FeralRobots.net

The other day I happened to be looking for the exact text of a recurring phrase from Ubik. I stumbled across a Metafilter thread about some P. K. Dick speeches and essays that very quickly settled into the usual dichotomy of Dick discussions: On the one side, "he was a crappy stylist who wrote the same story over and over again, and besides he was insane"; on the other "he was a visionary mystic who looked at the same issue a myriad of ways, and besides, what does 'insane' really mean, anyway?"

Recurring again and again was this notion that Dick was not an "idea writer": Not someone who thinks up new cool gadgets or really extrapolates where things are going. Visionary, yes, but not about society or technology.

I personally think people who dismiss Dick as a writer are missing out. But I also think that people who focus on him as a mystic are seriously missing out. I used to look at the Technovelgy newsfeed as often as I could find the time -- it's a great place to trace back new technology to the treatment SF writers gave to it when they first described it, usually 50 or more years ago. And one thing that I noticed quickly was that Phillip K. Dick stories seemed to be cited more than those of any other writer.*

Here are a few examples, off the top of my head and with some help from Technovelgy:

• Ubicomp (Ubik in a particularly annoying form, but all over the place in his work).
• Smart-missiles (The Zap Gun, and many other places).
• "Electric" pets (Do Androids Dream of Electric Sheep?).
• Personalized newspapers ("homeopapes") (Ubik and elsewhere).
• Insect-sized spy robots/cyborgs (Lies, Inc. and others).
• Insect-sized advertising robots/cyborgs (The Simulacra).
• Customized presentation of advertising in general (all over the place, it was one of his memes)
• Doors, weapons, etc. that are keyed to "cephalic patterns": They recognize you by recognizing the thing about you that is most uniquely you, your thoughts. (Lots of stories.)

Some themes emerge: Most notably for me is that he re-uses things relentlessly. It could be to save effort; more likely, it's because he's still exploring the idea and wants to put it into different situations. Note how he uses the fly as a spy in one setting, and as an advertising delivery vector (a meme-infection vector?) in another. Those aren't just different uses, they're diametrically opposed in one aspect; seeing them both here forces you to understand what they have in common: The idea of sneaking something up on you, whether it be for the purpose of theft (the spy-fly) or infection (the ad-fly).

So for my money, Dick is probably one of the better idea-writers to have worked in SF. I don't see this as separate from his mysticism (or paranoia). Some of these ideas (like the concept of being recognized by something and presented with a pitch that's specifically tailored to you) seem like attempts by technology to mimic a god -- or perhaps (and you have to think this would enter Dick's head, knowing Dick) intrusions of God into the machine. (In terms of his later thought, it might be the Demiurge revealing himself in moments of everyday life.)

__
*Please, don't go and count, this is just an impression, not science.

One of the nightmare climate change scenarios involves the sudden release of methane due to the decomposistion of methane hydrate beds in ocean sediments at northern latitudes. Methane is a very potent greenhouse gas, and there's some geological evidence that large-scale methane hydrate decomposition played a significant role in at least one previous warming event.

Well, it looks like it's happening.

"250 plumes" makes it sound big (and true enough, that's just what they saw, and that in only a small part of the ocean and a short period of time with limited equipment), but it's not that big, yet, really. And most of the gas is dissolving into the water before it gets to the service. So it's not such a big problem, right?

Well:

1. It's just the beginning. Who knows what happens as the sea ice degrades further and the ocean warms still more.
2. When methane dissolves into seawater, it makes the seawater more acid, which reduces the ocean's ability to buffer CO₂.

This may be one of those moments you normally need hindsight to see.

China Miéville on why crime novels end badly. This has particular resonance for me, because both my I and my wife are in-progress on crime novels. One reason is that...

