The image changed to black-and-white. I saw a fine probe push what looked like a tiny squid into position on a titanium field. It was a bullet-nosed lump with streaming filaments at the rear. It was a tenth of the size of the red blood cell, which in the vacuum of the scanning electron microscope was a wrinkled oval, like a gray raisin.
“Our camera is one ten-billionth of an inch in length. As you see, it is shaped like a squid,” Julia said. “Imaging takes place in the nose. Microtubules in the tail provide stabilization, like the tail of a kite. But they can also lash actively, and provide locomotion. Jerry, if we can turn the camera to see the nose… Okay, there. Thank you. Now, from the front, you see that indentation in the center? That is the miniature gallium arsenide photon detector, acting as a retina, and the surrounding banded area-sort of like a radial tire-is bioluminescent, and lights the area ahead. Within the nose itself you may be able to just make out a rather complex series of twisted molecules. That is our patented ATP cascade. You can think of it as a primitive brain, which controls the behavior of the camera-very limited behavior, true, but enough for our purposes.”
I heard a hiss of static, and a cough. The screen image opened a small window in the corner, and now showed Fritz Leidermeyer, in Germany. The investor shifted his enormous bulk. “I’m sorry, Ms. Forman. Tell me please where is the lens?”
“There is no lens.”
“How can you have a camera with no lens?”
“I’ll explain that as we go,” she said.
Watching, I said, “It must be a camera obscura.”
“Right,” she said, nodding.
Camera obscura-Latin for “dark room”-was the oldest imaging device known. The Romans had found that if you made a small hole in the wall of a dark room, an upside-down image of the exterior appeared on the opposite wall. That was because light coming through any small aperture was focused, as if by a lens. It was the same principle as a kid’s pinhole camera. It was why ever since Roman times, image-recording devices were called cameras. But in this case-
“What makes the aperture?” I said. “Is there a pinhole?”
“I thought you knew,” she said. “You’re responsible for that part.”
“Me?”
“Yes. Xymos licensed some agent-based algorithms that your team wrote.”
“No, I didn’t know. Which algorithms?”
“To control a particle network.”
“Your cameras are networked? All those little cameras communicate with each other?”
“Yes,” she said. “They’re a swarm, actually.” She was still smiling, amused by my reactions.
“A swarm.” I was thinking it over, trying to understand what she was telling me. Certainly my team had written a number of programs to control swarms of agents. Those programs were modeled on behavior of bees. The programs had many useful characteristics. Because swarms were composed of many agents, the swarm could respond to the environment in a robust way. Faced with new and unexpected conditions, the swarm programs didn’t crash; they just sort of flowed around the obstacles, and kept going.
But our programs worked by creating virtual agents inside the computer. Julia had created real agents in the real world. At first I didn’t see how our programs could be adapted to what she was doing.
“We use them for structure,” she said. “The program makes the swarm structure.” Of course. It was obvious that a single molecular camera was inadequate to register any sort of image. Therefore, the image must be a composite of millions of cameras, operating simultaneously. But the cameras would also have to be arranged in space in some orderly structure, probably a sphere. That was where the programming came in. But that in turn meant that Xymos must be generating the equivalent of-
“You’re making an eye.”
“Kind of. Yes.”
“But where’s the light source?”
“The bioluminescent perimeter.”
“That’s not enough light.”
“It is. Watch.”
Meanwhile, the onscreen Julia was turning smoothly, pointing to the intravenous line behind her. She lifted a syringe out of a nearby ice bucket. The barrel appeared to be filled with water. “This syringe,” she said, “contains approximately twenty million cameras in isotonic saline suspension. At the moment they exist as particles. But once they are injected into the bloodstream, their temperature will increase, and they will soon flock together, and form a meta-shape. Just like a flock of birds forms a V-shape.”
“What kind of a shape?” one of the VCs asked.
“A sphere,” she said. “With a small opening at one end. You might think of it as the equivalent of a blastula in embryology. But in effect the particles form an eye. And the image from that eye will be a composite of millions of photon detectors. Just as the human eye creates an image from its rods and cone cells.”
She turned to a monitor that showed an animation loop, repeated over and over again. The cameras entered the bloodstream as an untidy, disorganized mass, a kind of buzzing cloud within the blood. Immediately the blood flow flattened the cloud into an elongated streak. But within seconds, the streak began to coalesce into a spherical shape. That shape became more defined, until eventually it appeared almost solid.
“If this reminds you of an actual eye, there’s a reason. Here at Xymos we are explicitly imitating organic morphology,” Julia said. “Because we are designing with organic molecules, we are aware that courtesy of millions of years of evolution, the world around us has a stockpile of molecular arrangements that work. So we use them.”
“You don’t want to reinvent the wheel?” someone said.
“Exactly. Or the eyeball.”
She gave a signal, and the flat antenna was lowered until it was just inches above the waiting subject.
“This antenna will power the camera, and pick up the transmitted image,” she said. “The image can of course be digitally stored, intensified, manipulated, or anything else that you might do with digital data. Now, if there are no other questions, we can begin.” She fitted the syringe with a needle, and stuck it into a rubber stopper in the IV line.
