Dr. Dobb's is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.


Channels ▼
RSS

Things That Go Bump in the Night


October, 2004: Things That Go Bump in the Night

Michael is editor-at-large for DDJ. He can be contacted at mikeswaine.com.


Verisimilitude: The appearance or semblance of truth.

Verisimilitude is never achieved; the goal is just to get close enough—whether you're running for election or making a movie with nonhuman characters or writing software with a humanized interface. In the movies, when they show something that has human characteristics but is not human, it's usually either a cartoon animal or a scary monster: either cute or creepy. Or both.

In software, the effect can be similar.

I got to thinking about movie monsters the other day when the kitchen staff at Summer Jo's, my partner's restaurant, got into a discussion of scary monster flicks. "What's the scariest movie you ever saw?" Kate asked me. This was before I saw Fahrenheit 9/11, so I mumbled something about the Alien series with its wonderfully creepy H.R. Giger creatures.

But later I came up with another particularly memorable "creature." My all-time favorite science fiction screen villain has to be the robotic device called "Rover" that was regularly featured in the Orwellian (or was it Kafkaesque?) 1960s British TV show "The Prisoner," the brainchild of Patrick McGoohan.

Rover represented the enforcement power of the ubiquitous, anonymous, all-powerful, happy-faced authority that the McGoohan character fought against. Rover was this big, underinflated white beachball that would suddenly appear and bounce silently toward some terrified victim, plopping on him and suffocating him.

Strictly in the visual sense, it was a little like that scene of Michael Moore pursuing congressmen in Fahrenheit 9/11.

I always thought Rover looked wonderfully nonmechanical, not because you can't make a ball bounce via internal mechanisms, which of course you can, but because Rover's motion was so fluid and lifelike that it didn't seem consistent with the kind of mechanical innards that you'd imagine would be required.

Rover was both cute and creepy.

I have learned, in my constant quest to search out for you the answers to questions you didn't ask, that it was quite a challenge to work the magic that made Rover look so nonmechanical in its movements, given the state of 1960s special effects. Make that 1960s television special effects. No, make that 1960s British television special effects.

Rover was usually dragged around by more-or-less invisible wires attached to a fishing pole, or to McGoohan's shoe heels when Rover was to chase him. Sometimes they shot Rover while pulling it backwards, then the film was reversed to get the effect of forward movement with no wires out in front of Rover messing up the scene—and maybe tripping actors. This meant that any actors in the scene had to walk backwards in such a way that their movements looked natural when the film was reversed.

The particular squishy bounce that characterized Rover's movement, which I found both cute and creepy, was due to its being filled with a combination of air, helium, and water. It would hit the ground with an ominous squishiness, then bounce off with a delightful buoyancy.

The software designer inevitably aspires to a kind of humanness in user interaction, even when there is no intention to create some pseudohuman "agent." If the program is to interact with a human, it has to do so in a way that fits with human reaction times, memory limitations, weakness in dealing with logic but ability to deal with ambiguity, and emotional reactions to all sorts of situations. The better you do in addressing these user interface issues, the more humanlike your user interface will seem, even if you had no intention to imitate humans.

The same aspiration would seem to apply in robotic design—but the scope is broader. Robots don't just swap ASCII test streams with a single user; they interact with the whole physical world around them. This suggests that roboticists have a huge realm to explore in the shaping of robotic behavior—and huge opportunities to get it disastrously wrong.

(No, I didn't just see the movie I, Robot—but I have read the book.)

The Zorbing Menace

A new kind of robot that Japanese researchers have developed shows a clever way to implement robotic movement that looks to me very Roverlike.

The robots that Shinichi Hirai and Yuuta Sugiyama have built at Ritsumeikan University in Kusatsu are not beachballs, but wheels.

What's unusual about these robots is that they move by changing shape as their spokes are alternately heated and cooled. An electric current passes through the spoke near the bottom of the wheel, the spoke shrinks a little, and the wheel deforms in shape. This causes it to become unstable, and it moves in the direction of the shortened spoke. The current moves smoothly from spoke to spoke and the wheel continues to roll.

The rims are made of an elastic polymer and the spokes are of a shape memory alloy, a material that changes shape when heated and springs back to its old shape when cooled. The wheels move in a distinctive way that seems to me to be close to the way the "Prisoner" beachball moved—or a real beachball, for that matter. It's not easy to control: Inertia keeps it from changing direction quickly. But these little robots can scale 20 degree slopes and jump five times their diameter straight up.

Those who've seen the wheels say they could serve as the basis for new a kind of all-terrain vehicle. But I say, give the robot wheel two additional degrees of freedom in motion and remote wireless control and wrap a beachball around it and you've got yourself a Rover.

If you've ever seen "The Prisoner," you might not think that was such a great idea. Because of the difficulty of controlling them, there is a real danger of people being run down by such robots. Admittedly, that might not be a major danger if they restrict the things to the current 4-centimeter-diameter size.

