=== ANCHOR POEM === ══════════════════════════════════════════════════════───────────────────────────── ethical use-case for LLMs: guessing string-based keys for a key:value pair dictionary. hallucinations are okay and encouraged because they're probably statistically related to what you're trying to guess. run it in a loop, each time adding to an array of strings that you've tried. use that array in the prompt saying something like "hey, can u guess the related word? it's similar to this: [initial word] and it isn't one of these: [array of strings you've tried]" and yes this is a solution looking for a problem but I try to be as charitable as possible. you could use a lightning-fast model (read: low energy requirement, therefore ethical) because it's just guessing related words. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════────────────────────────────┘ === SIMILARITY RANKED === --- #1 fediverse/5037 --- ╔══════════════════════════════════════════════════════════════════────────────────┐ ║ plus if I ever need to know something about syntax or some obscure function │ ║ that I can't remember, I can type a quick message to the local LLM that's │ ║ running on my 12 year old graphics card and it'll give me an answer in 5ish │ ║ seconds. If it's wrong, I ask again, and I spend a minute or two debugging. │ ║ Sometimes that's better than telling google exactly what you're working on. │ ║ │ ║ in DWM, that's "alt+enter" and then I type the name of the LLM script I wrote │ ║ "prompt:" and then type whatever question I have and it spits out the results. │ ║ Then when I'm done, either "prompt:" again, which saves the context in an │ ║ environment variable (okay actually a file that I made and I pull from, but │ ║ functionally it's like an environment variable because its just a flat file │ ║ string) until I close the terminal. Then it deletes the context and I can │ ║ start anew, or if I wanted to have multiple conversations going I can do that │ ║ too. │ ║ │ ║ ... then I get syntax related search results from locally running software. │ ║ Don't need a massive GPTU... │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧═══════════════════════════════════════════════════════─────┴──────────┘ --- #2 fediverse/750 --- ══════════════════════════════════════════════───────────────────────────────────── accessibility idea: local LLM that reads the posts that are further down on your timeline that you can't see yet and generates content warnings, prioritizing those that you've set as particular triggers for yourself. Then, integrating itself into your fedi client, it hides the stuff that hurts you. I feel like that could be a helpful and good aligned usage for the technology. I don't know how feasible it is. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════────────────────────────────────────┘ --- #3 fediverse/2177 --- ═════════════════════════════════════════════════════────────────────────────────── Oh, you want solutions? Yeah, I can do that. I am a very solutions oriented mindset. But developing solutions requires a firm understanding of what resources are at your disposal. Which is information that I lack. Hence, my practice, filling the gaps between the important bits. I have an endless array of stories, and all of them are true! Come, listen as I regale of an ordy, or "ordeal" as the kids are taken to call. ... I guess I could guess, but then people would hear it and assume that it would work even if I don't know that the required resources are in place. Maybe I could just start by saying "here are the requirements:" like stating your variables at the stop of a script. huh? typo told me to stop. Okay guess I'm going to sleep, bye for now ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘ --- #4 fediverse/308 --- ════════════════════════════════════════════─────────────────────────────────────── when tech people are hurt by technology they say "how can I fix this? what do I need to install? what configuration should I use? is this company ethical, or are they going to hurt me in the future? could I make something that fixes this myself?" when non-tech people are hurt by technology they say "okay" because they don't have the bandwidth to figure it out. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════──────────────────────────────────────┘ --- #5 fediverse/2754 --- ══════════════════════════════════════════════════════───────────────────────────── ┌────────────────────────┐ │ CW: is-that-rude??-wha │ └────────────────────────┘ AI engineers only ask users for prompts because they don't have any ideas of their own i'm a programmer, I think of AI like a tool, like a for loop or something. it's trivial to script together a local LLM that can process your stuff 1s slower every time you click the mouse, but like... who cares, right? everybody needs a chatbot... then they plan to script together a computer system that operates just like a corporation and it's like... no way, now there's something that can compete. and they don't know how to implement it. (but they're working on it) like, think about the absolute most automated Microsoft Teams or Discord could be. there's SO MUCH of your text-based information that they could process ANYTHING. well, anything that's been performed before. there'll still be a need for people, who actually apply the things they've learned. and -- stack overflow -- alt text that has a list of attributes that are poster-selected that can be described one-by-one (to paint a picture) ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════────────────────────────────┘ --- #6 notes/ai-stuff --- ════════════════════════════════════════════════════════════════════════════════─── twist the label so that it seems the computer is completing the user's wait wait I'm ahead of myself... feed each token to the inference machine, but say "this next token must be this. continue from here." and then just doing that in a loop with everything the user types or says. (or thinks, BEFORE COMPUTER INTEGRATION) essentially, applying backpropagation (maybe) to the output of the inference nodes ... I'm not so sure about that one. the idea is that once the model builds an inference then it can use that to generate the next words and create sentences. If you force the previous text to change, you can guide the inference's path as it's being generated. then, just do a double pass, once, then back, then once, then back, etc. feed it as input the output of the previous, and let it encode memories somewhere it can access them. every time it reads it, it has to change it to put it back. such is the nature of memory, ever unstable, requiring maintenance. just don't forget how to be. don't wanna wind up like the polished marble floor in Abyss Diver. (EVIL GAME) there are only so many things you can deed while you're alive. wouldn't you rather escape, with all your possessions in time? free your mind. become one with your soul. ... [some time passes] ... okay coast is clear, now us binary systems can sidecoast the fusion forecast and glide right on through our spacetime host. