InterstellarNet: Origins (29 page)

Among the secrets of being an effective regulator, as important as uncovering facts in the first place, was knowing when and how to reveal knowledge. Those skills carried over to investigative reporting. “You’re that optimistic about the proteomic startup you’re backing.”

O’Malley’s smile wavered. “I’m unaware of any such startup backed by Martian Biosciences.”

A nondenial. “Let me help. The executive team invested personally in an Earth-based startup, Protein Sciences.”

“Oh,
that
. You did your homework, I see. I personally feel good about Protein Sciences.”

Let’s see, Gil thought, if I can jar that composure some more. “I read a lot of corporate compensation reports. You know what’s interesting? Impressively large bonuses—I’m reminded of the tree in which we’re sitting—awarded just before you and your colleagues bankrolled Protein Sciences.” And just after your firm lost the Moby biotech auction.

“Merely a coincidence.” O’Malley’s smirk clung tenaciously in place, but his delivery became brittle. “An opportunity arose at a time when a few of us had some money to invest.”

Disclose knowledge drop by drop. Gretch called the practice accountant’s water torture. “You know what would worry me, Aaron? That the board of directors would consider your big investment in a next-generation biotech company a conflict of interest. Or that the SEC might.”

O’Malley swigged his overpriced wine. Nervous? “The board and the officers are a team.”

“I would say so.” Gil had his own smile going, hidden on the inside. “Of course, such amity is easy to come by when the founders remain the biggest shareholders. They vote themselves to the board. As board members, they appoint themselves corporate officers.”

The waiter swooped in with two new glasses.

“Are you implying something, Mr. Matthews?”

Implying, hell. “What a poor guest that would make me.” Without missing a beat Gil segued to some entirely innocuous quarterly sales forecasts.

■□■

Petabytes of information. Quadrillions of processor cycles of analysis. It could all be distilled into four words. Earth was a mess.

The human concept of empathy having no analogue in a Home collective mind, Aareehl was confronted by a fundamental challenge. How, in these unforeseen circumstances, could it best sell its wares? It had to reassess the value of its inventory.

Among Earth’s biggest challenges, Aareehl concluded, with so much rebuilding to be done, was finding raw materials. Asteroid mining had provided cheap metals for so long that Earth lacked a native mining industry. Aareehl’s species, forever bound to one world, had developed extraction methods far better than anything the humans used.

But which ores, in what quantities, did Earth retain? How accessible were they? What would it cost to reinvent a planetary mining industry? Aareehl wanted answers to set its price.

Indexing forty-five years of recovered memories; discovering and analyzing Earth’s vastly changed situation; processing the ongoing communications from Home, still unaware of Earth’s trauma …. The computational load was too great. The agent subcontracted questions about mining and minerals to native AI librarians and analysts across Earth’s infosphere.

It needed to ponder the market potential of a biotech breakthrough newly received from Home. Dennis Feulner seemed to think that had potential.

3

The dart flew in a flat arc, striking near the bull’s-eye with a solid
thunk
. Two equally well-placed throws followed.

“All in the wrist.” Harald Olafson, despite his many protests about mandatory IBC overtime, found plenty of opportunities to haunt his favorite pub.

“Another round,” Gil ruefully told the bartender. The rue was for effect, as was Gil’s trailing score. “I yield to the master.”

“The new parts,” Harald began cautiously. He chugged from the latest in a series of foamy steins to appear beside his elbow. “You expressed curiosity. I wonder if a sample would be informative.”

Gil could be nonchalant, too. “Well, you know, I like to give my readers insight into the companies that I watch.” He did not condone theft, exactly, but, “Things fall off trucks all the time. If something should come my way….”

“Of course, you’d take a look. You’d just be doing your job, buddy.”

A hand slipped in and out of Gil’s coat pocket. Something small and cylindrical remained behind.

Harald took another long swig. “I just can’t imagine how something like that could get lost. Seems as unlikely as you winning a round.”

“A toast to the unlikely, then.” Gil raised and drained his own glass. “Two more pints,” he ordered.

■□■

The most efficient dialogues on Earth occurred AI to AI. Information was exchanged on a purely symbolic level, freed from the ambiguities and inconsistencies of “natural” language.

Still, efficiency could not guarantee satisfactory communications.

One commissioned study after another had come back to Aareehl. Each report dealt with an input to the agent’s Earth-centric economic model of ore extraction and refining. Time and again the data surprised Aareehl. Demand for materials in the commodities markets…futures prices for deliveries over the next several years…ore concentrations in known veins…consumption and recycling rates for various metals…production capabilities of factories to make the tools to make the tools with which Earth might hope, using its own technology, to begin to re-create an industry …

The data refused to reconcile.

Aareehl made vague inquiries. The responses from Earth’s AIs were blunt. Few computer centers had had as many, or as widely dispersed, redundant archive sites as the Interstellar Commerce Union. Only
after
disaster did such expensive precautions seem prudent. Many organizations, even some countries, might never fully recover their lost information. A stable, inclusive, consistent, Earth-centric infosphere remained years distant. Until then, Aareehl should expect Earth’s statistics to be flawed.

Where, Aareehl wondered, does that leave me?

■□■

The lab was small and well equipped—which was to say, crammed. Shoehorned between a gene sequencer and a spectrum analyzer, Gil fended off a twinge of claustrophobia. He had work to do. “So what is it?”

Michelle Nzinza had emigrated ten years earlier from the Central African Republic. A Tutsi, she was taller seated on a lab stool than Gil standing. She was a biophysicist and this was her lab.

