard to navigational systems and the ambulances, coming to late due to misdirection. All these effects need to be understood before massive AR applications will go mainstream and support more life-critical situations.

In terms of the information displayed, I see a major dichotomy between text and visual information. If the real-world view is relatively static, the subject has time to read and digest text, but if the real-world view is even moderately dynamic, the information MUST be graphic in order to be useful; there’s no time to digest text.

The next question is one of feedback from the contents of the subject’s real-world view to the AR content. If the AR computer can’t “see” what the subject sees, the AR content can’t be intimately tied to the subject’s real-world view. And even if the AR computer can “see” the real-world view, it’s tricky to superimpose the AR content on the view with enough placement accuracy to be effective. Significant misregistration or position tracking lag of the AR content will cause cognitive dissonance in the subject, resulting in confusion and disorientation. This could be disastrous in life-critical situations.

Speaking of which, there are some interesting legal issues. Suppose a firefighter has a HUD with AR, and the AR malfunctions. If the firefighter has come to trust the AR display, he/she may well make a mistake that could be injurious or even fatal to the subject or a victim he/she is trying to help. This is a legal can of worms.

Definately.
(Hardware) constraints are part of the limitations which we see right now within the Augmented Reality realm.

Like you refer to as well, robust vision on AR will develop the few coming years, past the hype, past several hardware constraints, past initial adoption by early majority and smartphone diffusion.
The applications we see now are also fairly easy to develop, the wait-and-see attitude is understanding from a business point of view, just like VC’s are not jumping on this tech right away. All signs that show this tech (for the mainstream) is in its initial phase.

“Artificial information about the environment and the objects in it can be stored and retrieved as an information layer on top of the real world view.”

Some of the weak apps you cite above are constrained by the ability to extract meaningful information about a user’s environment and objects in it. Simply relying on GPS to calibrate location (useless indoors, often inaccurate outdoors) and correlate with info embedded in, say, Google Maps, is akin to only using the internet for email when much more is available.

The availability of very low cost sensors combined with state-of-the-art ultra-low energy wireless communications capabilities like DASH7 (www.dash7.org) effectively means businesses can AR-enable assets and even people in ways that GPS will solve for only partially or not at all.

In 2010, I expect we’ll hear more about a more robust vision of AR that includes a larger set of potential options for enhancing your AR experience beyond GPS-based approaches.