Containment (29 page)

"Then we'll
all
do it," Zaire said. She looked at Cam.

"Listen carefully," Arik said. "At exactly eleven hundred hours in exactly ten days from now, the three of you are going to suit up and take a rover out to the wall. You're going to leave a locker open for me, and twenty minutes later, I'll meet you out there. We're not going to talk about this again between now and then. We're not going to debate it, and I'm not going to remind you. You just have to do it. All of you. You have to swear to me."

"Even if I wanted to help," Cam said, "there's no way all of us can get out of V1. Zaire and I can get the suits, but there's no way we can get Cadie through the Wrench Pod and into the airlock, and there's no way you'll be able to follow. Things have changed since your accident, Arik. People will stop us."

"There won't be anyone around to stop you," Arik said. "I'll take care of that."

"How?" Cam said.

"By doing the only thing that's guaranteed to bring every single person in V1 together into one place," Arik told them. "I'm going to solve AP."

The maglev stopped at the Life Pod platform.

"Don't let me down," Arik said as he stepped off the train. "I can promise you that we will never get another shot at this."

T
he term "red herring" originated from the practice of distracting hunting dogs from the scent of a fox or a badger with the pungent odor of a cured fish. It later came to be used in the areas of literature, science, and politics to refer to anything that lured attention away from the issue at hand.

Arik had come to believe that the term "artificial photosynthesis" was an innate red herring. Instead of describing a problem, it inherently suggested a solution. The issue at hand wasn't actually how to reverse engineer a photosynthetic plant's metabolic pathways — the real problem was how to generate large amounts of oxygen using readily available elements as quickly, cheaply, and efficiently as possible. But at some defining moment in V1's history, someone had been unable to escape the confines of their own experiences and imagination, and had confused inspiration with implementation. Emulating photosynthesis in an attempt to generate oxygen was exactly like trying to achieve flight by copying birds, and so far, the exercise had yielded just about the same level of results.

To solve AP (clearly a misnomer, but the phrase had become a convention), Arik knew that he had to play by a different set of rules. Plants' ability to produce their own food and expel oxygen as a byproduct had evolved over the course of more than a billion years, starting with the simplest forms of algal scum. Their technique had been judged and refined by what has always been the single constant and only valid measure of success: survival. Millions of years of minute and almost imperceptible genetic mutations had steered hundreds of thousands of species of plants down paths that had proven to be evolutionary dead ends, yet some small percentage of mutations had proven beneficial, compounding and accumulating into unimaginable complexity, specialization, and elegance. To reverse engineer photosynthesis in less than ten days was about as practical as trying to understand the principles of flight by taking apart a heavy Sagan rocket with a screwdriver. But using an electron core computing cloud to model and dramatically accelerate the processes that gave rise to photosynthesis in the first place might just be possible.

Arik's previous attempts to crack AP had consisted of building software models of the tulsi ferns that were cultivated in the dome along with their aeroponic life support systems, but even with the help of a team of extremely competent chemists and biologists, Arik felt like he would never be able to do better than what he perceived as a rough approximation. Computer models had always been rough approximations, and rough approximations had never been good enough. Weather had never been consistently and successfully predicted beyond a percentage of certainty, and models of the universe could only suggest general theories regarding both its origin and its eventual demise. Beyond a certain point in the development of computer processor technology, the problem was no longer a shortage of CPU cycles, but instead reflected humanity's inability to ask their binary counterparts the right questions. For years, electron computers had been sitting relatively idle waiting for their creators to finally pose a challenge worthy of their unimaginable faculties.

Arik understood that the effective use of an electron computer required a realistic and practical division of labor between computer and human. Humans were good at intuiting possibilities while computers excelled at testing those possibilities by the trillions. Therefore, rather than starting with something incredibly complex like modern day photosynthesis and trying to distill it down to something that could be modeled and programmed, Arik decided to start with a small number of simple elements and use the computer to see how they might evolve into more complex results. Rather than reverse engineering photosynthesis, Arik believed that he could
arrive
at photosynthesis — or perhaps something even better — through a process he liked to call
evolutionary engineering
.

Assembling the virtual environment for Arik's experiment was fairly straightforward since he already had most of the software models he needed. Years ago, he had built precise software abstractions of protons, neutrons, and electrons, and written algorithms to model gravitation, electromagnetism, and strong and weak nuclear forces. (Of course, this software had already existed for decades, but Arik only trusted code that he himself had authored.) He then used his low-level nuclear models to assemble higher-level models of all the elements in the periodic table, and then used those models to assemble still more complex molecular models. Since he began working in the Life Pod, Arik had also added to his software library several routines for simulating things like temperature, atmospheric pressure, and all wavelengths of light energy.

Arik had all the software models and most of the algorithms he needed; what he didn't have was an efficient way to combine and test trillions of permutations.

Random mixing and matching wasn't good enough. Arbitrarily combining models of the physical world in a virtual environment might eventually yield some interesting results, but even an electron computer could easily spend months or even years playing such a complex and unconstrained guessing game. Arik needed a much more intelligent approach. He needed an algorithm that knew how to pursue paths that were promising, and swiftly abandon those that were obvious dead ends. It had to understand how to build on success, learn from failure, branch out and pursue multiple possibilities simultaneously, and it needed to understand how to build increasingly complex systems out of proven simpler ones. Arik needed an algorithm intelligent enough to condense over a billion years of evolution into just a few days.

He wondered what it would be like to be able to actually witness and perceive evolution. He imagined entire species coming and going at the speed of soap bubbles forming and popping; ice caps expanding and receding at the rate of a rapid heart beat; continents drifting apart and dispersing like sea foam. Even watching a simple vine grow at as little as ten times its normal speed makes what we perceive as a dumb and static tendril look like the long thin appendage of an intelligent primate skillfully searching out better light and more secure anchorage.

