Guys in lab coats weird me out. I’m not sure why exactly, but I think it has something to do with the mad-scientist look. I want to think they are doctors, M.D’s, but they are not. They are imposters draped in white-coats of authority, doing something insidious with glass jars and latex gloves, meddling with the real world, playing god. Maybe it’s because their world is so far removed from my own. I deal in abstractions, bits and bytes that exist ephemerally, equations and computations. I try to avoid the real world as much as possible. So when I found out that my neighbor Tom worked in biotech I was kind of put-off. He was a great guy and all, but I didn’t think we would really ever click as buds, being so different professionally. What could we talk about other than sports and mowing the lawn?
Yet, he seemed intrigued about computers and would ask me questions from time to time. At first this did not seem abnormal. People usually ask me computer questions, what machine to buy and how to make certain software go. Years ago I actually had opinions on this kind of stuff. That was before the boom, before everyone had a computer on their desk (along with a car in their garage and a chicken in their pot.) Back then, answers came easy. Now everything is so complicated, too many choices that I avoid these kinds of questions like the plague. Still, out of neighborliness, I gave him as good of advice as I could at the time. This encouraged him to strike up even more conversations, which would have become my nightmare scenario, except the nature of the questions started to change and this intrigued me.
He wanted to know about computing algorithms, linked-lists, sorting, that sort of stuff. I figured he was taking some type of course. They probably did a lot of modeling on computers in his line of work. I mean look at that human genome project; thousands of machines used to crack that one. That kind of stuff I could really get into. I gave him some good answers and even let him borrow some of my books. They were only collecting dust in the garage anyway. This went on for a few months, and as it did the questions got more and more involved. He wanted to know about threading and parallel execution. This was right up my alley and so what used to be me dreading a prolonged discussion turned into me spouting off for two or more hours. His questions started pushing the limits of my experience, getting into technical areas that were off into research. After a while, all I could do was help him search for online papers and translate some of the technical jargon.
I assumed he was just overextending himself because it was all so new to him, not understanding which parts were practical and which bordered on science fiction. I doubted it had anything to do with his work. He monkeyed with test tubes not microchips. He was probably just in over his head. It would be like me boning up on quantum physics. As interesting as that is, I would never gain much more than a cursory understanding of the subject no matter how much time I spent staring at the books.
Eventually, the questions stopped. There was a period of weeks where Tom just failed to drop by after work, as was getting to be his routine. It was a good thing too, because I think my wife was starting to get perturbed at us.
And then one evening as I sat eating dinner there was a knock at the door. It was Tom. He looked a bit flustered, his hair unkempt and his face unshaven. I couldn’t tell if there was something wrong. For a moment, we just stared at each other. He started to speak, but was having trouble forming his thoughts into words.
“What can I do for you, Tom?” I said politely.
He shook his head out of frustration. “I can’t explain, but do you think you can come by the lab tomorrow. I’ve got this thing.” He just left it hanging there.
“What thing, Tom?”
“I can’t really say. I need your opinion. It’s technical.”
Great, I thought, another comrade in trouble. He probably had a computer virus and didn’t know what to do. I should have told him to take a hike, but I’m not that kind of guy. I just don’t know how to say no. Fortunately for me, I work in an industry that allows for flexible work hours.
“Sure, I guess. What time?”
The drive the next day was draining. I don’t think I fully awoke before scrambling into my clothes and heading out the door. I was under caffeinated and barely alert. The music on the radio was grating. I couldn’t find a decent station. It was either some annoying chatter or eerily chipper boy-band melody. I had to turn it off. That left the sound of the engine and the wheels rolling over gritty pavement. It felt like I was driving over my own head. It was early.
Tom worked at the research branch of Vivalux. I had never heard of it before either. The address he gave me was down the valley, along the most recent sprawl of the high-tech corridor. It was a brownish stonework slab building, put up in the rush during the Internet boom. Some of the buildings nearby still had placards outside listing off crazy named start-ups. Almost none of them existed anymore; the offices dark, the parking lots empty. It was a concrete ghost town.
There weren’t many cars around the Vivalux building either, though I was glad to see Tom’s blue Sable parked near the front door. He must have gotten in real early.
I waited by the reception desk for at least fifteen minutes, and I don’t think I ever did see one. Another employee passed by the inside of a pair of glass doors and saw me standing there mulling over whether to sit down in the orange-fabric chair or just keep standing. He took pity and let me in. He seemed to know exactly where I needed to go and lead me there.
