Slashdot Mirror
The Crazy-Tiny Next Generation of Computers
An anonymous reader writes University of Michigan professors are about to release the design files for a one-cubic-millimeter computer, or mote. They have finally reached a goal set in 1997, when UC Berkeley professor Kristopher Pister coined the term "smart dust" and envisioned computers blanketing the Earth. Such motes are likely to play a key role in the much-ballyhooed Internet of Things. From the article: "When Prabal Dutta accidentally drops a computer, nothing breaks. There’s no crash. The only sound you might hear is a prolonged groan. That’s because these computers are just one cubic millimeter in size, and once they hit the floor, they’re gone. 'We just lose them,' Dutta says. 'It’s worse than jewelry.' To drive the point home, Dutta, an assistant professor of electrical engineering at the University of Michigan, emails me a photo of 50 of these computers. They barely fill a thimble halfway to its brim."
Great. There are some days where I forget where I've put my smartphone. So now I can expect to lose my entire computer because it dropped and I might have vacuumed it up with the dust bunnies?
My sci-fi novel, Ghost Thief, is now available from Amazon.com.
a [drool] beowulf cluster of these!
No Beowulf clusters yet?
-- I have monkeys in my pants.
I read about this a week ago. I was not impressed. Basically a lot of marketing bullshit and no huge breakthroughs.
Strip any small CPU of it's plastic and guess what you have? Well, a tiny silicon die.
They will release the blueprint so that anyone with a $50 million lab can build them? How nice...
And they think these things are going to measure the real energy costs of my house? I have news for you. The energy costs of all houses in the world have probably doubled only because of all the projects to measure these same energy costs.
Sorry, but no.
Dust on the desktop! Oh wait, I already have that.
/shuffles around....
Solving Unix problems since 1989...
That article is all about the miniaturization process they went through. Wake me up when the hardware specs are available: CPU speed, amount of RAM, wireless connectivity and range, etc.
I have serious doubts that these things will become popular anytime soon (if ever), especially if their per-unit cost is more than a few cents. Their size, coupled with the "if you lose sight of it, consider it lost forever" joke (read: warning), makes them seem impractical.
They should scale it back up to the size of that quarter.
You knew there was one.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
That's *one* IoT... but how does that relate to my lightbulbs that track me around the house or my garage door opener that lets me open it remotely from my Apple Watch after seeing who's standing outside?
The IoT is about networking commodity hardware and aggregating telemetry and sensor data remotely. For some reason, it seems to have significant overlap with Cloud Computing such that we really have a CloT with access control nightmares.
Funny thing is, vending machines were on the Internet almost 20 years ago. This was useful for the parent's illustration (service tech knows what to restock and when, and if the machine's out of service / bil cartridge is full / etc). But we didn't call it the IoT back then; just the Internet. That was part of the original vision, before .com got involved and morphed it into some sort of a "display your web browser banner here" place.
In other words, the IoT is closer to the original concept of the Internet than what most people have thought of as "the Internet" for the past decade or so. A bunch of internetworked hardware talking to each other and to humans, all around the world.
Apparently they charged him $80 an hour to fix his washing machine.
APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
My point exactly (the last one)... making the devices respond to signals, and making the concentration point "in the cloud" means that people hacking into your home computer is a thing of the past -- all they have to do is get your Apple/FaceBook/Google ID, and suddenly they've got access/control for every device you own.
Vacuum cleaner won't be chasing you, but your lights will be tracking you and your power meter might just send an extra few amps to your digital doorknob just as you go to open it....
Point. Said more generally, does IoT mean that the most common failure will be some malfunction in the "I" part of the device? That more complexity inevitably leads to more points of failure?
Will this be a pattern similar to that followed by CFLs? Early IoT devices will be buggy, but the bugs will be ironed out, followed by a short Golden Age, where the prices have fallen and the devices essentially last forever, followed by the inevitable Value Engineering, after which things fail randomly and often, with error modes never seen in non-IoT devices?
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
Also, Stanislaw Lem featured a kind of smart dust in his 1980s novel.
Escher was the first MC and Giger invented the HR department.
We just lose them
This will be great for security. /sarcasm
That's SCADA or any of a large number of remote-access monitoring systems (many running over IP).
IoT is not over the Internet. It's always (for those I've seen selling it) a private network of things. NoT. And that's what you should think of it. When they start pushing for actual open connections to the things (everyone has 1M IPv6 addresses at their house, and every door knob, appliance and widget in the house has a unique static IP that the owner (or anyone else) can connect to), then it'll be an Internet of things.
Right now, it's a closed network of things. What you describe is "remote monitoring". Nothing more, nothing less. I've seen IoT used when describing batched video downloads over closed WiFi from fleet vehicles to a private server that's not connected to the Internet in any way. IoT, like "cloud" has no useful definition or meaning.
Learn to love Alaska
The concept of smart dust is much older than Pister and the 1990's. Stanislaw Lem used the idea already in the early 1960's in his stories.
That's *one* IoT... but how does that relate to my lightbulbs that track me around the house or my garage door opener that lets me open it remotely from my Apple Watch after seeing who's standing outside?
Well, as I keep saying, the IoT is not really about whether you fridge or garage door are on the internet; these are just gimmicks to entertain you and lure you into thinking that it is 'cool' and therefore somehow OK. And I'm not sure there is all that much intent to spy on people, in most cases - it is more that these devices are becoming easy and obscenely cheap to produce, and it is very easy to persuade yourself to thinking "what's the harm?" in incorporating them into all kinds of every day objects - paper documents 'for security', wrapping 'to track goods throughout production', etc etc. In many cases they are meant to be no more than a "better barcode", and there is no malice behind; but computers being so much more than just passive markers means that they can be used for a host of things that they were never intended for, and that is the big worry, in my opinion. It is certainly something we have to apply some thought to - it is technically possible to produce mote computers that include capabilities like networking, microphone, possibly camera and other environment sensors. There are scenarios in which these things may be beneficial, but the potential for abuse is also great.
This is the year of the Linux Dust Top!
And just to annoy everyone reading my article, I didn't even bother to include that particular photo in it.
Get free satoshi (Bitcoin) and Dogecoins
The mote localizers were in A Deepness In The Sky. That was in 99.