ubiquitous computing environments (was: <nettime> From Tactical Media to Digital Multitudes)

[Ana Viseu <ana.viseu@utoronto.ca>]

[At a time when Nettimers are engaged in a discussion regarding the 
connections between physical and digital worlds, here goes an article (from 
PCWorld) with a commercial vision for one possible way of interweaving both 
worlds. It is far from the activist 'vision' that Geert and Florian 
propose, but it is one that is rather pervasive within tech development 
circles, and thus one that we should keep attention to. It has been 
emphasized by a recent speech by Bill Gates at Comdex, where he announced 
that Microsoft will start developing "smart personal objects" and thus 
moving from an emphasis on 'personal computers' to one on 'personal 
computing' where the digitality is hidden. (see for instance 
<http://news.bbc.co.uk/2/hi/business/2487787.stm> and 
<http://www.nytimes.com/2002/11/18/technology/18GATE.html>. All the best. 
Ana Viseu]



Future Gear: Tiny Chips, Everywhere

With petite sensors and radio transmitters, every object can have an 
identity and even think for itself.

"Objects are going to be increasingly given budgets and responsibility, so 
they can make their own decisions," says Paul Mackinaw of 
AccentureTechnology Labs. I recently visited Accenture to check out 
prototypes of new products and to meet Mackinaw, who has the dream job of 
imagining new worlds enabled by technology, then trying to build them.

Accenture's vision of the future is called Reality Online. It presumes that 
people, products, and even clumps of dirt can have digital identities that 
share information over networks and use this information to make decisions. 
Among the many applications Accenture imagines are store shelves that know 
when they are out of an item and order more, crates that can tell the 
receiver if they've been dropped during shipment, and dirt that can sense 
if crops are getting too dry and turn on sprinklers.

For Everything, a Chip

Critical to the development of Accenture's omniscient new world is the 
profusion of very tiny, very inexpensive computer chips. Even today, a chip 
smaller than the size of a match head can hold a unique identifying number 
and a tiny radio transmitter. Anyone who uses an electronic security badge 
or card is carrying such a device. When you wave your card by the reader, a 
radio signal energizes the chip, allowing it to transmit a unique ID number 
back to the reader. If your number is in a database of people allowed to 
enter, the door is unlocked.

Small enough to fit in a credit card, the chips are extremely cheap: They 
cost about 17 cents apiece. Accenture envisions putting them on everything 
from car parts to shampoo bottles. And the tags wouldn't have to hold much 
data. The Auto-ID Center, based at the Massachusetts Institute of 
Technology, is proposing a 96-bit key as sufficient to develop enough 
unique identifiers for all the products on the planet. (By comparison, a 
56-bit key could identify every grain of rice on earth, according to 
Accenture's Mackinaw.)

With the transmitter chips in place, a company could track each product as 
it moves through a factory. Shippers would know what items--and how many of 
them--are in every crate or container, and stores would know if they are 
receiving what they've ordered. When customers hit the checkout counter, 
they could simply pass their cart by a radio receiver that would register 
every item and tally the bill.

That's certainly neat, but other applications are downright creepy. In 
Accenture's world, the tags stay on after you buy a product. If you are 
strutting down the street in your fancy new shoes, for example, a passerby 
with a radio receiver-equipped PDA could find out the brand. You could no 
longer pass off your cheap loafers as Prada. (Unless, perhaps, the vendor 
also creates forged ID chips.)

Mackinaw demonstrated this with a table of tagged products--books, a tie, 
and a scarf--and a Handspring Visor with a radio receiver. In his example, 
the Visor checked an onboard database of just a few product ID numbers. But 
in the future, he envisions the PDA linked wirelessly to databases 
containing the universe of ID numbers. With sensors in place, Accenture 
also imagines everyone turning into a salesperson. Got a new purse your 
friend likes? Don't just tell her the brand. Make sure she scans your purse 
so you get a commission from the retailer.

You might foil the system by yanking out the sensors. Accenture also 
envisions a system that would allow you to specify that the IDs on your 
shoes are confidential. But if we're already having trouble getting spam 
filters to work, what are the chances of mastering access privileges to our 
clothing?

What about embedding chips inside people? It's certainly possible, says 
Mackinaw, but he sees their placement in plastic cards as more likely.

The Nth Sense

Putting an ID number on a chip is just the beginning. Researchers are 
already building chips with sensors that can read conditions like 
temperature, vibration, or moisture. In one of the simpler applications, 
car brake pads might have sensors to warn you if they are getting thin. 
They could even communicate with a Web-connected onboard computer that 
schedules a trip to the mechanic.

How would your car find a convenient time for the appointment? If you set 
up the service, the car's computer could access your online calendar. Just 
like a brake pad or a bottle of shampoo, you, too, could be tracked online.

In much larger applications, thousands or millions of sensors strewn about 
would automatically combine into networks. The University of California at 
Berkeley has already shrunk these sensors down to about 12 cubic 
millimeters, which is smaller than Lincoln's head on a penny. The sensors 
could be commercially available in about two years. The ultimate goal is to 
approximate a grain of sand, creating a type of sensor called Smart Dust. 
Mackinaw says we may see such chips in about five years.

These minuscule sensors could be spread over a farm field to measure crop 
conditions and order extra water or fertilizer if needed. They could even 
be mixed in with paint to signal when it starts peeling. Too small to 
contain batteries, they would draw power from heat or sunlight to produce a 
signal just strong enough to reach the next chip. Together, the dust would 
form a peer-to-peer network to aggregate data and pass it on to a receiver.

Reality Offline

In theory, all this can come to pass. We have the technology, or we will 
shortly. In practice, we'll probably see some implementations of it. 
Electronically tracking certain products--especially expensive items like 
TVs or computers--makes sense. And large companies with sufficient budgets 
to hire qualified technicians can make it possible. We already use lasers 
to read UPC symbols, so why not add radio IDs that provide still more 
detailed information?

But I wonder how big a role consumers would want to play. Although most 
people would appreciate a dashboard warning before their brakes become 
dangerous, would many of us agree to grant the brake pads access to our 
online calendars? Would many people even bother to have an online calendar, 
and keep it up to date?

And how likely are customers to use PDAs for tracking products that move 
through the world? Most PDA users engage only its most basic functions, and 
just synchronizing data is a challenge for some. Will those people see 
radio receivers and subscriptions to product-tracking databases as killer 
apps? And even if they do, will all the manufacturers and retailers agree 
on an interlocking database network, delivered as a Web service, to track 
billions, even trillions, of items? I suspect that, for the foreseeable 
future, your Prada secret is safe.


http://www.pcworld.com/features/article/0,aid,106403,00.asp
Sean Captain
PCWorld.com
October 30, 2002


[ - - - - - - - - - - - - - - - - - ]
Tudo vale a pena se a alma não é pequena.
http://fcis.oise.utoronto.ca/~aviseu

http://privacy.openflows.org
[ - - - - - - - - - - - - - - - - - ]

#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net