The doors into the lab were wide and swung both ways, like those in a hospital where gurneys burst through periodically, pushed by green suited orderlies. I expected to see a bustle of activity on the inside, hoards of hygienically sealed technicians manipulating test-tubes, with elaborate glasswork paraphernalia strung about the room delivering bubbling liquid through curling pipes; beakers, brackets and Bunsen burners. But it was just a large white room. The place made me nervous.
Tom was there with one other guy, a gray bearded fellow that reminded me of a distant uncle. There wasn’t much in the room, except for some stainless steel cylinders, a microscope and a bank of PC’s. The sight of the CPU’s should have calmed me, that being my element and all. But the place just seemed to have an unfriendly vibe. Maybe it was the white on white linoleum floor, the institutional grey walls or the grid work of high-glare fluorescent lamps. Everywhere else the building looked modern, but this room was sterile. It must have been built to match someone’s fanatical belief that workplaces should be dreary.
But that was not it at all. It was not the room that made me anxious. It was the lab coats. They both were dressed in sparkling white lab coats, encasing their bodies in the starched robes of science, covering everything but a hint of color near their feet and a drab knot of tie at the neck. I was definitely out of my element here, no matter how many servers, flat screens and mechanical rodents. If there had been a recliner, a beer and a bag of chips I still would not have been able to shake the overwhelming urge to escape for my life.
Tom waved me over. He still looked haggard, even with the clean attire. His eyes were set deep and the line of his mouth never came to a rest.
Tom introduced me to Dr. Salinger. He was the head of the facility, and it was his research they were all working on. Salinger smiled and said something polite, but he did not shake my hand. He did not seem to recognize the gesture and just went back to work at the microscope, noting numbers on a clipboard pad. I shrugged it off as par for the course. I’m used to social ineptness. Half the guys I work with don’t even bathe. It was probably the same here.
“So what do you want me to look at?” I said, my eyes spying the digital hardware. With any luck it would be something like the bedlam3 virus that spread around just about everywhere last month. That one was easy to find, just a disguised file in the system directory. I could clean it up in no time. If it was anything else, I brought a disk with full-on antivirus.
“We’ve got something strange going on here,” Tom said pointing to the apparatus around the room. “I just can’t explain it. It’s just not my field.”
“If it’s the software, I can take a look.” He probably just miss keyed some data in a spreadsheet. I see it all the time. People, especially your smarter types, get in to trouble like this easily. They think because they are experts at something like science that it immediately qualifies them as geniuses at all ‘lower’ pursuits. So they tend to do everything for themselves, building complicated spreadsheets, writing software. After all, it’s easy isn’t it? They read a book once.
“No, no,” Tom said. “It’s not the software, it’s the experiment. Let me show you.”
Of course, none of this made any sense to me at the time. If it was not software related what did he need me for? What was I doing here at 7 am on a Thursday? I started to wonder if they had a coke machine somewhere nearby when Tom lead me over to a white board.
“I should explain what we are doing.” His mouth curled in a grin. “I suppose you don’t know a lot about genetic computing?”
“You mean a genetic algorithm, right?” Of course, I knew a lot about genetic algorithms. I’d been using them in side-projects for years. The idea is to make software simulate an evolutionary framework. You encode potential solutions to your problem space as a series of numbers, kind of like software DNA. You take a lot of variations of these and try them out, scoring them based on some criteria of appropriateness. The low scorers are thrown out and the high-scorers are breed with each other, producing new patterns that get tested yet again each new generation. The idea is that through this process, better fit solutions emerge. It doesn’t necessarily lead to the best solution, but often it’s a lot faster than using brute force to try out every combination. Sometimes the search space is unimaginably huge and it would be impractical to do anything else.
“No, not software,” he said. “Genetic computing, with protein sequences, chromosomes, that kind of stuff.”
I almost didn’t hear the words. For a moment it sounded like gibberish coming from his lips and then it sunk in. It had gotten it backwards.
“You mean you’re building computers in a Petri-dish?” I had read something about this in a magazine once, Popular-Science or Wired. I guess the idea was to get a bunch of strands of made-up DNA to do the computational work for you. I didn’t really understand it at the time. Somehow you could do useful work by swishing these strands in a broth of proteins. The parts that combined would have detectable effects. I guess it didn’t seem like much use to me. It was like claiming a single NAND gate was a computer. Some scientist had figured out how to perform a simple math operation using this process. It still was not much better, a whole lot of hoo-ha over a whole lot of nothing.
