“And you tested it on yourself?”
“Yeah.”
“In the middle of the night.”
“Well...yeah. If that’s relevant.”
I do my best to tell the story of my flawless new inversion, in circles that wind around this point — I already told you that I learned from the best — but eventually she clocks it. I expect her to be angry, and I almost flinch when she puts her hands on my shoulders and gives them just a tiny squeeze. “Oh, god, you poor thing. Are you okay?”
I bring my hand to my forehead. In truth, I’ve been waiting for my brain to start oozing out of my ears. Last night, there had been no question: who else was I supposed to test the inversion on? But when I brought it to the clinic this morning, nobody from the Big Three thought twice about volunteering to be patient number two. I should have known that they’d be game. Given their risk tolerance, and given their alternative.
“I’m fine. I am.”
Deng produces a pen-light from her waist pack. “Here. Here. I want to check your pupils.”
“No, I’m...” But I lean back as she wiggles the light into my eye. “Seriously. You’ve made your point.”
“Which is? Blink.”
“Which is that it’s a dumb fucking idea to infect yourself with debris.”
“Which is that I worry about you.” She clicks her tongue as I scowl. “Oh come. Not like that. I want you to track my finger now. I only mean, I want you to be happy. I want you to be safe.” Stows the pen light. “Tea?”
After she offers me some hot water, as I’m digging in her boxes of chamomile for a stray Earl Grey, she walks over to her whiteboards and makes what I consider a big show of wiping them clean of her sketches and calculations. Her meaning could not be clearer: I guess we won’t be needing these any more. But I don’t know how to square it. I keep waiting for her to drop a “but” on me. But you need a theoretical grounding. But we need to talk to the IRB. But we still can’t publish yet. But it’s still just a little piece of you.
I know she wants to say it. I watch her almost say it once or twice, I can read it from her lips as she blows at steam. Hell, I almost say it for her, as though I don’t want her to give ground, like I wouldn’t know what to do with it. Of course she’s right about it. But maybe she’s run the numbers and knows that if she brings it up, I’m going to walk away. Or maybe Dr. Ren called her at four in the morning to twist her arm.
“Yes, it might be for the best.” She snaps out of several minutes of calculating silence and springs to her feet. “Mona, this is the fun part. I think you’re really going to like this.”
The faculty aren’t hoarding the best scanners for themselves. Not really. They’re breaking them in, they insist between chuckles. Working out the kinks. Still, there’s nowhere on campus outside of the faculty lounge where you can use a late-model Kanwei Pulsar. The last door on the left is for a tiny scanner room, not really big enough for the both of us. It’s dimly lit, the suggestion of sunlight through dense conifers. Ambient music pulses quietly from the walls. And, god, the Pulsar! In another, lower-stakes mood, I’d be all over this thing. It has citrus-scented moisture-exchange padding where the clinic’s beds have sweat-stained foam. It has chrome-plated controls on the armrests. It has fucking mahogany accents, man. I dunno.
Deng watches my reaction with contained amusement. “Kanwei claims they’ve nailed prose with the Pulsar. If you’re going to be writing, you should try it.”
Prose doesn’t mix so well with loop-lock. Writing in this way amounts to tiling in with a vague intention and letting the scanner probe your mind with a language model. It’s remarkably fast and eerily passive. Personally I doubt that Kanwei has nailed prose, but it can’t be worse than the first few assignments I handed in written on student-center Gliders, five or six years old even back then.
“It might take a few tries, but I’m told it’s very fast. I’ll be waiting outside.”
So I close the door to the booth. A stillness descends, punctuated by the scanner’s clicks and beeps. I do know how to write this paper. Most of it will be the model’s doing, the guardrails of academic prose. But I can supply the rest. I have a mathematician’s sense, imperfect but immediate, of how my inversion works. At academic distance, it doesn’t feel quite like a weapon of mass destruction. Maybe, for once, I’ve just done something good.