And anyway, you can get run down by a giant beachball while watching zorbing. You do know about zorbing, right? The New Zealand sport, or activity, in which you strap yourself into a transparent ball and roll down hillsides? Zorbing seems to be the answer to the question, "What do you do to maintain the level of excitement in your life after you've been an extra in The Lord of the Rings?"

New Scientist magazine (July 3, 2004) has a story on the wheelbots and more information can be found at http://www.ritsumei .ac.jp/se/~hirai/research/softrobot-e.html. The web site shows stills from a video of a wheel rolling, hill climbing, and jumping. My source of "Prisoner" trivia is The Prisoner by Dave Rogers (Barnes & Noble, 1990; ISBN 1566191637). And an explanation of zorbing and a video of a rash documentary maker (not Michael Moore) being run over while making a video of zorbing can be found at http://www.zorb.com/.

The Grey Goo Menace

The grey goo menace is no longer a menace, it seems.

You know the grey goo theory: Out-of-control, self-replicating nanoscale robots consume all life on earth as they build more of themselves by ingesting all the organic raw material around them. Like Michael Moore in a donut shop.

It was nanotech pioneer Eric Drexler who unleashed this meme, although he never took the idea as seriously as some of those who picked up the meme and ran with it. Now he has decided that it's time to put the grey goo menace meme to bed. There is no grey goo menace, he says. He's sorry he ever coined the term.

Others have challenged the possibility of the grey goo scenario before now, setting off a lively discussion of the pros and cons of robotoxic ecophagy. Drexler's argument for rejecting the grey goo menace goes something like this:

Yes, early notions about molecular manufacturing involved huge numbers of tiny robots called "assemblers" that would self-replicate and work together on large building projects, like social insects. And it was reasonable to anticipate the kinds of problems that insects cause us.

But that was then and this is now, and nanotech is no longer focusing on creating little self replicators. Self replication turns out not to be a particularly efficient way to build products. Nobody, Drexler says, is going to develop these self-replicating nanobots because, for any practical purpose, you'd be better served by a factory of dumb robotic arms (as in limbs, not weapons). He says that the current model of nanotech manufacturing is "Ralph Merkle's convergent assembly architecture for a factory that joins small parts to form larger ones, starting with nanoscale blocks." He says that the nanodevices in such a factory would be similar to conveyor belts and assembly robots in conventional factories today. "If you pulled one out, it would be as inert as a light bulb pulled from its socket."

Factory robots like this don't go rogue and threaten humanity.

So he says. It happens that my brother is a factory robot troubleshooter. He's told me horror stories of dumb robot arms running amok and flinging mechanical parts around the factory. Robots have already broken all three of Asimov's Laws of Robotics. Robots have killed people. But that's not threatening humanity, that's just whacking a few innocent bystanders.

So forget about grey goo, Drexler says. But before you get a chance to heave that sigh of relief, he gives the story this unnerving twist: It's important to put the false grey goo threat behind us in order to fret about the immediate and real threats of nanotechnology. Economic upheaval, environmental catastrophe, an escalating arms race, nanoterrorism, suppression of basic rights in the face of real or imagined threats, and other more or less credible menaces of nanotech gone wrong are envisioned at the Center for Responsible Nanotechnology (http://www.crnano.org/).

You can read Drexler's argument in "Safe Exponential Manufacturing," actually by Chris Phoenix and K. Eric Drexler (http://www.iop.org/EJ/abstract/0957-4484/15/8/001/). The book that popularized nanotech and unleashed the grey goo meme is Drexler's Engines of Creation (Anchor Books, 1986, 1990; ISBN 0385199732).

Fantastic Voyage

The word "nanotech" is everywhere these days, but finding actual nanotech news involves sorting through a lot of announcements of developments and products that are nano in name only. I try to track this stuff, but Jeffrey Harrow not only tracks it, he shares what he learns regularly at http://www.theharrowgroup.com/.

Jeffrey sees a lot of promise in nanotech, but in this month's column, I'm focusing on menaces, perceived or real, so I'll let you read what Jeffrey has to say about such encouraging topics as how nanotech might achieve the Holy Grail of giving us RAM that is simultaneously nonvolatile, fast, and cheap.

I suppose the movie Fantastic Voyage can't be called a scary flick, but I always found the idea of tiny people boating up my veins kind of creepy. Even if one of them is Raquel Welch. Back in those days, Raquel Welch was enough of a threat to heterosexual male blood pressure anyway, without contemplating her making intimate contact with one's personal platelets. I find equally creepy the nanotech scenario of tiny cell-repairing robots doing the same, a scenario that many advocates of nanotech have taken very seriously.

Now it's beginning to look like this nanoscenario will also fall to the dumb-is-better rule. Researchers at Rice University are in fact injecting living subjects (rodents, not people) with medical nanostuff, but these nanothings are not smart (or sexy) robotic plumbers or robotic surgeons that unclog arteries or repair cells. They're just subcell-size inert spheres of silica and gold.

