“This is called traffic analysis, and it’s surprisingly powerful. In this animation, for example, none of us had access to the actual text of the German messages emanating from this part of the front. Yet, I’m pretty sure we all got the picture that a battle happened, that the Germans won the battle, and that they advanced to new locations on the map.”
The web of lines grew brighter and the underlying geography faded into darkness until they were all just looking at a black wall reticulated with a complex pattern of glowing lines.
“How does this apply to the subject matter of this conference? Well, the message traffic within the Process has vastly higher bandwidth than anything that existed in World War Two. But we have computers that are vastly more capable of analyzing it.”
The black background grew brighter, turning pale gray and taking on some complexity and definition. The pattern of bright lines was still the same, but now it spanned a different geography: the undulating folds of a human brain. New lines flashed into existence and old ones winked out.
“Where does that actually get us? Well, that’s a pretty deep question about consciousness and epistemology, as some of our colleagues at this and previous MonsterCons—excuse me, ACTANSSes—have shared with us. But I’ll give you a quick, glib answer: it gets us nowhere. Imagine, for the sake of argument, that we had a way to wiretap every single axon within a living human brain, so that we could make a record of every firing of every single nerve cell.”
The animation zoomed in deeply until a single nerve cell filled most of the screen. Leading away from it was its main output channel: an axon. Clamped onto that was a cartoon alligator clip with a wire running away to a teletype. Whenever the nerve cell fired, a pulse of light ran down the axon, a signal shot up the wire, and the teletype chunked out a line of data.
The animation then zoomed back out, slowly at first, revealing that every single nerve cell in the brain had its own alligator clip and its own teletype. Then the zoom accelerated until the whole brain was depicted on the screen. Below it was a window showing all of the lines of data coming out of all the teletypes at once; the information was zooming by so quickly that it was just a blur.
“We’d have a hell of a lot of data. But would we really know what the brain was thinking? Would we know that it was seeing red, or doing arithmetic, or feeling sad? Despite some recent advances in pattern recognition on neural networks, the answer is basically no. But we can still perform traffic analysis and try to draw conclusions about what sorts of activities are going on within the Process. Just as the Allies were able to sift through mountains of gibberish to get an idea of what the Germans were doing and where, we may be able to analyze what we know about Process-related message traffic to make some guesses about what the Process is doing. There are various ways of attacking that problem. Matilda is going to talk about one of them. She’s drawing on some of the foundational mathematical techniques of physics to detect what appears to be spatial thinking inside of the Process. Matilda?”
Dr. Matilda Napolitano stepped forward and waited for a buzz to die down. Many of the attendees weren’t quite sure they’d heard Sophia correctly. Spatial thinking? Even the word “thinking” was controversial here, and usually deployed in scare quotes, as many skeptics doubted whether the Process’s activities could be classified as such. To talk about a particular type of thinking was a bold move.
Matilda was forty-five, stylishly dressed, and a little nervous. Her English, though accented, was perfect—she’d spent half of her life in the big English universities before landing a prestigious appointment at Turin.
“In order to do physics,” she began, “we have to deal with space and time. I’m here to talk about space. In the old days, we took the existence of space as a given and we didn’t do a lot of hard thinking about its structure. What space actually was. How it behaved. By the time Einstein came on the scene, the groundwork had already been laid by Minkowski and Lorentz for a reappraisal of the fundamental nature of space, and new mathematical notations and techniques were in place, ready for him to apply them to the problem of curvature of space-time.” As she spoke, still images of Newton and the others appeared on the screen, interspersed with pictures of geometrical proofs from Euclid, mathematical formulas, and ending with Einstein, superimposed on a graphic of a black hole bending space like a ball bearing on a rubber sheet.
“I’m not here to give you all a modern physics seminar,” Matilda said, “but my point is that we do have ways of representing space now, and thinking about it with mathematics. Starting with an idea that is so simple that most of you won’t even consider it to be an idea. Namely that each point in space has something to do with the other points that are nearby to it, but progressively less to do with points that are farther away from it.” She backed this up with an arrestingly simple image consisting of a piece of graph paper, with dots marked in ink where some of the lines intersected. “If we oversimplify the situation by likening space to a sheet of graph paper, then each point is connected immediately to four neighbors, north, south, east, and west of it. A little bit farther away are the corner points in the northwest, the southeast, and so on.” The image was slowly zooming back to show more and more of the graph-paper universe. “Let’s imagine we have a creature that is moving around on the graph paper.” A cartoon penguin appeared at the center point. “It goes north, then northwest, then back east a little bit, and so on.” The penguin had gone into movement, waddling around on the graph paper just as she described it. “One moment it is here, the next moment it is there, and so on. This is all so obvious that we don’t even think about it. But what if we scramble the points?”
In the animation, the points went into motion and moved all over the place, each going to a new position, like cards in a deck being shuffled.