Even the visor smells nice. I paw at the shiny knobs, dialing in my settings, and then the Kanwei logo hovers in front of my eyes. STAND BY FOR TILES. The magnetized whine of the scanner encroaches and overtakes…
And I understand the Sunflower Sieve debris better, too. I understand it with textbook words like pathological subdomain and measure-zero Kakei expansions. It spills and twinkles from the negative space of Deng’s equations. I won’t think about that.
The nearly ultrasonic whine takes on even higher, tangy, citrus harmonics as an absolute wall of tangerine assails my eyes. The tryptamine hits: I feel a wave of intricate bliss, a vivid image of Deng smiling knowingly — the trip almost gets away from me before I can focus on the color, willing it to burst into four, and then again and again, primer tiles looping more than a hundred times a second, until I'm right on the verge of it. The sync is crisp, the colors bursting with skylike luminescence, the computer firmly but smoothly under my control, and I under its. The building blocks of my reality in loop-lock with the Pulsar are like a gemstone dust, almost fine enough to let real emotions swirl misty around me.
I could get used to this, I think. And then I’m back. There was the flashing, oscillating sensation of the language model scrubbing my channels. A blinding, rainbow-road prismatism. Now, just the sound of blood in my ears. Barely thirty seconds have passed, so little time that the DMT has only just peaked. I let the trip fade, birdsong made of bells made of birdsong, and then rise from the scanner to see just what it is that I’ve written.
A Generalized Inversion for Sunflower-Class Debris
Xu Maoli (xu.mona@yins.edu), Deng Jinghan (deng@yins.edu)
Introduction, yadda yadda. Background and motivation, stop me if you’ve heard this before.
without the usual considerations of dimensionality (Deng and Liang, 2057). However, the algorithm uses an unusual variation of the isotropic binding operator introduced by Deng (2061) which yields a modular residue upon application to the binding graph for a fundamentally incompatible tileset. It further applies a pre-binding notch filter (again due to Deng, 2062)
It gets weird about halfway down the first page. My advisor has authored hundreds of papers, and I’ve read a tiny fraction of them. But in loop-lock I can churn through all of them in a few minutes. And the thing that is both Mona and the machine...is utterly convinced that the Sunflower Sieve is Deng’s work. The foundation was laid years and years ago, in paper after paper, and someone’s finally put them together.
(Deng and Liang, 2057). However, the Sieve’s unusual choice of pathological subdomain presents a second challenge: irregularly spaced attractors in the k-space of its debris that, when the Waldmann operator is applied to the binding graph by , again produce elements of the pathological subdomain. We therefore consider the debris as a truncated warning that it’s coming (Deng, 2059), we all feel it and yet none of us can form the words to fence it in the tide is too high and still rising () and if any of us are to escape it will be in pieces
...oh, for fuck’s sake. I put myself back into the scanner and try again from scratch. This draft is more promising...until it’s not.
introduce the notion of “smeared” seed sequences which are computed via a moving window deep below deck moving up and down. Are you feeling it (Deng et al., 2060) the tension on the line the hook in your mouth the porthole convex. Known as hyperlagmites, these structures in the Mirror Sea
I swig half a bottle of chilled water. On a heavy linen paper towel I put down some of Deng’s equations in cramped hand, steering my mind away from all that, towards dry and objective formalism. Take three:
third case in which has we can choose arbitrarily whether the binding curve is concave () or convex () (Deng, 2063). Recovering this degree of freedom reveals runes unseen. The cardinality of is deeper than you thought, but then you’ve already seen stolen sunlight twinkling on their edges you’ve been there you’ve been them it cannot (without loss of generality) be otherwise. It builds because it builds rises because it rises loops because it loops and it will overtake silent and silky the mirror sea will be the sea they crawled out of (Deng, 2064)
I make myself delete this draft before I read any further. It descends into shanzi, characters with no fixed meaning, radicals echoing with sinister suggestion. Whatever gnarled half-truth is trying to climb out of me and onto the page, reading it back is only going to make it worse. I climb back in, ready to try again.
