You would still need the power to transmit a live feed from multiple cameras and sensors. That will still take a battery with some grunt.

I am a firm believer that the end design for most robots will be a combination like you suggest. That the robots will be fairly stupid and controlled by a central powerful computer. But if the robot is moving with any haste then it will need instant feedback loops running between its sensors and motors.

But even in this scenario on a farm the central computer will ideally be battery powered even if it isn't mobile. In this case I could see it also being solar powered. Just sitting way out in the field directing and fuelling a bunch of its little mobile friends.

Also there will be other uses for robots where transmitting to a central powerful computer isn't really an option. Robots that are sent way out into the field. Say water quality sampling along a river or flying huge distances doing air quality surveys.

So autonomous robots will really come into their own when they have the portable power to actually do stuff without breaking the bank.

I build robots and they all suck, they suck because they don't have enough power. I could potentially load them up with $1,000 worth of Lithium based batteries or two tons worth of lead acid batteries but for a robot that I want to follow my cat I am not sure that it is worth it. If I want to build a real robot that will go out in to the real world and do real things then I need batteries. It is one thing to have smooth rolling robots running over a smooth surface and not using much power. But to have an agricultural robot weeding its way through a clumpy muddy farm right after a heavy rain, I need some serious power.

So batteries force robot designers to make many compromises: They can compromise sales by making the robot too expensive, they can compromise how much work it can do by a small battery, they can compromise the computing power to save power, they can compromise functionally to save power.

Of all the problems the one that bothers me the most is compromising computing power; it is very nice to have two or more HD cameras feeding their data to one or more GPUs that crunch what the robot is seeing in real time and plan the optimal solution also in real time. Also other sensors such as radar or laser scanners can be energy gobblers.

For instance I would be curious to find how much Asimo's battery cost, and how long it lasts.

So it is battery technology that is the last piece of the puzzle to adding independent robots to our lives in a substantial way.

When these were first released they were silverlight only. I wanted to watch them but there was a zero percent chance I would use silverlight. It is wonderful that these are now available for all the sensible people who don't drink the microsoft koolaid.

The rule should also be: they are not a cop if it isn't recording. They are just some joe shmoe waving a gun without it. Thus if they arrest or detain some innocent person without it then they are charged with kidnapping, extortion, waving a gun in public, etc.

A frustrating friend of mine who periodically calls me for computer help but will argue with any help I offer got nailed by one of these guys. Except that he argued with them the whole time and wouldn't follow their instructions. The only thing that ended up being changed was that he deleted his browser icon from his desktop.

I had my dog chipped but the vet said that if I ever let my iPhone run out of battery the implant will explode. I think the implant was made by the same company.

The worst thing they can do is to secure it and then depend upon the security working. Thus the system should be designed so that if it is hacked every other Monday that it can survive. There have been a number of recent (last 20 years) events that have shown that single points of failure can have devastating effects. So make sure that if terrible things happen that a lesser grid can be maintained manually.

A great example of this would be a local grocery store chain's SAP system failed shortly before Christmas(some years ago). They were so dependant upon it that their ability to order stuff and manage inventory was pretty much non existent. So the store ended up looking like some kind of soviet grocery store where the only goods on the shelves were pretty much those that are managed by the distributors themselves; things like milk.

This grocery store hopefully has learned from this and now has some kind of manual backup plan where a store manager can actually phone in his orders and crudely manage the store's needs in the case of another serious computer outage.

The same with the grid. Ideally they set some sort of minimal functionality emergency plan whereby humans can crudely manage the system as opposed to a system that either works perfectly by computer or doesn't work at all.

But I worry far less about hackers and far more about system design failures and Carrington events.

Where I live (Nova Scotia) basically doesn't have earthquakes. So the risk here would be Tsunami from a distant earthquake. Interestingly enough if there were a Tsunami the configuration of the seafloor would cause it to be massive and wipe everything out for 10 or more miles inland.

I am not sure how many bottles of water I would need for that scenario.

I guess this is why this fine academic institution has never crossed my radar. I have never heard it mentioned in any publication, any citation, any contest win. I am not saying that they don't publish squat but that nothing they have published managed to catch my attention. And when I read something in Nature, etc I will check to see which institution the various authors are from to mentally compile a list of intellectually active institutions.

So as far as I can tell this place is the intellectual opposite of say, MIT.

I suspect that if most people were faced with the choice of paying for all ad-driven sites would simply not go to most sites. I could live with a only a few sites, StackOverflow being a huge one, a mapping web site, a classified ads site, etc. Do I really need to watch russian drivers crash into each other?

How did you do as compared to say the CFO or the head of marketing(assuming equal time in the company)? I am not saying that techies lose every time but that often when you see a set of technical and business co founders that often the technical founder is gone by the time things go public.