... crime novels are not what they say they are. They are not, for a start, realist novels. Holmes’s intoxicating and ludicrous taxonomies derived from scuffs on a walking stick are not acts of ratiocination but of bravura magical thinking. (Not that they, or other ‘deductions’, are necessarily ‘illogical’, or don’t make sense of the evidence, but that they precisely do so: they make it into sense. The sense follows the detection, in these stories, not, whatever the claim, vice versa.) The various manly Virgils who appear ex nihilo to escort Marlowe through his oneiric purgatories are not characters, but eloquent opacities in man-shape: much more interesting. Dalgliesh’s irresistibility to hyperrealised moral panics du jour – the poor man manages to contract SARS – is an elegiac opera of Holland Park angst, rather than any quotidian gazette of a policeman’s unhappy lot. Detective fiction is a fiction of dreams. Not only is this no bad thing, it is precisely what makes it so indispensable.

Of course this is even more true of speculative fiction as it is of crime novels (and of course Miéville knows this, in this context that's obviously part of his point).

Miéville goes on to say something else that's very interesting:

Secondly, detective novels are not novels of detection, still less of revelation, still less of solution. Those are all necessary, but not only are they insufficient, but they are in certain ways regrettable. These are novels of potentiality. Quantum narratives. Their power isn’t in their final acts, but in the profusion of superpositions before them, the could-bes, what-ifs and never-knows. Until that final chapter, each of those is as real and true as all the others, jostling realities all dreamed up by the crime, none trapped in vulgar facticity. That’s why the most important sentence in a murder mystery isn’t the one starting ‘The murderer is…’ – which no matter how necessary and fabulously executed is an act of unspeakable narrative winnowing -  but is the snarled expostulation halfway through: ‘Everyone’s a suspect.’ Quite. When all those suspects become one certainty, it’s a collapse, and a let-down. How can it not be? We’ve been banished from an Eden of oscillation.

And yet, and yet, the crime novels that end without a solution -- that let us stay in our 'Eden of oscillation' -- are almost universally reviled. As are the ones that mess too much with the conventions. I wonder if we feel a need to retreat from candyland; or perhaps it's the driving force of the deeper quest narrative that Miéville alludes to above.

When I first encountered Sherlock Holmes as a child, I took lovers of Holmes at their word and took it seriously as 'scientific detection.' But even at the age of 10 or 11 I could see how ridiculous it was at that level. All you have to do to turn it into parody is change the name of the protagonist to, say, Herlock Sholmes, and the very idea of somone publishing monographs on the taxonomy of cigar ash instantly acquires enormous comic potential.

Technorati Tags: China Miéville, crime fiction

Theatre director Natasha Williams on theatrical interpretation:

"The idea is that everybody dies," she says. "The playwright has to die in the director. And then actors have to die in the play because they have to let go their ego to create a world that is a world of its own. And then of course [the] play goes into [the] audience's mind, and everyone understands it their own way.

"So, you know, it's a food chain. It's a theatrical food chain."

... then the riddle would be easy to answer. Transhumanism and particularly extropianism have long seemed to me to be mostly about technophilia. About love of the product: Maybe even loving it so much that you want to make yourself more like it. In one sense, it's of a piece with gun-nuttery: You can have my Blackberry* when you pry it from my cold, dead, extra-jointed thumb. 

On the other hand, I've always felt that people might be a little too attached to the present, so I find it hard to quarrel with designer/artist Marcia Nolte:

If we look back into the history of evolution, we see that our body adapts to changing circumstances. Today we see that these circumstances often adapt to our body. In this case the design is usually reacting on the individual needs and less on surviving.

Well, she's got a point. For better or worse, we're driving our own evolution, now, and while I'm really skeptical about our ability to make design decisions that don't bite us in the ass a hundred (or even fifty) years from now, letting the changes accrue randomly as an aggregation of consumer designs makes even less sense.

I'm not really sure what Nolte is trying to do with these images. As quietly disturbing as they are (and that ill-ease springs as much from the gestalt of the photos as from the strangeness of the body modifications), I suspect that they are meant to provoke thought. So let's think about it. 

What does it mean to alter ourselves to suit our products? Arguably we've done that, as part of a long, slow process that was initiated when we picked up a stone and threw it at prey or predator. But it was a truly slow process, and there wasn't really anything conscious about it. If you looked hard enough, you could probably find examples of selective breeding for desirable traits -- it should be easy to find them in slave populations, but you could almost certainly find them in general populations, as well. It would happen just as surely through social constraints on marriage suitability as it would through selective breeding among slaves.

