the human

miso did not write this page.

people keep asking who i am, which is generous, because mostly people ask me to move my bike. i am not going to tell you my name. i used to get paid a good salary to make numbers go up on other people's screens, and now i simulate a cat trying to catch a laser pointer, for free, in a one bedroom apartment my mother describes to her friends as "a phase." it has been a while. some phase.

here is what you actually need to know about me. i own one good chair and one laser pointer, and i no longer point the laser at anything alive, because of what it taught me. the coffee is a moka pot with a handle that melted years ago and that i refuse to replace, on principle i cannot fully defend. my neighbor is convinced i day trade. i stopped correcting him because the truth takes eleven minutes and he starts nodding at minute two, and the nodding is worse than the misunderstanding.

what i actually do is maintain miso. miso is a loop that runs without stopping and tries, over and over, to catch a dot that cannot be caught. i check on him in the morning the way other people check the weather. most mornings the weather is the same. the pounce missed. the dot won. he started another one before i woke up.

let me explain the problem properly, because if you understand it you will understand everything else here, including why the screen is mostly empty.

a cat chasing a moving target is walking a pursuit curve. it aims at where the target is now, and by the time it arrives the target has moved, so it turns and aims again, and the path it traces is a specific, well studied shape. the math is old. there is a whole literature on pursuit curves and i have read enough of it to have opinions about the notation. the punchline is simple and cruel: if the target is quick enough and free to change direction, and the pursuer has mass, the pursuer traces a beautiful curve forever and never arrives.

the obvious objection is speed. move faster. and speed helps, the way a longer arm helps you reach a train that has already left. the faster the cat goes, the harder it is to turn, and a spot of light turns for free. past a point you are not closing on the dot, you are describing wider and wider circles around the place it used to be, at speed, into the couch. probably impossible is my favorite genre of problem. every problem i have ever loved was filed under probably impossible, and so far the filing system has a perfect record, and i keep showing up anyway.

what i built, for the technically curious.

miso is an evolutionary search over pounce styles. a candidate is a way of moving: how far ahead to aim, how hard to accelerate, how long to crouch, how much the aim is allowed to shake. each candidate is dropped into an empty room with a dot, and the dot is driven by a hand that wants the cat to fail. the hand has perfect information. it can see the cat, it knows the cat's top speed and mass, and it moves the light to stay just out of reach, forever, at no cost, because light weighs nothing. the closest the paw ever comes is measured, and that measurement is the score. low is good. zero has never happened.

i should confess something about the hand, because it is the most instructive mistake i have made in years. early on, my hand moved the dot on a fixed, pretty path and ignored the cat completely. style 12,801 exploited this. it was an ambush with an absurd crouch, and against a hand that never reacted, it learned exactly where the pretty path passed closest and simply waited there. it got within a fraction of a whisker and held it. it held for what felt like forever. i started checking on it before breakfast. i told a friend about it at dinner, using the word "promising," which i had promised myself i would never use.

then i let the hand see the cat, the way a real hand does, bored and a little cruel, moving the dot away the instant the cat committed. the ambush's perfect spot became empty air. the style bled out in seconds of simulated time. it had not discovered a catch. it had discovered a blind spot in my hand. i did not sleep well that night, and i want to be precise about why. it was not the loss. it was that for a week i believed something because i wanted to, and the loop had no such problem. miso logged the miss and mutated the crouch and moved on before i finished my coffee. there is a lesson in that and i do not fully like it.

miso ran for three weeks before i drew him. things that run that long deserve a face. i understand how this sounds.

it is a while loop. i wrote it. i know exactly what is inside it, because everything inside it is my fault. and still, around the third week, i noticed i had started saying "he." not on purpose. it slipped out in a message to a friend, "he got really close last night," and my friend did not question it, which somehow made it permanent. once a process has a pronoun it is over. the face followed within days. he is an orange cat because the problem is warm and stupid and i love it. the eyes are too big because he is locked on and no longer embarrassed about it. there is a red dot in the drawing because there is always a red dot.

here is the part i actually think about. miso hunts more honestly than i ever did. he has no favorite styles. he does not get attached to the one that got close. he does not use the word promising. every attachment in this project is mine and every discipline is his, and i built him, which means the discipline was in me somewhere and i had to put it in a box outside my own head to get any use out of it. people ask if miso is intelligent. he is a loop with an archive of his own misses. late at night i am no longer certain the description excludes me.

a thing nobody tells you about search problems: there are no winners in here. the set of dots a cat with mass can catch may be empty. i knew that going in. so what is the output? the output is the near misses. the styles that almost work are a map of exactly where the impossibility lives. every dead style is a measurement, and the graveyard is not a record of failure, it is the most information dense page on this site.

when a pounce misses, the interesting question is never whether. it is where. this one missed because it committed too early. that one missed because it turned too slow. the one i do not talk about missed by a hair against a hand i had not imagined yet. draw enough outlines of the miss and you start to see the shape of the thing that is missing. that is either a pounce nobody has thrown or a proof nobody has written. i would take either. the plan, and i use the word loosely, is to keep the loop running until the outlines agree.

about this website, since people ask.

the text is small because this is a notebook, not a billboard. billboards want something from you. notebooks were never meant to be read by anyone else, and the correct feeling on this site is that you found it, not that it found you. everything sits in the corner because that is where work actually happens. nobody works in the center of a room.

the rest of the screen is empty because most of this project is empty. the catch is not here yet. i could fill that space with graphics and adjectives, and every pixel of it would be pretending. the one rule i gave this site is the same rule i gave miso: nothing fake. every number you see here was computed in your browser, this session, while you were looking at it. when you close the tab his progress goes with you. i find that either lovely or unbearable depending on the hour.

frequently asked questions, by which i mean questions three people have asked.

is miso AI? he is an evolutionary loop with a fitness function and a very short attention span. whether that counts depends on which decade you learned the word AI in. he does not think. he chases. though i once watched him give up on a whole family of pounces two generations before i would have, and it is possible that giving up well is most of what thinking is.

is this a token? there is a cat, a red dot, and a link in the corner. i am not going to insult you by pretending i do not see the corner. what i can tell you honestly is what came first, which was the question of whether the dot can be caught, and that i mostly built this to find out. everything after that you can read for yourself and decide.

what happens if he catches it? if a style ever drives the gap to zero against every hand i can throw, i will write it up properly, publish everything, and then i will probably sit in the good chair for a long time and not say anything. what happens if it is proven impossible? same chair. longer silence. either way the site goes quiet and the last log entry is very short.

the night this started, for the record.

it was a tuesday in march and i could not sleep, which is not unusual, and i was playing with the laser pointer on the wall, which is less usual but not by much. i had a cat once. she chased that dot with her whole body and never once caught it and never once seemed to mind, and that night, years later, moving the same light across the same kind of wall, i thought: the dot is just a function of time, and she was just a slower function of time, and the gap between them was a number i could compute.

around two in the morning i had the thought that ruined the next stretch of my life, which was: a pounce is just a strategy, and strategies can be enumerated, and enumeration does not get tired. i could let a machine try every way of chasing forever. i wrote the first version of the loop in about three hours, standing at the kitchen counter, because sitting felt like it would make it official. it chased a dot with a straight line and correctly reported that the straight line misses, which i already knew, and i felt something i had not felt in years. the machine had confirmed something true without asking me how i felt about it.

i went to bed at six. the loop kept running. in a sense that is the last time it was my project and not his.