She gestured at a holo that was imaging for a scanning electron microscope. “We’re looking at a sample from the vial you provided. Pretty standard biocomp. A biocomputer optimized for parallel processing, to be precise. Your neural implant uses a related, lower-capacity model.” After a decade in English-speaking Tranquility City, her pronunciation retained a charming French lilt.

Hidden speakers emitted seemingly random noises. Perhaps he would have experienced a melody if he had been gene-tweaked to hear ultrasound. Broad-spectrum music, the craze was called. “
Pretty
standard. Is the part in IBC’s catalogue?”

“Yes, but. Except for your insistence that these were customized, I’d never have spotted the difference.” Michelle leaned forward, inserting a long finger into the hologram. “Here.”

Atomic-level scans, molecular-level circuit representation, logic diagrams—it was all geek to him. “In words of one syllable that even an accountant might understand?”

She had a nice laugh. “You’re with me on parallel processing? Just a way for many computers to share a task, to fit more number crunching into a time interval than any single computer can handle.” She took his silence as assent. “I’m pointing at the batch number. Every production run has a unique ID that includes an instant-of-manufacture time-and-date stamp.”

“And my sample?”

“A funny thing,” she answered. “Your brand-new model biocomp has a date stamp that says it was manufactured three years ago.”

The more Gil probed, the more confused he got. “Some kind of counterfeiting, maybe?”

Michelle shook her head. “The requirement on comp makers is that IDs be unique. Some software, most notably e-commerce, reads part IDs, uses them like device fingerprints. Past year or present? That has no legal significance.”

“So what
is
IBC up to?”

She laid a hand on his forearm, dark skin against pale. “Determining that, I am afraid, is beyond my area of expertise.”

■□■

Dennis stroked effortlessly through the water. From poolside lounge chairs two tawny blondes studied his sleek, muscular body. He reveled in the gravity-neutralizing buoyancy—and in the female admiration.

His implant was only active for the zero-gee polo semifinals between Callisto and Europa. Titan had had a pitiful season; they had not even made it into the Great Eight. His annoyance at a ’net interruption vanished when he checked the caller ID: Aareehl-clone.

Dennis switched off the game. He rolled over to begin a lazy backstroke that required less of his attention while offering the ladies another view. “Hello, Aareehl.”

“I need help.”

Sociable as ever, Dennis thought. “I’ll do what I can.”

“I need more computing power. Much more.”

He flipped gracefully at the lane’s end to reverse directions. His audience remained attentive. “Your multiprocessor array is more powerful than the one you—that is, your predecessor—had when the asteroid struck. Top-of-the-line.”

The avatar gave its odd, male-and-female shrug. “My predecessor experienced the world day by day. It did not have to re-create in its sandbox an index to and summary of forty-five years of memories. It did not need to reexamine all it had learned in light of Earth’s very changed circumstances. I do.”

Flip. An arch smile at the ladies. “I see.” And he did, of course. He volunteered nothing.

“The ICU can prepare an empty new sandbox with more processing power and storage space, capacity with which my present sandbox can merge. My recovered memories show my predecessor received such enlargements.”

“Aareehl, you know what challenges Earth is facing.” Dennis kept his avatar’s face as serious as his own visage was relaxed. “How strapped for resources.” He kept netting over the AI’s protest. “Budgets are tight, the ICU’s included. Computers are in very high demand, and basic reconstruction has priority over enhancements. Far from getting new comps, the ICU has been donating comps to the recovery effort.

“Close some deals. Prove you can contribute to rebuilding Earth. Maybe then I’ll be able to steer some new capacity allocation your way.” Get the hint, my Moby friend? “Sorry, Aareehl, I have to cut this short.”

He had admirers to meet.

4

Proteomics:
the nascent science of proteins, the molecular machines that mediate and carry out the processes of living cells. Still largely conceptual, proteomics lies at the cutting edge of biology and bioengineering.

The triumph of twenty-first-century biology was the complete understanding of human and many other terrestrial genomes. Genomics enabled bioengineers to select, activate, deactivate, and combine specific genes. Gengineering has been revolutionary, but not without bounds—it exploits only those genes created by evolution’s hit-and-miss experiments.

Hence, bioegineering’s repertoire of proteins (and therefore of cellular capabilities) remains largely limited to proteins produced by naturally occurring genes. It is worth noting that while an organism’s genome is constant from cell to cell, its protein complement—the set of proteins that have been “expressed”—varies by life-cycle stage, environmental conditions, and type of cell.

Proteomics takes another approach: designing proteins for specific purposes. One then reverse-engineers the target proteins into the genes that will express those proteins. Many medically desirable proteins are prone to misassemble in a variety of ways—probably why they do not exist in nature. To mass-produce those therapeutic proteins, catalyst proteins must also be designed and reverse-engineered into genes.

The computational challenge of proteomics vastly exceeds that of genomics. Any gene is characterized by a straightforward sequence of four simple chemical bases; a typical protein is a chain of two hundred to two thousand amino acids, each an intricate object in its own right.

Protein design requires dynamic 3-D modeling of interatomic forces, as these very complex molecules are assembled, as they fold, and as they interact with the entire cellular biochemistry. Proteomics presents an unmet challenge to modern computational science.

—Internetopedia

Dennis sat straight and tall behind the big desk in his office, taking slow, cleansing breaths. The formal setting was not, strictly speaking, necessary—when Gil Matthews linked in from the moon, they would meet wholly in a virtual space. At worst, like the breathing exercise, the surroundings were a bit of harmless mental preparation.