“Something like that.” Tom replied, and then he sprouted that grin again. “Though, more like in a test-tube or flask. That’s what we were trying to do, only more advanced.”
Of course, that probably meant they had figured out multiplication where the last guy had only done addition. It still didn’t tell me why I was here. This was clearly out of my league. I was quickly trying to think of an excuse, some way to get me out of here and on my way to Krispy Kremes when Tom started drawing on the board.
“Normally, this process is a simple matter of recombination. You have a variety of sequenced chains and how they interact with the input set. By that I mean by what they produce. From this you determine the result of your computation. You usually get a variety of noise, but on average those cancel out and you are left with the majority response.”
Tom drew some circles and lines, squiggles for DNA strands and little square for proteins. For looking and sounding so scatterbrained he was very meticulous at the board.
“But this is only so interesting. The real trick is to get the computations to continue. No one’s ever done that before. The idea is to get feedback into the system. Without feedback you basically just get predictable responses, and any student with a text book and calculator could have figured those out.
So we’ve been looking at ways of using DNA strands as algorithms, ones that produce new proteins that become again information in the input set. You have to design your sequences pretty clearly so they don’t combine at the wrong stages. Different segments for different time slots and so on.”
“And this is what you’ve done?” I had to interject. I hate it when guys start to teach me things. It’s like they are preaching at me. “So you do series of calculations? Is that it? What about loops and branches?” I acted skeptical, but the truth is I just did not care. The whole subject bored me. If you wanted to calculate something, you should just use a PC. That’s what they are there for. Messing with chromosomes is so Dr. Moreau.
“Well, yah, those are the tricky ones. There are a lot of other labs out there trying to find the right sequences that will do things like that. Dr. Salinger had a better idea.”
I just stared at him. I think my eyebrow twitched.
“Dr. Salinger conjectured that the purpose of DNA in all forms of life is already to perform deep computations. So instead of trying to find appropriate sequences by throwing together pieces at random, we should learn from DNA that already exists. We started out by studying simple strands from microbes, but have worked up over the last few years to stands from higher order animals.”
“Like insects?” I tried to sound interested. I just hoped it wasn’t dinosaurs.
“We did for a while, yes. But we made our breakthrough when we moved up to primates. I think the Dr. really wanted to jump right to human DNA, but with all the stem cell nonsense he couldn’t get the authorization.”
“But aren’t they like near identical. I mean, even the insect DNA?”
“Yes, nearly, but apparently the differences do make a difference. The sequence may not be far off, but outcome of the computation is. You don’t look like a bug do you?”
I didn’t even deign to respond to that one. “Okay, so where do I come in?”
“I’m getting to that.” He seemed perturbed that I would even suggest that he hurry along. “What we found is that primate DNA does indeed compute. Now, what I mean by that is that we get a series of re-combinations. We’ve find that many different species give a multi-period response. It was difficult to see with microbes, but higher order forms sometimes yield ten to twenty cycles before stabilizing.”
“So they are doing something?”
“Precisely, they are computing long sequence instructions. But with primate DNA we found something even more.”
What could be more exciting? I yawned.
“Primate DNA keeps computing. It doesn’t stabilize, and it’s not periodic.”
Now, you might have thought that would make no sense to me, being the lab-coat speak that it was. But I know a lot about programming, from my earlier years as a BBS hacker to my later days as a professional code-slinger. Things that compute tend to be periodic, like pseudo random number generators. The less periodic the algorithms are, the more complex they are said to be. For some applications, complexity is very valuable, data encryption for one.
“But that’s not why you are here. You are here, because we found something else. There is some form of communication.”
“What do you mean? Like, interacting with each other’s computation? I mean because there all in the same soup right and so one guy’s output becomes another guy’s input. Wouldn’t that just be expected?”
“Yes, normally, but they are communicating across the room.”
I couldn’t tell if Tom was serious or just ribbing me. I hoped it was the latter, because I just did not want to go where he was leading me. The whole thing was too surreal already.
“You mean between test tubes?”
“Yes.” This time Tom smiled wide.