Deng raps quietly on the door. “Are you all right in there?”
“I’m fine,” I shout back. Sweating now. Knowing that when I next walk through the cafeteria, I’m going to feel that debris hanging in midair, overflowing from the tiny crevices between word and gesture. I’m going to remember vividly how I saw it chattering silently in the Mirror Sea in just the same way. And the loop is going to tighten. The gap between here and there is going to feel like no space at all. Last time, with Cai by my side, I loved this feeling, loved it all too much. Now I climb into the scanner again and fire a massive magnetic charge into my channels, trying to burn it away.
“This is exciting,” Deng’s muttering to herself an hour or so later, hands flying at the lab server’s keyboard. She’s corresponding author, which means that it’s her job to fill out the various declarations and disclaimers of the submission process for the Soup — the open-access archive for the field of neikotics. My job is to pace uncertainly around her office, swigging from a water bottle. I watch her compose a two-paragraph abstract without hitting the backspace key once.
“Jesus Christ,” I exclaim, looking over her shoulder.
“I’ve written a lot of these, you know, in my time.”
“No, I mean — your inbox. Do you ever clear that out?”
Deng has “99+” unread messages on the Soup. She smirks, clicks and scrolls lazily through them. “When you’re one of the…” she begins, and stops when she can’t find a self-effacing way to finish the thought. “I mean to say, a person in my position attracts a lot of crackpots with their crackpot theories. It’s not really a good way to get ahold of me.”
To say the Soup is full of crackpots is an understatement. They’re firing thousands of synthetic papers an hour into the Soup, and why shouldn’t they? Some of the great guidestones of neikotic theory were produced in this rock-tumbler. Waldmann’s Folding Lemma was a synthetic paper, an offhand mathematician’s footnote about choice of sign that bounced blindly and abrasively off all the nonsense in the archive until its contours were clear. The computers are great at sifting these diamonds from the silt, and a synthetic paper that racks up ten thousand or so citations is usually worth at least a glance by a human being.
Later on, Deng lets me click the button. Submitting to the Soup is a pretty visual affair; I watch my paper drop into the citation graph, losing track of that single star in a spiderwebbed milky-way. Within a minute, there’s a flash of activity as the synthesizers get ahold of it and deem it worthy, tying A Generalized Inversion for Sunflower-Class debris in silken strands to the rest of the field. No doubt Deng Jinghan’s name has a lot to do with that. A relieved exhale, almost a sob, as my work is woven into the story of neikotics. I’m published. I finally did it.
After ten citations, in about an hour, it will be cryptographically infeasible to pull the paper from the Soup. My paper will have circumnavigated the globe, from server to server; it will be on there forever. And I tell myself that then I will say something. But in the meantime...
Behind Deng’s big stack of tea boxes she’s apparently been stashing several pricey blue bottles of Yanghe. She twists one open with some difficulty and sloshes some into her teacups, and then we drink, laughing and coughing as the baijiu stings our throats. I sigh on borrowed contentment and cast lazy glances around her office, which is half-illuminated by a stolen salt lamp. There’s a gunmetal-grey piece of a Navy loop-lock rig. A wall of degrees and certificates and commendations. Only one photograph. Deng’s mother, who died the year we moved here.
I think we both know there’s more to say here, and the drinking accelerates a little as we both try to figure out what it is. The liquor softens everything as I watch the citation count tick into the hundreds. I consider that I could just let it go for once, do the peaceable thing, break the loop.
Or.
“You’re cited a lot, you know, in my paper.”
“Hmm?” Deng hears this mid-sip, and swallows funny. “Yes. I noticed that. It makes me look a little immodest, I suppose, but this is nearly inevitable in student work. The advisor looming large. You’re giving me a look.”
Clicking through citations now. “Deng. Deng. Deng and Liang. Li and Deng...”
“Different Deng.” She pours again for the both of us. “But I take your point.”