What we have not had the capacity to do before now -- what we will soon have the capacity to do -- is to change ourselves to suit our environment. In its most extreme forms, this kind of extropianism (or is it transhumanism?) forces us to contemplate where the boundaries of humanity lie. James Blish pushed the edges of this envelope in his 'pantropy' stories, which were driven by a seeding metaphor: We "normal" humans "seed the stars" with beings more suited to the places we land. Is it a largely arborial world? Make them like monkeys. Aquatic? Seals. No large scale life yet? Make them near-microscopic.

The gauntlet hasn't often been thrown as forcefully since. But the extreme cases can make us think it's an easier problem than it really is. Body mods like extra joints in the thumb are trivial compared to actually changing the way the brain works, for example, and that's the big missing link, from my perspective. For me, the key human thing about humans has always been how our brains work. (I'm not a human-exceptionalist: This is also, to me, the key fox-thing about foxes or bat-thing about bats.) One of the things that extropians miss is that changing the brain in even small ways is liable to fundamentally change our human-ness.

Most people don't really have any grasp of what a homeostatic balancing act the brain-mind really is. Tweak the serotonin a little bit, for example, and you can turn a stable and solid person into a basket case in a heartbeat. So imagine if we were to throw off the timing by adding more efficient nanotech-driven neurons, for example. We might throw the recipient into a massive feedback loop that crashed their brain.

Some people are going to get seriously messed-up by all this. I suppose evolution requires casualties.

[via Dezeen]

--
*Gotta be a Blackberry. Extropianism is too hard-core for an iPhone. Anyway, if he were modified for an iPhone, the tip of her model's thumb would be really, really tiny.

Vernor Vinge on cheating at the Turing Test:

As with past computer progress, the achievement of some goals will lead to interesting disputes and insights. Consider two of [Rodney] Brooks's challenges: manual dexterity at the level of a 6-year-old child and object-recognition capability at the level of a 2-year-old. Both tasks would be much easier if objects in the environment possessed sensors and effectors and could communicate. For example, the target of a robot's hand could provide location and orientation data, even URLs for specialized manipulation libraries. Where the target has effectors as well as sensors, it could cooperate in the solution of kinematics issues. By the standards of today, such a distributed solution would clearly be cheating. But embedded microprocessors are increasingly widespread. Their coordinated presence may become the assumed environment. In fact, such coordination is much like relationships that have evolved between living things. [link added]

I don't think Vinge goes nearly far enough. I've thought for at least twenty years that our standards for machine sentience (or any non-human sentience, for that matter) were hopelessly anthropomorphic. For most purposes, "Artificial Intelligence" should have been called "Artificial Human-Like Intelligence."

I lack sympathy for the idea that we have only ourselves to judge it by. People have been dealing with beings of different intelligence for as long as there have been human beings. (We call them "animals.") I'd like to think we've figured out something by now -- or at least figured out how to figure it out -- about how to judge intelligence without provoking the shades of Alan Turing and G. B. Shaw to peals of ghostly laughter.

To Vinge's immediate point: If we require of the machine that it be like us, we are placing an irrelevant constraint on our evaluation of its intelligence. Imagine if machines were evaluating us -- think how short we'd fall in their estimation with our slow electro-chemical communications network, the lossy construction of our visual apparatus. And how slowly we caluclate!

Technorati Tags: turing-test, pygmalion

The Advent of Malicious Circuits | Beyond the Beyond from Wired.com

(((There's a new one. You bake the malware right into the hardware, then release the hardware into the wild. With keystroke loggers, spamware and trojans built right into the chip itself, you're home-free against software-based detection.)))

(((How do you get victims to buy your subverted chips, though? That one seems pretty obvious: product forgery. Sell 'em your China-based Appl3 H-phone. If the price is right, they'll go for it -- and with the fraud money you'd pull down from a scheme like this, you could give the hardware away -- even pay fools to take it.)))