“We have encrypted the map—lost track of which point is supposed to go where. Disaster! First the penguin is here.” The penguin popped up in the upper right-hand corner. “Then suddenly it teleports to this other random-seeming point.” It showed up on the far left edge. “Then it’s here.” It jumped to the central region of the graph. “We can’t make sense of it right away. We need to gather data—to perform traffic analysis. And if we do this enough, we can see trends. We see clear evidence that this point, and this point, must be very close to each other—perhaps, directly adjacent—because they seem to have a lot to do with each other; when one lights up, the other tends to light up at the same time, or a little before or after. So we can begin to make sense of this overall pattern of traffic as if it were a three-dimensional manifold. Whereas other geometries don’t seem to make so much sense. And when these patterns persist we can become very sure that we are seeing something real.”
“Matilda, are you doing this with all of the data coming out of the Process?” someone asked.
“Not all. No,” she said. “The older data—from the first two or so years that the Process was running—can be understood quite clearly as traffic on a neural network. We don’t know what it’s ‘thinking,’” she hastened to add, making quotes with her fingers, “but we can see that the patterns of traffic are what we would expect from a neural network.”
Sophia broke in. “Yes, we started in on this line of research after the New Allocations grew really large.”
“New Allocations” was the admittedly vague term for a phenomenon that had started to become evident late in the first year of the Process and that had been growing exponentially since then. The Process had started with sufficient memory to store its own neural network. At first, it had seemed content with that. But then it had begun to requisition more and more additional resources from the systems to which it had access.
“NMA or NPA?” someone asked.
“M,” Sophia said. For the Process had allocated two kinds of resources for itself as time went on: first memory (hence “NMA,” or “New Memory Allocations”) and later processing power (hence “NPA”). These tended to leapfrog each other; the Process wou
ld suddenly grab a lot of memory and grow into it for a while, then demand a new plateau of processing power.
“We’ve known for a while that the use of memory in the NMA doesn’t follow the same pattern as that in the original Process,” another questioner pointed out. “If I may be permitted to anthropomorphize, the Process is using the NMA for some purpose other than simulating its own neural net.”
“Agreed. And that’s precisely what motivated this research,” Sophia responded. “We asked ourselves, if the memory in the NMA isn’t organized—isn’t used—like neural network memory, then just how is it used? What does it look like?”
“And our answer,” Matilda said, “is that it is used in a spacelike way. It is being used to keep track of a fictional or virtual space with permanent characteristics.”
“Like graph paper?” someone asked. Their tone was not so much skeptical as confused.
“Like a spatial manifold. For which graph paper is a metaphor,” Matilda said. “Because we have access to a vast amount of data, we can say much more about the structure of this space. About what is in it. Sophia and I have brought you all here to show you the results. If you would now please all back up, making a clear area in the middle of this space, I will turn on the spatial simulation so that you can see it.”
The attendees somewhat noisily went into motion, picking their bags up off the floor and shuffling away from one another, clearing a space that grew larger and larger until they were standing around the edges of the court in a ragged oval. Sophia and Matilda had remained closer to the center, standing near midcourt. Matilda’s hands moved in front of her as she worked with some sort of virtual user interface that was only visible to her.
A graph-paper pattern appeared superimposed on the floor. “Just a test pattern,” Matilda said. “Now, the point cloud.”
In one instant, several million motes of green light appeared in the space above the tennis court.
Even though the points were all the same color, and not connected to each other, they were immediately recognizable to the eye as together representing a landscape. In some places, mostly around the edges, the green points were on the floor. In others, in the interior, they were perhaps waist high on a typical person’s body. So the general sense—once everyone had had a few minutes to walk around and view it from different angles—was that they were looking at a three-dimensional terrain map of an island or a continent, bounded all around by the sea, with high land in the middle.
“Ireland,” someone declared.
“Much too mountainous!” someone else scoffed. “Just possibly the Isle of Man? It has a mountain in the middle.”
“This is far larger than that, if you look at all the rivers, the mountain ranges,” said a third. “I’m going to say the North Island of New Zealand? Hard to tell unless we get a ladder and look straight down on it.”
“This is no place that exists on Earth,” said a man speaking in a mild tone of voice but with a calm authority that brooked no argument. Heads turned toward a bald-headed, shaggy-bearded man, who looked to be in his sixties. Through most of ACTANSS 3 he had made a habit of sitting quietly in the back of the room, sans wearable, sometimes paying close attention and sometimes humming to himself as his mind apparently wandered. Through most of the current presentation he had shown a kind of mildly amused boredom, as if finding the subject matter too infantile for words, but when the landscape had appeared in the middle of the room he had been galvanized, and had practically knocked one person over while striding through the point cloud to focus his attention on an interesting feature.
“Pluto’s correct,” Sophia announced. “This doesn’t match against any known—”
“There’s no way that it could. That is obvious by inspection,” Pluto announced.
Sophia turned to glance over her shoulder at the bemused Matilda and winked at her as if to say, I told you so! Then she said to Pluto, “Would you like to explain what you mean by that?”
“It’s impressionistic. Not a physics-driven map. The alluvial formations are all wrong. These mountain valleys are V-shaped rather than U-shaped, as they ought to be—no understanding of how glaciation shapes them. The mountains are just high places in the landscape—they have no history. No exposed sedimentary layers, no evidence of volcanism.” Pluto snorted. “This is programmer art. It reminds me of the first maps of T’Rain that Dodge sketched, when he was trying to recruit me.”