“...Deng, Deng, Deng et al., Deng and Hong. There’s just like an incredible amount of detail on the structural building blocks of the Sunflower Sieve in my paper. Stuff I didn’t know I already knew before I got in the scanner.”
“Well, that’s the beauty of it.”
“But I just feel like...” I sigh. “It’s stupid. I’m sorry.”
She surprises me here. She takes what I’m trying to say, holds it with ungloved hands up to the light, with a bit of a bemused smile. “You’re saying that you think I have something to do with the Sunflower Sieve.”
“No, no, no.” Yes, yes, yes. But the drink dulls the gnaw in my gut. “Nothing like that.”
“I’m beginning to think you might be right.” She pours for us again, raises her teacup as though in toast, but then: “Pop quiz, Mona. The Standard Neikotic Basis has how many primal forms?”
“Twelve,” I reply, a little insulted, after the baijiu clears my throat. “Everyone knows that.”
“Why?”
“Well...” Fuck. “Because you can’t really get away with fewer? That’s what makes it a basis, roughly speaking.”
She brandishes her hard copy. “Third page and, ahem, second column. Read for me please?”
...forensic decomposition of 132 debris samples shows that the Sunflower Sieve uses a 9-piece basis proposed to and ultimately rejected by INSI in favor of the now-ubiquitous SNB-12...
Deng tosses me a whiteboard marker. The shapes flow easy off the felt tip. “Okay. This one zigzags like this. And then this one hooks left, right, and then it would come out of the board. And then the last one...”
I can’t quite remember the last one. But I recognize it, and my heart skips a little when Deng comes in beside me, in green, and draws out the little pinwheel nexus. “Yes!”
She caps her marker with the ghost of a smile. “This. This was my idea. What you were taught as mathematical fact was really a decision by committee some fifteen years ago. The tileset we use today, SNB-12, was standardized instead of a different set, SNB-9, which had only nine basis forms. The nine-piece version feels a little alien to us now, but it had its advantages. Fewer forms to memorize, for one thing. Gets you into loop-lock faster. Some composite forms were a little harder to express, but once you got your head around those — it sings.”
“But it lost out, the way the harder thing always does, even if it’s more elegant,” I surmise.
“Worse is better. You see it. The nine-piece basis was completely brilliant but nearly impossible to work with, they said.” Deng scoffs. “What’s so funny?”
The paranoid wind has not quite gone slack in my sails.
“So but you did know, the entire time, what this thing is actually made of.”
She makes a little equivocating gesture with her hand. “That overstates it. I had an inkling. Truth told, it unsettles me, to see this still out there. It means one of the old guard is probably behind this.”
She huffs, rubs her nose pensively, as though that were a complete thought. I grip my chair. “Well...who?”
“Your guess is as good as mine. Old as dirt and rich as sin, and still likely a practicing neikonaut after all these years. It does narrow the possibility space in a rather...tantalizing way, don’t you think?”
“Guan Zhumi?” The Haojie founder. “You don’t think...?”
“Oh, lord. If I had to guess, from what I know of the man, I doubt he’s run anything other than pleasure protocols on a scanner anytime recently. But I’d be fascinated to be shown wrong. One more?” She wiggles the bottle. “You know, as it happens, there’s an INSI working group session happening on campus tomorrow morning. And not a few of the old guard among them. They can’t really stop me from showing up and they usually don’t try. Would you like to join me? Read the room? Compare some notes?”
“I can’t.” It creeps up on me, the icy feeling I know I’m due for. “I have my session.”
Deng chews her lip in acknowledgement. “Still? After all this time? You’ve been taking your medication.”
“Yeah.” This is my one consistent lie, and I think she knows it. The antipattern agents that I’m supposed to be taking make it hard to fish patterns from the noise. They’re supposed to prevent me from seeing things that aren’t there in the Mirror Sea, but they also make it impossible to enter loop-lock. “Still. They say I’m doing really well. That I might not need to come back in much longer.”
“I think that time has long since arrived,” she tuts. “Well, hurry back. And don’t let the Weather Bureau drag you down to their level.”