Many moons ago, I started working on a novel about a brainwashed super-agent. (Who bears an eerie and entirely accidental resemblance to Jason Bourne, of whom I was blissfully ignorant at the time. Not that he was entirely original; just that he was more like a mashup of Piers Anthony's "agents" and Graham Greene's "Professor D." But that's beside the point.) He would have been hoodwinked into joining, then trained and brainwashed and turned into a psychotic robot to be sent out on dangerous missions, then brought back and brainwashed again for the next job. (That's the part I cribbed from Anthony.) My first sketches on that idea date to the summer of 1980. They're pretty bad. In summer and fall of '80, I worked through two drafts of a novella that outlined that character and situated him relative to the great technical bureaucracy that I imagined him serving, and ultimately defying, like a cancer cell.

The character stuck in my brain and I started fleshing out the parts of the idea that had to do with how you would organize a great invisible intelligence enterprise. I'd created a pretty coherent vision of how the whole system worked and started sketching some much better stuff as long ago as 1984; I'd settled on the idea that his "master" was a sentient but profoundly alien AI as early as about 1988 or so; by early 1992, I'd worked out how the system-monster communicated without being noticed, by hiding its traffic as noise packets on the Internet; more sophisticated messages could be "book-coded" in Usenet messages. By 1995, while working at Kodak, I merged that nightmare vision with another, based loosely on the legend of Volund/Wayland Smith, and concocted a grand, long-range story of a conflict between two new sentiences, once accidental, the other planned, and neither seeming very human. The driver on the "planned" side was a small but powerful firmware vendor called TranScalar Systems, who designed chips for communications applications. Their chips were in everythign, and spyware was in all their chips. That gave their sentient monitoring system absolute control over dat streams (as long as the government-owned monster didn't realize it was there).

I'd done almost all of this while being more or less completely ignorant of cyberpunk. I didn't read any Gibson until about 1999, no Sterling until '98. So on the one hand, I reinvented some wheels. On the other, I had some ideas that I now know never really got much traction. The Sprawl trilogy, for example, is thick with deeply alien AIs, but that vision never caught on -- the moder post-cyberpunk transhumanist AI is cloyingly human, as a rule.

Since then, one by one, most of the stuff I dreamed up has hit the mainstream. In 1998's The Saint, Simon uses a Usenet-based "book code" to trade messages regarding his contracts. Jason Bourne, who'd been there all along, of course, entered my consciousness in the early oughts. And now I learn that the idea of hardware-embedding malware is finally making the mainstream. I'm a visionary without visible portfolio. It's my own damn fault for not writing it earlier, of course.

From Usenix:

Abstract

Hidden malicious circuits provide an attacker with a stealthy attack vector. As they occupy a layer below the entire software stack, malicious circuits can bypass traditional defensive techniques. Yet current work on trojan circuits considers only simple attacks against the hardware itself, and straightforward defenses. More complex designs that attack the software are unexplored, as are the countermeasures an attacker may take to bypass proposed defenses.

We present the design and implementation of Illinois Malicious Processors (IMPs). There is a substantial design space in malicious circuitry; we show that an attacker, rather than designing one speci?c attack, can instead design hardware to support attacks. Such ?exible hardware allows powerful, general purpose attacks, while remaining surprisingly low in the amount of additional hardware.

We show two such hardware designs, and implement them in a real system. Further, we show three powerful attacks using this hardware, including a login backdoor that gives an attacker complete and highlevel access to the machine. This login attack requires only 1341 additional gates: gates that can be used for other attacks as well. Malicious processors are more practical, more ?exible, and harder to detect than an initial analysis would suggest.

1 Introduction

1.1 Motivation

Attackers may be able to insert covertly circuitry into integrated circuits (ICs) used in today’s computerbased systems; a recent Department of Defense report [16] identi?es several current trends that contribute to this threat.

First, it has become economically infeasible to procure high performance ICs other than through commercial suppliers. Second, these commercial suppliers are increasingly moving the design, manufacturing, and testing stages of IC production to a diverse set of countries, making securing the IC supply chain infeasible. (((Uh-oh.))) Together, commercialofftheshelf (COTS) procurement and global production lead to an “enormous and increasing” opportunity for attack [16].

Maliciously modi?ed devices are already a reality. In 2006, Apple shipped iPods infected with the RavMonE virus [4].

....