“How do you know that? How do you know anything is going on at all with these things? All you could possibly know is that they keep on churning, but you don’t know what any of it means. It’s all just random noise unless you know what they are computing. How do you even determine there’s communication?”
It was absurd of course. I knew that these guys had gotten in over their heads. They obviously misunderstood the whole paradigm of computation. Sure they were good with their chem and bio, but when it came to math and data, they were extreme beginners. If I were the overly arrogant type, I might just start lecturing them on information theory.
“Because they synchronize,” Tom said.
He gestured with his hands for me to follow him, but I just stood there, my brain spinning in infinite recursion, a mental double-take, a needle skipping tracks with max-headroom reverb.
On the counter nearby he lifted the stainless steel hood to reveal a rack holding a fat glass tube. Inside a brackish liquid rested calmly. Tom gleamed at it as if it were the seventh wonder of the world. But it was just a pale green liquid, with nothing interesting about it, except when it changed into pale blue. It pulsed between colors, a slow melancholy throb.
“The output set is visible, which is extremely rare already.” Tom stepped across the room and lifted the lid over an identical container. It too throbbed, now orange, then red. “The colors represent the height of the recombinant wave. Some are ahead of the others, so they kind of overlap making the smooth fading transitions. It is the crest colors that matter. Those are the real outputs.”
I stared at both tubes in a line. They were ten, fifteen feet about, and they pulsed in unison, the same colors in lock step, indiscernible.
Tom pulled a syringe from a drawer and used it to draw up a clear liquid from a plain white plastic bottle that had been labeled with masking tape and a magic marker.
“This is a biased input broth. We use it to initiate computation. Generally, this works for DNA with shorter periods. Specific inputs lead to predicable outputs.” He took the syringe and placed it over the far tube, and then after making sure he had my attention, he squirted the liquid inside.
A milky cloud formed inside the glass, everything fading away to white, and then the color started to re-coalesce around orange. Not one but both. The tube next to me also turned orange, and they both seemed to stabilize on that for a moment, but then they were off again, switching to green and then purple.
I must have stared at the tubes for a long while, my mind numb to anything else, but then I caught my reflection in the glass, my mouth hung open. I had to look away.
“Tell me how it does it,” Tom said. “How does one process interact with the other? There must be some exchange, some signal, some protocol but I cannot find it.”
I think I must have been too fazed to even respond. After a dead-long minute or two Tom waved his hands in front of my face.
“I don’t think you need me,” I said. “You need a quantum physicist.”
Tom smiled. “I was thinking that myself. Maybe there was some parallel universe thing going on, or some shared higher-dimensional link between the different strands.”
Of course, what I was thinking was ‘bull-pucky.’ Still, I didn’t have any idea what was going on, so it might as well have been sub-atomic hijinks. Surely, Tom didn’t get me down here to ruminate on some science-fantasy trip. Obviously, he wanted me to read the code.
“Let me see the data.” I said.
Tom nodded and led me to the PC console at the base of the computer rack. “We’ve decoded the genome into these files. That takes up most of the disks. But we’ve also isolated important regions that appear to do most of the work.”
He also had data showing combinations and splitting of different proteins across these sequences at different points in time. I did not know what any of it meant, yet, but I was now in my element. It was just another core dump from some processor needing to be debugged. I was a whiz at this sort of thing. It did not matter that none of the instructions were familiar to me. After you learn a few different instruction sets, you find that they are all pretty much the same. If you took x86 instructions and encrypted them using the best cipher, but showed me the core dump, I could eventually figure it out. There were only so many commonly repeating primitives used in computation. If this chain of molecules could add, subtract, shift and roll then I would eventually figure it out, given enough time.
Of course, given enough time a bunch of monkeys with typewriters could produce Shakespeare. Thinking about that made me laugh. I suddenly realized what this whole nonsense was all about. These genious probably thought a billion monkey chromosomes communicating with billions more out in the void was somehow equivalent to the pseudo-science notion of parallel universes, quantum computing and the whole ball of wax. Heck, if all these monkeys were computing something, they might as well be computing shakespeare. This was the kind of nonsense you got when you took a bunch of highly educated people and rammed a bunch of dreamy notions into their heads about reality and the universe; nothing more than comic book metaphysics.