“Mark time.”
“Zero point zero.”
“Here we go.”
She pushed the plunger down quickly. “As you see, I’m doing it fast,” she said. “There’s nothing delicate about our procedure. You can’t hurt anything. If the microturbulence generated by the flow through the needle rips the tubules from a few thousand cameras, it doesn’t matter. We have millions more. Plenty to do the job.” She withdrew the needle. “Okay? Generally we have to wait about ten seconds for the shape to form, and then we should begin getting an image… Ah, looks like something is coming now… And here it is.” The scene showed the camera moving forward at considerable speed through what looked like an asteroid field. Except the asteroids were red cells, bouncy purplish bags moving in a clear, slightly yellowish liquid. An occasional much larger white cell shot forward, filled the screen for a moment, then was gone. What I was seeing looked more like a video game than a medical image.
“Julia,” I said, “this is pretty amazing.”
Beside me, Julia snuggled closer and smiled. “I thought you might be impressed.” Onscreen, Julia was saying, “We’ve entered a vein, so the red cells are not oxygenated. Right now our camera is moving toward the heart. You’ll see the vessels enlarging as we move up the venous system… Yes, now we are approaching the heart… You can see the pulsations in the bloodstream that result from the ventricular contractions…” It was true, I could see the camera pause, then move forward, then pause. She had an audio feed of the beating heart. On the table, the subject lay motionless, with the flat antenna just over his body.
“We’re coming to the right atrium, and we should see the mitral valve. We activate the flagella to slow the camera. There the valve is now. We are in the heart.” I saw the red flaps, like a mouth opening and closing, and then the camera shot through, into the ventricle, and out again. “Now we are going to the lungs, where you will see what no one has ever witnessed before. The oxygenation of the cells.”
As I watched, the blood vessel narrowed swiftly, and then the cells plumped up, and popped brilliantly red, one after another. It was extremely quick; in less than a second, they were all red. “The red cells have now been oxygenated,” Julia said, “and we are on our way back to the heart.”
I turned to Julia in the bed. “This is really fantastic stuff,” I said.
But her eyes were closed, and she was breathing gently.
“Julia?”
She was asleep.
Julia had always tended to fall asleep while watching TV. Falling asleep during your own presentation was reasonable enough; after all, she’d already seen it. And it was pretty late. I was tired myself. I decided I could watch the rest of the demo another time. It seemed pretty lengthy for a demo, anyhow. How long had I been watching so far? When I turned to switch off the TV, I looked down at the time code running at the bottom of the image. Numbers were spinning, ticking off hundredths of a second. Other numbers to the left, not spinning. I frowned. One of them was the date. I hadn’t noticed it before, because it was in international format, with the year first, the day, and the month. It read 02.21.09.
September 21.
Yesterday.
She’d recorded this demo yesterday, not today.
I turned off the TV, and turned off the bedside light. I lay down on the pillow and tried to sleep.
DAY 2
9:02 A.M.
We needed skim milk, Toastie-Os, Pop-Tarts, Jell-O, dishwasher detergent-and something else, but I couldn’t read my own writing. I stood in the supermarket aisle at nine o’clock in the morning, puzzling over my notes. A voice said, “Hey, Jack. How’s it hanging?” I looked up to see Ricky Morse, one of the division heads at Xymos. “Hey, Ricky. How are you?” I shook his hand, genuinely glad to see him. I was always glad to see Ricky. Tanned, with blond crewcut hair and a big grin, he could easily be taken for a surfer were it not for his SourceForge 3.1 T-shirt. Ricky was only a few years younger than I was, but he had an air of perpetual youthfulness. I’d given him his first job, right out of college, and he’d rapidly moved into management. With his cheerful personality and upbeat manner, Ricky made an ideal project manager, even though he tended to underplay problems, and give management unrealistic expectations about when a project would be finished. According to Julia, that had sometimes caused trouble at Xymos; Ricky tended to make promises he couldn’t keep. And sometimes he didn’t quite tell the truth. But he was so cheerful and appealing that everyone always forgave him. At least, I always did, when he worked for me. I had become quite fond of him, and thought of him almost as a younger brother. I’d recommended him for his job at Xymos.
Ricky was pushing a shopping cart filled with disposable diapers in big plastic bundles; he had a young baby at home, too. I asked him why he was shopping and not at the office. “Mary’s got the flu, and the maid’s in Guatemala. So I told her I’d pick up some things.”
“I see you’ve got Huggies,” I said. “I always get Pampers, myself.”
“I find Huggies absorb more,” he said. “And Pampers are too tight. They pinch the baby’s leg.”
“But Pampers have a layer that takes moisture away, and keeps the bottom dry,” I said. “I have fewer rashes with Pampers.”
“Whenever I use them, the adhesive tabs tend to pull off. And with a big load, it tends to leak out the leg, which makes extra work for me. I don’t know, I just find Huggies are higher quality.”
A woman glanced at us as she pushed past with her shopping cart. We started to laugh, thinking we must sound like we were in a commercial.