As Arik worked, he gradually began to repartition his life. He no longer thought in term of day and night, and he stopped trying to keep track of mealtimes. He slept when he was exhausted and ate when he was weak. The only schedules he observed were the cycles imposed by pain pills and stimulants. Cadie stopped asking him when or if he would be home, and started bringing him changes of clothes and boxed meals, instead.

In place of his own circadian rhythm, Arik adopted the rhythm of the computer. For reasons he didn't fully understand, his program seemed to reach milestones at predictable intervals. A milestone was defined as the completion of a simulation that yielded no less than a tenth of a percent more oxygen than a previous milestone. Every time a milestone was reached, Arik assembled an experiment in a borosilicate tube and brought it down to the dome to physically validate the results.

Between experiments, Arik tried to understand the formulas that his program was producing. With Cadie's help, he had been able to make sense of the first few milestones, but the output had rapidly grown too complex for either of them to really comprehend. With every milestone, the computer was getting closer to solving AP, and Arik and Cadie were getting farther from understanding how.

The definition of when AP was technically solved was somewhat subjective. Subha had suggested that nothing less than the production of one dioxygen molecule for every molecule of carbon dioxide that went into the process constituted a viable solution. She called it the "One-to-One Rule." Arik and Cadie had verified a 1:1 ratio milestone over forty-eight hours ago, and with time to spare, Arik allowed the computer to continue. If increasing photosynthetic oxygen production by a factor of two didn't guarantee him an audience with everyone in V1, nothing would.

When the computer predictably and unceremoniously achieved the 2:1 milestone, Arik assembled two verification experiments. He and Cadie each carried one down the corridor and into the dome. Both of the absorption disks turned blue within moments of being exposed to sunlight, and the oximeters sealed inside the tubes verified the results of the process. Cadie touched Arik's pale and gaunt face. His pupils were artificially dilated from his regimen of stimulants, and he squinted in the bright natural light. Neither of them spoke, but they shared a weary and almost unbelieving smile as acknowledgment of what they had accomplished.

Arik prepared four more borosilicate tubes. He sent three of them home with Cadie and kept one for himself. Subha was aware that Arik and Cadie had been putting in long hours, and, sensing they must be nearing a discovery, had started extending her workdays, as well. She accepted Arik's connection request almost instantly, and he could hear the anticipation in her voice when she agreed to meet him in the dome.

Before he left his lab, Arik stopped the computer simulation, deleted the program, and permanently erased every piece of data it had produced and recorded.

C
adie and Arik had an early dinner, and probably for the first time since the day they were married, went to bed at the same time and without opening their workspaces. Arik told Cadie that everything had gone well with Subha, and that the assembly would happen on schedule. He knew that Cadie wanted to ask him about the things he'd said on the maglev, but she didn't. They dimmed the wall lights and talked about Haná instead, about what projects they might work on next, about really needing to reconnect with Cam and Zaire, and frankly with each other, as well. And then they just laid beside each other in the warm glow of the walls and didn't talk at all.

Before they turned the lights all the way down, Cadie brought Arik a sedative. She told him he needed to get back on a regular sleep schedule, that she couldn't stand seeing him so exhausted and frail. A few minutes after he swallowed the pill, he started to say something about his lips and nose feeling numb, but was asleep before he could finish the thought.

He was awakened by a feeling of pressure in his abdomen which turned into a severe, stabbing pain when he moved. He'd been asleep for almost eleven hours, and Cadie was gone. He put his feet on the floor and pushed himself up, clutching his stomach with one hand and steadying himself against the bed with the other. His clothes were damp with perspiration and he was shivering. He wiped his nose and saw blood on the back of his hand, and when he got to the bathroom, he found he was bleeding from the inside, as well.

The latency period was over. The cells of his digestive tract were dying and he was shedding dead intestinal tissue. He knew he wouldn't be able to eat again until he was treated, and that the biggest risk over the next 48 hours was dehydration and shock from a water-electrolyte imbalance. Since his body was gradually losing its ability to absorb nutrients, he knew he would eventually need to take in fluids intravenously, but for the next two days, he needed to find ways to keep himself alive with whatever he had.

If he could figure out how much radiation he'd been exposed to, he would have a better idea of what to expect and how he might be able to treat himself, but since he hadn't worn a radiation badge, there was no way to know. At least there was no obvious way. In Arik's experience, there was almost always a way to recover lost information, to calculate a missing piece of data, to derive whatever you didn't know from whatever it was that you did. One of Arik's core beliefs was that the inability to solve most problems was ultimately due to a failure of the imagination.

He could have tested his clothing or a piece of his equipment, but he had sealed everything up in lead-lined pouches and dropped them down the hazardous waste chute in the Wrench Pod. If he could get a sensor outside now, he might be able to estimate what his exposure had been, but going back to the Wrench Pod before ensuring it was empty was far too risky. He thought about trying to get a reading off the robotic rover, but it would have been fully decontaminated in the airlock before being allowed back inside. With Cadie's help, it would probably be possible to estimate his exposure based on the rate at which his cells were dying, but he doubted they had the time to gather enough data to identify a meaningful trend. And the less she knew about his condition, the better.

Searching for a way to extrapolate a missing piece of data was sometimes like searching for your glasses. It often took the exercise of retracing your steps before finally reaching up and finding that they had been perched on your head all along. It suddenly occurred to Arik that Malyshka would have picked up radiation levels as part of her atmospheric analysis during the second part of her mission. Arik had downloaded the data after getting back from the ERP, but before he had a chance to examine the results, he began to experience the onset of radiation sickness. The information he needed had been right in front of him all along.