This is Joe Gregorio's writings (archives), projects and status updates.


Several jobs ago I worked at a company that made RFID readers. It was a fascinating technology and always seemed poised to take off, but in all the years that I worked there, and the time since, RFID tag readers haven't taken off, at least not in the way that everyone in the RFID industry expected. Of course, the company I worked for had revenue plans that incorporated a hockey-stick growth pattern, and that hockey-stick never materialized.

Now don't get me wrong, there are good markets for RFID tags, mostly when line-of-sight isn't achievable, but first let me clarify by what I mean when I say RFID tags. There are the simple anti-theft based systems that are widely deployed today, which are incredibly simple and only detect if a tag is in the field or if it's not in the field. I am not talking about those systems. The RFID tags I'm talking about carry information, such as serial numbers, and can even have new information written to them. That's the group of readers that I worked with, and it's that group that has never really taken off. There were grand plans for how they would allow tracking individual items from manufacturer, through Distribution Centers (DCs), to the market shelf, and finally as you checked out. They were supposed to supplant barcodes. Instead, while they have gained traction in niches, such as for tagging pets and livestock, and tolls, such as the E-ZPass system, they have not achieved ubiquity. They still could, deploying any such new technology takes time, but I don't think RFID tags will ever achieve the ubiquity of the barcode.

As I've thought about this over the years I've concluded that the promise of RFID was eclipsed by another technology out there that's poised to become more and more disruptive, not only to RFID, but to a host of technologies, and that's the CCD.

CCD stands for charged-coupled device. From Wikipedia:

A charge-coupled device (CCD) is a device for the movement of electrical charge, usually from within the device to an area where the charge can be manipulated, for example conversion into a digital value. This is achieved by "shifting" the signals between stages within the device one at a time. Technically, CCDs are implemented as shift registers that move charge between capacitive bins in the device, with the shift allowing for the transfer of charge between bins.

Often the device is integrated with a sensor, such as a photoelectric device to produce the charge that is being read, thus making the CCD a major technology where the conversion of images into a digital signal is required. Although CCDs are not the only technology to allow for light detection, CCDs are widely used in professional, medical, and scientific applications where high-quality image data is required.

So do all your UPS packages have RFID tags in them? No, and the short answer is that ink is cheaper than an RFID tag. The slightly longer answer is that CCD prices have fallen and a 2D barcode called MaxiCode is easy to print on a shipping label and easy to read with a CCD camera.

CCDs are chips and appear to be subject to Moore's Law. They went from being esoteric to being common place in a very short time. Today my children's Nintendo DSi has two of them, one facing them, the other facing out. They're dirt cheap, getting better resolution all the time (or getting smaller), and are being integrated everywhere; just try to buy a phone that doesn't have one. That price curve will continue, making them more and more attractive as a solution.

Popular Photography has an article on What Photography Will Look Like By 2060, which is interesting and covers some of the CCD trends and how they may affect how we take pictures, and what we consider photography.

But photography is not the only place they appear, and nor is it the barcode-like applications that will have the biggest effect.

For example, the Wii remote works by using a CCD, this time it is one that only sees in the infrared range, and coupled with a bar with two infrared LEDs on it that sits on (or below) the television, it allows the Wii system to sense the Wii remotes distance and angle with respect to the TV. This technolgy has been reversed, at least by hobbyists, into an interesting technolgy that tracks the user's head location.

Microsoft seems to have taken that idea one step further with Project Natal.

But that's certainly not the end, and you can imagine more and more places where the CCD will solve problems, or be applied as a solution. For example, signing a credit card receipt is annoying, particularly when I've already slid my card, and even more so when I have to sign on one of those cheesy digital signing tablets. Why not use CCDs with facial recognition technology? In this case the technology can be a lot simpler than general facial recognition since it already knows whose face it is looking for, and the images can be stored for later analysis, and even used to apprehend people using stolen cards.

Of course, I can't go into this area without mentioning Sixth Sense technology, which seems to be the new name for "augmented reality" meets "wearable computers".

Like I mentioned, DSi's have two cameras built in to them, and my new T400 laptop has one. Will my next laptop, in two years time, have two CCD cameras built in, to give it binocular vision? Will I appear in 3D in video chats? Will you appear as 3D in video chats to me because my PC also tracks my head position as I participate? What other ways will cheap and ubiquitous CCDs change our world and how we interact with it?

The ideas come from thinking about what you get when you point a camera at an object.

Point one at my face and I can video conference.

Put them on a car and point them out and you have a backup camera. Buy why restrict it to just backing up? Why isn't the rear-view mirror a full panorama of the environment around the whole car stitched together from a dozen CCD cameras?

That's pointing out from the car, point them at the car and the possibilities are different. Put them next to highways to monitor rushhour traffic. Point them at your license plate and you have either an automatic red-light running ticket writing machine, or a new toll system, where a camera based system that reads license plates could be used instead of the current RFID based solutions.

Put them on your house pointing outwards and you have a security system. Point them into the house and you have a system that turns the lights and HVAC off in rooms that are empty. Think how much better it would be than those motion sensing systems in some meeting rooms today, where the lights switch off in the room and everyone waves their hands in the air like a bunch of drunk pelicans trying to get the lights back.

If I hang one over my kitchen table will it be able to count calories for me? Can I hang one over my desk and not need to buy a scanner? How about one in the bathroom? How much health information could you extract from an image taken every morning? Could it track my weight? Detect signs of depression? Obviously there are security and privacy concerns.

Sure some of these potential applications will require more processing power and much higher resolution cameras, but the resolutions are increasing, ala Moore's Law, and who said the computation needs to be co-located with the camera? Just realize that this is still early days, and that CCDs will continue to shrink, continue to get cheaper, and will continue to show up in more and more places. What are the benefits? What are the challenges? What other technologies will get disrupted with cheap CCDs?

Update: As has been pointed out in the comments, I'm using "CCD" as a place-holder for "digital imaging technology" regardless of the underlying technology, which includes CCDs and CMOS devices.