The spheres are small enough to leak through the characteristically weak walls of blood vessels in cancer cells, invading the tumor's cell tissue. The balance of silica and gold can be tuned to make the nanospheres especially sensitive to near-infrared radiation. Then you just hit the target area of the body with near-infrared radiation and microwave the tumors out of existence. No chemo, no dangerous radiation, and 100-percent effective. All the usual cautions about preliminary results apply, but what a breakthrough this will be if it's as good as it sounds. The researcher is Naomi Halas and you can check out her work at http://www.ece.rice.edu/~halas/.

The Quantum Horror

Quantum computing is not just a new computing technology, but a fundamentally different model of computing. Right down at the level of information representation, the difference is profound. Where a conventional register of n bits can hold any one of 2**n values, superposition allows a quantum register of n qbits (quantum bits) to hold all 2**n values simultaneously. What this means is hard to wrap one's mind around, but it's definitely a new paradigm.

The reason that some find quantum computing scary is one particular application of QC: quantum cryptography. Because quantum computing can, in theory, solve intractable mathematical problems like factoring large numbers in manageable time, practical quantum crypto is expected to make current public-key crypto obsolete. The uncrackable becomes crackable.

Alternatively, new quantum crypto algorithms can be developed that are far tougher than current algorithms to crack. And the bad guys get crypto that is really, truly uncrackable. Also a menace.

And it's not just theory: Quantum crypto systems have already been built and put on the market. They're a little expensive, but they exist. Quantum computing is coming. It may not yet be ready for your desktop, but you can afford the next best thing: a quantum computer simulator.

The Fraunhofer Institute of Computer Architecture and Software Technology (http://www.qc.fraunhofer.de/) has put a 31-qbit quantum computer simulator up on the Web. You can try it out right now. (But you only get to use 27 of the qbits.)

Since the simulator is not a real quantum computer, you won't get any of the predicted advantages of quantum computing, but you can try out quantum algorithms. And maybe get some valuable skills for that next job in quantum programming.

The iPod Threat

The greatest threat to humanity today, though, isn't uncrackable quantum cryptography or the imminent crackability of all current cryptography or tiny Raquel Welches in your veins or rolling, bouncing, or self-reproducing robots, but the Apple iPod.

At least it's a threat to corporate data security, or so The Gartner Group thinks. Or do they groupthink at the Group? Anyway, they've made the shocking discovery that iPods can be used to store and transport data. Could employees walk out the door with data on their iPods? Yes. Should companies ban iPods from the workplace to prevent this? Yes, says the Group.

By this logic, all sorts of portable devices should be banned, including memory sticks. Hard to police this policy, I'd say. To be fair, the Group is not proposing that companies frisk employees at the door for memory sticks and iPods, but just that they ban such devices from the company network. Not a crazy idea, just a hard one to enforce, as are all attempts to bottle up information.

I hope the report doesn't stigmatize iPod use in the workplace because, and I say this on the basis of no research whatsoever, if your workspace is a cubicle, an iPod with earbuds is a productivity enhancer.

Unrelated to any of the above, here's a little puzzle. What obsolete machinery is suggested by the following letters?

ISO and the URL

At the risk of sending you off on the wrong track, I'll suggest that you think about locomotives, focusing on the rail sound in south Ireland. I'll publish the answer in a future column after you've had a chance to mull it over. Members of my immediate family and a certain Apple employee in Hawaii are not eligible; I won't mention any names, but you know who you are, Mike Morton. And while I won't post the answer at http://www.swaine.com/, I will put links there related to this month's topics.

DDJ


Related Reading


More Insights






Currently we allow the following HTML tags in comments:

Single tags

These tags can be used alone and don't need an ending tag.

<br> Defines a single line break

<hr> Defines a horizontal line

Matching tags

These require an ending tag - e.g. <i>italic text</i>

<a> Defines an anchor

<b> Defines bold text

<big> Defines big text

<blockquote> Defines a long quotation

<caption> Defines a table caption

<cite> Defines a citation

<code> Defines computer code text

<em> Defines emphasized text

<fieldset> Defines a border around elements in a form

<h1> This is heading 1

<h2> This is heading 2

<h3> This is heading 3

<h4> This is heading 4

<h5> This is heading 5

<h6> This is heading 6

<i> Defines italic text

<p> Defines a paragraph

<pre> Defines preformatted text

<q> Defines a short quotation

<samp> Defines sample computer code text

<small> Defines small text

<span> Defines a section in a document

<s> Defines strikethrough text

<strike> Defines strikethrough text

<strong> Defines strong text

<sub> Defines subscripted text

<sup> Defines superscripted text

<u> Defines underlined text

Dr. Dobb's encourages readers to engage in spirited, healthy debate, including taking us to task. However, Dr. Dobb's moderates all comments posted to our site, and reserves the right to modify or remove any content that it determines to be derogatory, offensive, inflammatory, vulgar, irrelevant/off-topic, racist or obvious marketing or spam. Dr. Dobb's further reserves the right to disable the profile of any commenter participating in said activities.

 
Disqus Tips To upload an avatar photo, first complete your Disqus profile. | View the list of supported HTML tags you can use to style comments. | Please read our commenting policy.