Using modi?ed hardware provides attackers with a fundamental advantage compared to softwarebased attacks. Due to the lower level of control offered, attackers can more easily avoid detection and prevention. The recent SubVirt project shows how to use virtualmachine monitors to gain control over the operating system (OS) [11].

This lower level of control makes defending against the attack far more dif?cult, as the attacker has control over all of the software stack above. There is no layer below the hardware, thus giving such an attack a fundamental advantage over the defense.

Although some initial work has been done on this problem in the security community, our understanding of malicious circuits is limited.

IBM developed a “trojan circuit” to steal encryption keys [3]. By selectively disabling portions of an encryption circuit they cause the encryption key to be leaked. This is the best example of an attack implemented in hardware that we are aware of (...)

Indeed, a single hardcoded attack in hardware greatly understates the power of malicious circuitry. This style of attack is an attack designed in hardware; nobody has designed hardware to support attacks. The design space of malicious circuitry is unexplored, outside of simple, specialpurpose, hardcoded attacks. Responding to the threat of trojan circuits requires considering a variety of possible malicious designs; further, it requires anticipating and considering the attacker’s countermoves against our defenses. Without such consideration, we remain open to attack by malicious circuits....

Paolo Bacigalupi, interviewed at Wired Science, on getting unsolicited feedback from someone who can't be ignored:

Harlan Ellison called me up out of the blue.  It was soon after the short story had come out and I was in my house mopping the floor and I get this phone call and this man on the other end was like 'This is Harlan Ellison, do you know who I am?' and I was like 'Yeah, yeah, um yeah.' So he says, 'Go get your story.'  So I do. He then proceeds to basically critique every single aspect of my entire story.

He starts out by saying 'At first I thought that you were some sort of professional writing under a pseudonym because, you know, nobody has a name like Bacigalupi, I know the Abbot and Costello routine blah blah blah...' He goes off about how Paolo Bacigalupi is obviously a pseudonym or a joke name of some sort. Now he's getting a bit worked up. He says, 'You know, I thought you were a professional, and then I got to page 5 and right down there at the bottom you used the word jerked... and then 2 sentences later you used the word jerky--you took all of the power out of the fucking word!'

I'm sitting there on the line sort of terrified of this man just haranguing me. At the end of that whole conversation - a conversation in which he critiques, line by line, my entire story - he finishes up by saying, 'Well you got some potential, but don't write in genre, it's a waste of time. Don't get stuck in it like I got stuck in it.' And then he hangs up.

That was the last thing that I heard from this guy--I don't know what it was--sort of like a love tap I guess, but I actually sort of got to me. I proceeded to write a bunch of stories that weren't science fiction. I wrote historical fiction novels set in China, I went on and wrote a landscape... I don't know what you call it... sort of landscape porn I guess is the best word for it.  You know, one of those love of place and the rural west sort of stories. Then I wrote a mystery/western story and none of those genres is related to sci-fi in any way, shape or form, and none of them sold.

At the end of all of that, I'm sitting there with all of the rejection letters in my hands and thinking: Well, you know, actually I kind of liked writing science fiction and then I went back into it and started doing the short stories, and that's when I started writing things like "The Fluted Girl," and "The People of Sand and Slag" and started finding my niche. It's been a long process.

A bit from a talk last year by Vinge -- 3pointD.com » Blog Archive » Vernor Vinge Paints the Future at AGC:

I am convinced that the day we really get high resolution heads up displays, most people who nowadays are carrying a bluetooth earphone and microphone would have no problem with wearing eyeglasses that gave them a heads up display of something like 4,000 by 4,000 if the infrastructure had moved along in concert. Then high resolution HUDs could be exploited. ...

I'm unclear on what the infrastructure issue is, here. Bandwidth? That's only a problem if you assume one is transmitting bitmaps at full res in real time. I don't think that's likely, at least at first. Look to the way that humans transfer this kind of information via the extremely efficient lossy compression scheme we call "language".

.... That’s an example of a highly disruptive technology. It essentially destroys all other display technology except as emergency backups.

If you were able to get localization that was really good, you could imagine setting this up so that if your wearbale knew where you were looking, what the orientation of your head was and where your eyeballs were tracking, then in addition to being able to produce the world’s best display, as good as the worlds’ best desktop display, you could actually overlay things in the environment.