“That’s a compelling statement,” Sophia said.
Pluto realized that everyone was listening to him and clammed up. He became interested in a particular river mouth and began tracing it upstream. The room was nearly silent for several minutes as the attendees feasted their eyes on the new continent. Small clusters formed to mutter and point at interesting features. The green points winked and shifted. “Are they moving?” someone asked. “Or is it just me?”
“This is a live display,” Matilda confirmed. “Small shifts are to be expected as new data come in. But its overall shape has changed very little in the past few months.”
Some attendees waded in to view features in the interior of the continent. Those people looked as if their bodies had been truncated below knee or thigh level. But viewed from too close, the points just looked like an amorphous, green Milky Way. “This is the highest resolution we could attain with the time and the processing power we had,” Sophia remarked. “So it’s like a pointillist painting. Stand back and it’s fine.”
An unfamiliar voice, electronically distorted, sounded from somewhere near the middle of the continent. Heads turned toward Elmo Shepherd’s Metatron. The voice, relayed from the other side of the world, had emerged from the speaker array mounted to its blank face. “What is this?” he asked.
Sophia responded, “The bright patch in the middle?”
“Yes.”
“It’s bright because there is a much higher concentration of points in that zone.”
“I can see that.”
“That’s because we have a hundred times as much data from that patch as from anywhere else,” Sophia explained. “It is where almost all of the computational activity is focused.”
“Can it be zoomed in, then?”
“We expected someone might ask,” Sophia said. “I suggest you all close your eyes for a few moments, or put your wearables up on your head. The display is going to change and some may feel disoriented.” She nodded at Matilda.
Matilda waited for a count of three and then began to manipulate the user interface again. Those who had ignored Sophia’s warning were treated to a sudden, dizzying zoom as the tiny bright patch in the center of the display became much larger, banishing most of the outlying points to the outer darkness. When the display settled down, everyone spent a few moments taking in the new vista.
It was a cloud of green points, just as before, but these depicted a different landscape: a hill rising from a flat plain. Atop the hill was a structure made of straight lines and vertical walls, seemingly man-made. On the plain below was a sparkling grid pattern, reminiscent of a cityscape viewed from an airplane window at night: clearly gridded, with neat boxes rising from the grid.
“I could not be more confused as to what we are looking at right now,” announced Gloria Waterhouse. She was a philanthropist, connected with her family’s foundation, nontechnical, and always the first to stand up and request commonsense explanations when the discourse became too academic. She’d assigned herself that role and she did it well. “We’ve seen what looks like an imaginary island, with beaches and rivers and mountains, I guess. Now this is a small part of the island with a town on it. I get that. But what the hell is it? It looks like it came from a computer graphics experiment four, five decades ago when everything was primitive.”
“This configuration of points is the only way to make sense of the data emerging from the New Allocations,” Matilda said.
“And it’s consistent?” Gloria asked.
“This has been built up over months,” Sophia answer
ed. “It changes a little from moment to moment, but it’s generally stable.”
“What are the red sparks?” asked the Metatron. Here and there, the green point cloud was flecked with momentary red sparks, like tiny lightning bolts.
“We use red as a debug tool to mark new data coming in,” Sophia explained. “Real-time stuff.”
“Then perhaps there is some error in the model,” said the Metatron, and pointed with its open hand to the black void above the town grid. A cloud of red sparks was suspended there, a bit like a fireworks burst, no spark enduring for longer than a fraction of a second. But the cloud as a whole persisted, and as they watched, it began to move through space toward the large structure on the top of the hill.
30
Presently Egdod grew weary of being looked at and flew back to the Palace. This was still empty, which struck him as being somewhat wrong, but he had seen little need to make improvements. The souls of Town, which, compared to him, had so little power to change things, had put objects in their houses of which he knew the names: chair, table. Ward had done likewise in his Gatehouse. Just to add some variety, Egdod now shaped the adamant of the floor up into a chair, and experimented with sitting down on it. As his body never grew tired, this was no different from standing up or flying except insofar as it added some variety. He made another chair where some other soul might sit. Since the other souls were smaller, he made the chair smaller too. Between the two chairs he raised up a table. It was all without any real point and it was not as satisfying as cultivating plants in his Garden or building mountains, deserts, and swamps. So he soon quit and went out to the Garden to see what new plants had sprouted. Some had appeared with leaves that were colors other than green, and ofttimes those were arranged in clusters, which were called flowers. Their purpose, other than adding variety and pleasing whoever gazed upon them, was not clear to him yet. But in the many years since Egdod had first become conscious and extracted his soul from chaos, he had seen countless things that had no clear purpose. Each time he did, he was confirmed in his belief that all of the things he was making had been familiar to him in the place he must have abided when he had been alive. He must have lived in a place that was shaped like Town. Other souls must have lived there too. His world must have contained leaves, trees, and flowers. Season had followed season, and water had beat upon rocks. The emergence of these things was not an act of creating new out of nothing, but a kind of slow remembrance. The souls building chairs in their houses in Town were likewise trying to remember things that they had once known.
Fall; or, Dodge in Hell Page 39