The hours drug on, though it’s not like it was my first long haul session. You usually have to spend this sort of effort when the code has really crashed, gone fandango on the core, when the instruction pointer gets flung far into memory and starts interpreting random bytes as well intentioned code. Data gets trashed, code gets overwritten, and the pointer keeps bouncing around like a wild dervish scrambling up the sand whenever it glances against the ground. But if you are brave and diligent, tracking every load, every store, recording states before and after, you eventually start to see a pattern. You start to glimpse the celestial hand reaching down into the system, driving the system toward its ultimate goal.
I must have been there most of the day. I lost all track of time and never once did I start to think that maybe I’d be missed at my own office. Tom kept bringing me cokes, and Salinger kept wandering around behind me, fiddling, tinkering, trying to determine if I had made any progress. Though, I doubt he would have noticed when I did. By the time I stopped to glance up, I had accumulated a pile of notes and doodles an inch thick. Somewhere in that stack, an exact recording of my entire brain process for the last ten hours was inscribed using hashes and squiggles, loops and arrows.
I put the pencil down and pushed my chair back.
Salinger was at the other end of the room now, but he stopped what he was doing and looked my way. Tom was walking back in through the double doors carrying a fresh supply of red twelve ounce cans.
“I’ve got it.” I said, matter-of-factly, and then I shook my head.
Finally, Salinger spoke. “This is it?” He said, shocking me with an awkward European accent. “You’ve broken the code?”
Tom handed me a can. I popped it open and took a swig.
“It took me a long time, but once I figured out the branching mechanism that lead me to see how it used the proteins as both messages and memory cells.”
This time it was Tom who had his mouth hung wide. I think Salinger was starting to drool. “And this is how it communicates?”
“Well, no.” I said. “It doesn’t communicate at all. Not across the room anyway. I mean how could it?”
Tom stammered, “But then.”
“Look, you said the sequence was random, and non periodic. Sequences like that are not predictable, so you have no way of knowing how either system will advance.”
“But we changed one and the other adapted.” Tom said.
“What you are telling me is that you don’t actually know that you changed anything? It is random!”
“But they are synchronized,” said Salinger.
I sighed and ran my fingers through my sagging hair. “True, it appears that way, but that’s only if you don’t take other things into account, like the fact that the strands represent an exactly identical program, and that there may be something else that is synchronizing them.”
“I don’t understand.”
“Sounds, vibrations, sub-atomic resonance, any number of things could be providing the agitation necessary to keep them in sync, even the light waves themselves could be causing the necessary interference.”
I think that got to them. It did make sense, after all. Tom’s shoulders sank. I think all the strength fled him. He had to sit down on a stool. “I guess it could be something like that,” he said.
Still Salinger persisted. “But you figured something out, yes?”
Of course I did, and for a guy like me it was even better that science fiction.
“The colors, switching back and forth; I know why they do that.” Now it was my turn for the big fat grin. “I finally figured out how the code is broken down into computing fragments, like little subroutines. Once I got the major blocks figured out, I could start to piece together what it was trying to do, what its overall purpose was, and that’s when I realized that the subroutine was not supposed to be there at all.”
Tom pinched the rim of his nose and then folded his arms across his chest. Salinger just nodded his head and took notes.
“You see, most all of it hung together. I could see the signature of the design, the style of the code, and that’s when I realized that the color flasher was all different. It did not belong. So I tried to find how it got activated and that took quite a while. What I found was a series of rare inadvertent states that when processed together finally lead the code right into this strange bunch of bits. I would have figured it as some loophole or bug in the code, but the flasher is so elegant, it had to be designed.”
I could tell Tom’s heart was now aflutter by the way he bit his lip and squirmed in his seat.
“Well, designed may be too strong of word. There could be any number of explanations on how the thing got there. What is important is less how than why.”
I breathed in deeply; wanting to make sure I got his out as clearly as possible. “You see,” I said. “It has no particular purpose. It’s just flash and almost impossible to come by under normal circumstance. In the world of computation, there is only one thing that it could be. It’s an Easter Egg.”
With that I left them, Tom still perched on his stool, Salinger confused, pondering my words, probably wondering if he understood my English. I walked out of the room through the double doors and headed for my car.
The truth is, I didn’t know why the thing was there any more than you might fathom a guess. I just knew what I saw; the function was a marvel and an oddity. It was either some cosmic joke being played upon us by some jovial deity, or more likely, a joke played upon Salinger and Tom by some crafty colleague, some other guy in a white-lab coat playing god.
So where ever it came from, as a gag, at least was a good one.