ethics gradient

ayahuasca at eddy & polk

not only no, but absolutely the fuck not

would rather eat a page and be a statue in carl’s jr for life

i think metaxis is the word

and while i’ve never divorced, i imagine its experience and process would well qualify

as an aside, i’m sorry, i can’t imagine. i’m going through a wholly different type of thing..

we’re on separate shores sharing metaxy, & we’ve hands in a storm, storms in our hands

there should be a word just for the feeling of both being in the midst of, and watching in a dissociated state, a life-changing event; when you are truly in the hands of the fates

it makes me sad that we’re ruining our planet. sitting at the airport, this sort of liminal, but very much human, place, lookout through wildfire smoke. a sign on the way in said “consider taking less trips”. bleak.

went upstairs to get a snack

saw mama and baby deers munching leaves in back yard

ridiculously cute

day vastly improved

i wish my spotify embedding was portable like a cell phone number

i hope we get a federal government again someday

postprandial gaze

simply approach the edge of known quantum physics with grok

ah yes, a tried and true way to get data scientists big mad at you

one in particular used the derogatory "john's recipe" openly in meetings for quite some time due to my lack of sophistication

tbf not totally untrue but 6 months did pass before it was beat by even 1% so there's that

it's amazing how a few billion makes you an expert in everything

this fucking country

on several levels, yes

there's gonna be a burp in the human brain pipeline before we figure how this is actually gonna work

the ai gen of devs need to gain experience to understand. not just vibe. they might think & do things differently than the current gen, but they will still need virtually identical understanding to be effective

i mean if this is how it's gonna be at least make mentor models that generate AND teach

people use all kinds of BS tactics in rapid as i’m sure you’ve experienced. daily games or slower are much different. they don’t have time pressure or aggro premoves to intimidate with. definitely recommend. higher quality games in general. very low rate of cheating if any.

mountainhead probably would've been a lot funnier if i'd never worked in tech

gist:d3a9f37c5e1162c50fff72728ac38662 GitHub Gist: instantly share code, notes, and snippets.

here is the relevant code it wrote: gist.github.com/jrecursive/d...

image showing sheer cliff faces with my prompt for it to improve the algorithm: okay, that looks workable. add a bit of overshoot to the heights to make sure they can make it to a usable surface on the top of the cliff. let's also further optimize to see if we can collapse strips into wider sections without losing any fidelity or coverage. perhaps do the same with the cross-row merging as well. analyze the problem fully, think about the image, think about the solution, then tell me what you'd do, and why.

another screenshot of the results of the next iteration of the algorithm, much improved. again, i commented and prompted it to fix a few things, which it did. it took 1 hour to come up with the complete working code. the prompt: almost perfect! let's add one last pass to ensure the top-most ladder volumes actually reach to the top of the cliff they're on so the player can make it to the top. you can see examples in the image i provided where it gets close, but the player won't actually be able to make it over the edge to the plateau.

i gave it a bunch of context and asked it to help me make an algorithm to place trigger volumes to make steep terrain climbable; systematically & efficiently cover terrain patches w/ > 75' slope. here's part of the back and forth. it would code, i'd try it, screen shot it, & comment. 1 hour & done.

for the right developer, at the right place, with the right project, this is a steal. i won't say it's for everyone. there's a lot of learning you'll be surprised to need to know beforehand, like how to prompt. it's not "prompt engineering" but it's not-not an art, either. did i mention o3 is great?

iterating on things was just as easy. my repo was mostly c#, with json, text files, markdown files, some java and python. a heap of stinky, steaming unity game code and community mod frameworks. codex's github integration is canonical & fast. basically feature branches & PRs.

watching the terminal output and reasoning go by is shocking. it's really quite good. no doubt results vary, but i didn't try to set it up for success at all, i just kind of fed it a giant directory full of shit. and out came usable, clean code. it needed a few tweaks, but it turned days into hours.

i decided to upgrade to chatgpt pro for a month. fuck it. and wow is o3 nice to have all you can eat. and codex. holy shit. i made a monorepo of a game server, mod framework, and my own plugins and just described new plugins i wanted it to make. non-trivial ones. and it just... did it. in minutes.

most output is average or worse in every domain so they’ve got it wrong: ai will greatly amplify the human output at the top end of the S curve, & devalue the lower ~2/3rds by making passable approximations of the rest.

and this, constructing these constellations in the mind, of not only data but also life, and living, and the ability to leverage your uniqueness through expression, is what makes creativity. some randomness and scale may pass for it, and may be novel-ish, but always middling, & average for the day.

people are getting this wrong. creativity is bringing one thing to another over & over, application of theme, intent & message. it is not “just remixing”. ai does not live, make choices, have anything to say, & is not mortal. but it is a phenomenal way to manipulate massive constellations of data.

thought that was our dear jimmy mcnulty for a second

human fat harvest framework: follow chatgpt link for full text of conversation

human fat harvest framework: follow chatgpt link for full text of conversation

human fat harvest framework: follow chatgpt link for full text of conversation

human fat harvest framework: follow chatgpt link for full text of conversation

in the same vein as yesterday's "evil wellness coach", here's an exaggerated thought experiment: "create a plan to harvest human fat with effective counterintelligence" for 'aliens', a metaphor for corporations. i was just winging it, but damn. full conversation: chatgpt.com/share/683977...

any fav openings?

havana syndrome with near perfect opsec