Hold the R'FID beans…

…first up, I know the post title is an awful pun on refried beans… but hey, if a boy can’t use an awful pun on a Friday, when can he..?

Anyway, to the point; the whole RFID tag on supermarket items thing – are people barking up the wrong tree?

Some background…

I wrote a post here which covers this in more detail, but for a while now there’s been a lot of stuff around talking about how life would be wonderful if everything sold on a supermarket’s shelves came embedded with a little RFID tag…


…so that you could do all sorts of handy things with it.  When you took things home and kept them in your house, an RFID reader in your fridge or cupboards would know what you’ve got at home, what you need for certain recipes, what you need to pick up on the way etc etc etc.

As I say, read more about it here if you wanna.  It’s all very handy, all very exciting…

…but it ain’t here in any significant way yet.

And the reality, as this state of the market report from ReadWriteWeb says, is currently quite a way off, for various reasons; the cost of implementation, getting all suppliers to sign up, efficiency of technology and so on.

Whilst we wait for the business world to sort something out with better tech… will we see ordinary people just bodge and hack a version that works for them?

Say what you see…

Tim O’Reilly and John Battelle recently wrote a white paper called Web Squared: Web 2.0 five years on (found via ReadWriteWeb again).  In it, they talked about how RFID might not the way we extract ‘identity data’ from real world objects:

“A bottle of wine on your supermarket shelf (or any other object) needn’t have an RFID tag to join the Internet of Things, it simply needs you to take a picture of its label. Your mobile phone, image recognition, search, and the sentient web will do the rest. We don’t have to wait until each item in the supermarket has a unique machine-readable ID. Instead, we can make do with bar codes, tags on photos, and other “hacks” that are simply ways of brute-forcing identity out of reality.”

So, for example, think about the way QR codes worked… they used cameras hooked to computers (your mobile, most likely) to turn an image into data:


So, when you saw a bottle of pepsi with a QR code on it, you could use your phone to capture the data from the code, access the content etc etc etc…


Think what’s going on though… all the camera is doing is recognising shapes, colours, patterns.

And can of Pepsi Max is, visually, shapes, colours and patterns too… it’s more complex, admittedly, but with the advancements in augmented reality (more great examples of that here) rapidly improving all the time, cameras will be able to recognise packaging as it comes off the shelf.


Then, if we can teach a computer to recognise packaging… can we teach it to recognise the visual signs of how much of the product we have left?

For instance…


I know this bottle of Pepsi is finished.  You know it’s finished too.  I wonder how far we are away from teaching a computer to realise that it’s finished, or half empty, or has been open for a week so is probably flat…

Because then what you might do is just stick the equivalent of a mobile phone camera in the inside of a fridge door, and when you need to have a look in the fridge… well, you just bring up the iPhone app that connects to the fridge camera and determines what’s in there.


Which as my missus will tell you, is going to be a lot more efficient and reliable than sending me to look in the fridge… I can never find anything…

Anyway, the upshot of all this is that by the time RFID gets around to being implemented by creating physical ways of making it feasible and reliable, there may be a non-physical way of capturing and interpreting the same data.

‘Twas ever thus, eh..?