Ricky said loudly, “So hey, how about those Giants?” to the woman’s back as she continued down the aisle.
“Fuckin’ A, are they great or what?” I said, scratching myself. We laughed, then pushed our carts down the aisle together. Ricky said, “Want to know the truth? Mary likes Huggies, and that’s the end of the conversation.”
“I know that one,” I said.
Ricky looked at my cart, and said, “I see you buy organic skim milk…”
“Stop it,” I said. “How are things at the office?”
“You know, they’re pretty damn good,” he said. “The technology’s coming along nicely, if I say so myself. We demoed for the money guys the other day, and it went well.”
“Julia’s doing okay?” I said, as casually as I could.
“Yeah, she’s doing great. Far as I know,” Ricky said.
I glanced at him. Was he suddenly reserved? Was his face set, the muscles controlled? Was he concealing something? I couldn’t tell.
“Actually, I rarely see her,” Ricky said. “She’s not around much these days.”
“I don’t see much of her either,” I said.
“Yeah, she’s spending a lot of time out at the fab complex. That’s where the action is now.” Ricky glanced quickly at me. “You know, because of the new fabrication processes.”
The Xymos fab building had been completed in record time, considering how complex it was. The fabrication building was where they assembled molecules from individual atoms. Sticking the molecule fragments together like Lego blocks. Much of this work was carried out in a vacuum, and required extremely strong magnetic fields. So the fab building had tremendous pump assemblies, and powerful chillers to cool the magnets. But according to Julia, a lot of the technology was specific to that building; nothing like it had ever been built before. I said, “It’s amazing they got the building up so fast.”
“Well, we kept the pressure on. Molecular Dynamics is breathing down our necks. We’ve got our fab up and running, and we’ve got patent applications by the truckload. But those guys at MolDyne and NanoTech can’t be far behind us. A few months. Maybe six months, if we’re lucky.”
“So you’re doing molecular assembly at the plant now?” I said.
“You got it, Jack. Full-bore molecular assembly. We have been for a few weeks now.”
“I didn’t know Julia was interested in that stuff.” With her background in psychology, I’d always regarded Julia as a people person.
“She’s taken a real interest in the technology, I can tell you. Also, they’re doing a lot of programming up there, too,” he said. “You know. Iterative cycles as they refine the manufacturing.”
I nodded. “What kind of programming?” I said.
“Distributed processing. Multi-agent nets. That’s how we keep the individual units coordinated, working together.”
“This is all to make the medical camera?”
“Yes.” He paused. “Among other things.” He glanced at me uneasily, as if he might be breaking his confidentiality agreement.
“You don’t have to say,” I said.
“No, no,” he said quickly. “Jeez, you and I go way back, Jack.” He slapped me on the shoulder. “And you got a spouse in management. I mean, what the hell.” But he still looked uneasy. His face didn’t match his words. And his eyes slid away from me when he said the word “spouse.”
The conversation was coming to an end, and I felt filled with tension, the kind of awkward tension when you think another guy knows something and isn’t telling you-because he’s embarrassed, because he doesn’t know how to put it, because he doesn’t want to get involved, because it’s too dangerous even to mention, because he thinks it’s your job to figure it out for yourself. Especially when it’s something about your wife. Like she’s screwing around. He’s looking at you like you’re the walking wounded, it’s night of the living dead, but he won’t tell you. In my experience, guys never tell other guys when they know something about their wives. But women always tell other women, if they know of a husband’s infidelity.
That’s just how it is.
But I was feeling so tense I wanted to-
“Hey, look at the time,” Ricky said, giving me a big grin. “I’m late, Mary’ll kill me, I’ve got to run. She’s already annoyed because I have to spend the next few days at the fab facility. So I’ll be out of town while the maid’s gone…” He shrugged. “You know how it is.”
“Yeah, I do. Good luck.”
“Hey, man. Take care.”
We shook hands. Murmured another good-bye. Ricky rolled his cart around the corner of the aisle, and was gone.
Sometimes you can’t think about painful things, you can’t make your mind focus on them. Your brain just slips away, no thank you, let’s change the subject. That was happening to me now. I couldn’t think about Julia, so I started thinking about what Ricky had told me about their fabrication plant. And I decided it probably made sense, even though it went against the conventional wisdom about nanotechnology.
There was a long-standing fantasy among nanotechnologists that once somebody figured out how to manufacture at the atomic level, it would be like running the four-minute mile. Everybody would do it, unleashing a flood of wonderful molecular creations rolling off assembly lines all around the world. In a matter of days, human life would be changed by this marvelous new technology. As soon as somebody figured out how to do it.
But of course that would never happen. The very idea was absurd. Because in essence, molecular manufacturing wasn’t so different from computer manufacturing or flow-valve manufacturing or automobile manufacturing or any other kind of manufacturing. It took a while to get it right. In fact, assembling atoms to make a new molecule was closely analogous to compiling a computer program from individual lines of code. And computer code never compiled, the first time out. The programmers always had to go back and fix the lines. And even after it was compiled, a computer program never ever worked right the first time. Or the second time. Or the hundredth time. It had to be debugged, and debugged again, and again. And again.