There could be some interesting localization artifacts, here, as different localization schemes amplified one another's errors -- or just introduced strangenesses. The avatar floating six degrees to the left or right of the speaker, for example. A person's modified nose floating in the air above his/her head. Again, this is an argument for a vector-based system, offloading the localization to the local system: Let my eyes localize the stuff, don't make the broader system do it.

The term for that in academic circles is augmented reality. In that situation, having the processing power that’s involved with the network infrastructure I just described becomes very very useful, because you could in an ad hoc way overlay those portions of reality that you wanted to.

In an auditorium like this you could make the walls look like whatever you wanted, you could make the speaker look like a clown, and since everything was networked, you and your friends could get together and agree on what things looked like. The notion of consensual imaging becomes very very important, and again this is actually a very disruptive technology, if it were finally to happen. It blows away all discussion of large three-dimensional display technologies.

This is the really, really fun part -- in a project I'm working on right now, I call this "digging". A technology's adoption phase has phases of its own, and the one that's most interesting to me right now are what I could call the primary and secondary negotiation phases. The primary phase is the part where the bleeding edge early adopters figure out amongst themselves what consistutes and appropriate use of the tech. The secondary phase happens while the tech is going mainstream, and involves complex interactions between the bleeding and leading edges and the critical-mass bulk right behind the leading edge.

It's really that critical mass bulk that will drive everything. I don't take it as given that as goes the bleeding edge, so goes the leading -- or as goes the leading edge, so goes the critical mass. These are different populations, and I think even a trivial reading of how trends develop is liable to show that the leading edges don't determine how the body of the wave flows, or even predict it, so much as they give clues. To understand those clues, you need to understand the composition of the bleeding/leading edge communities, and how those communities relate to the critical mass adopters.

Here's how I'm thinking the adoption phases are likely to shake out with regard to consensus avatars. In the primary negotiation phase, digging (i.e., hacking someone's consensus avatar) would be regarded as a form of vandalism; in the secondary negotiation phase, it would be regarded as 'play' -- at least, that would be the socially acceptable way to regard it. Anyone who persisted in regarding it as vandalism would be be regarded as old-fashioned, behind the curve. You'd have to buy in (at least superficially) to keep up.

Naturally this is just the tip of the iceberg. We would still have to figure out how augmented "reality" would play out for ordinary things -- things that are (superficially) not as charged as presentation of self.

In a TED talk, Stewart Brand pointed out that all over the world, poor villages — the same villages that Jeffrey Sachs seems to want to preserve — are vanishing. The people who lived in them have moved to squatter cities, where, according to Brand, there is zero unemployment and a much better life. Because Jeffrey Sachs’ interest in poor African villages seems to be recent, I am not surprised that he may end up on the wrong side of the helped/didn’t help ledger.

Seth Roberts @ Scientific Blogging | The world's best scientists. The internet's smartest readers

There is "zero unemployment" in Brand's squatter cities because those who do not (or cannot) work, die. It's really got nothing to do with innovation or with economic growth or opportunity. It's got to do with it being a fundamentally unforgiving environment.

He may be right about the demise of experts (at least, if we forget for a moment that Brand et al are setting themselves up as really nothing more than alternative experts). But I don't really think squatter cities have anything much of value to tell us about it one way or another, especially if what we're relying on is the strange argument that they're wonderful places without human problems.

The germs of truth in Brand's arguments (yes, the countryside is depopulating, yes, people in squatter cities are continually innovating in response to the highly challenging survival environment) obscures a deeper truth: People will engage in endlessly inventive rationalizations to justify their activities.

Squatter cities are, more often than not, squalid places that are rife with disease, where the oppressions of tradition are replaced with oppression by the strong/clever/zealous/amoral. Better? Worse? And by what (and whose) criteria?

Are they also rife with innovation? Sure; it's necessary to survive in that kind of environment. Do people experience joy, happiness, wonder, and live full and rewarding lives there? Absolutely; people will tend to make a world where they can do that, wherever they live. This reactionary defense of "squatter cities", though, smacks of free-marketism at its silliest: That is good which provokes the most change. Amen.