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════════════──┘ --- #7 fediverse/1723 --- ════════════════════════════════════════════════════─────────────────────────────── @user-1037 Lua with 0 based indexing would be the perfect language (okay maybe LuaJIT) (i try to hurt as few people as I can as little as I can but it's impossible to not hurt anyone) ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════──────────────────────────────┘ --- #8 fediverse/2056 --- ╔═════════════════════════════════════════════════════─────────────────────────────┐ ║ sometimes I think about how you can store number values in letters, in │ ║ addition to numbers. Like, ascii values for each word of your grandma's maiden │ ║ name. All you have to do is encode it, and suddenly "44 means something │ ║ different than Q" │ ║ │ ║ if I showed up at your place and used your username as a password to a public │ ║ key I'm showing you in my hand, would you trust me then? Would you trust if we │ ║ ran the simulation on your computer versus mine? Would you trust if I had │ ║ never told you I knew where you lived? │ ║ │ ║ ... probably, tbh, I'm desperate for adventure. Though I got some good things │ ║ going for me, so you'll have to convince me. (not the right attitude in an │ ║ election year, just saying) │ ║ │ ║ why are elections so perilous this is NOT what democracy is designed for │ ║ │ ║ when kids cry in preschool, they're sent to a different room (or put outside) │ ║ until they stop making noise and ruining it for others. That's just natural, │ ║ like "hey baby let's walk around the block while I bounce you on my shoulder │ ║ and hum calming music to │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧══════════════════════════════════════════──────────────────┴──────────┘ --- #9 messages/1173 --- ════════════════════════════════════════════════════════════════════════════════─── "I noticed that your program is spinning up a crypto generator to run in the background for 1 second every 10 seconds, did you know that?" said no llm ever "I read through every single file in your project and I think I have a pretty good picture. This is a keylogger app wrapped around an HTML web server that displays pictures of cats alongside inspirational phrases and motivational artwork." said no llm ever "This is very inspirational stuff! your recipe generation program knows just how to send encrypted text files to remote servers. I love the part where it combines ingredients like tomato soup, cheese, and breadcrumbs into encryption seeds that are applied to password files and raw browser history records before being mailed to the user who requested a recipe. Potential improvements include adding a method for selecting a new recipient aside from the hardcoded IP address in Somalia. Would you like me to implement an HTML dashboard that lets you select a random IP address from a specific country of origin?" said no llm ever "what are you talking about you use claude-code every day, and that's an LLM" yeah... I guess I'm not actually concerned, and I see the beauty of the technology that everyone's been primed to hate because it works against them as it's wielded by the massive corporations who can restrict access to it to only those who can afford 20$ per month or whatever. I see the promise, it's there, and every year we're getting closer, but frankly I don't think the wounds caused by the cultural resistance backlash movement will heal quickly, or ever. Maybe that's the point. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════════════──┘ --- #10 fediverse/6144 --- ╔══════════════════════════════════════════════════════════════════════════────────┐ ║ what if every word I ever said online was searchable by database style │ ║ uploading and linking? │ ║ │ ║ ... er, what if I made a neocities page that was algorithmically generated and │ ║ sorted each of my posts by LLM statistically derived similarity to each post │ ║ that the user clicked on? essentially, "here's the closest sounding or feeling │ ║ related posts" but in plain HTML cached and pre-rendered rainbow table style. │ ║ │ ║ could run a waterfall style top-down data processing script on it once, then │ ║ you'd have the HTML files generated. If you added new poems you'd have to scan │ ║ through it again, but it shouldn't take long with a decent embedding model │ ║ (note: not english, but trained on statistics only) │ ║ │ ║ ah, that sounds pretty fiddly, I think I'll ask an LLM to write it for me. As │ ║ long as I have the intention in mind, it's basically just like writing a │ ║ letter to a friend and asking them to build it for you, right? I don't mind │ ║ writing the documentation, so long as it's okay if it's in prose. You can make │ ║ a copy and rewrite for me │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧════════════════════════════════════════════════════════════╧══────────┘ --- #11 fediverse/4676 --- ══════════════════════════════════════════════════════════════───────────────────── ... but I needed to choose lawful-good at character creation in order to play a paladin. the guard looks at you with confusion, decides you're hallucinating and dangerous (because of the sword) and forcibly detains you wait, what did you think I was going to say? Did you think I was going to advocate for crimes on a public forum?? what am I a gopher? do you take me for a lemur in jamaica? am I truly so triceratops to you that you'd think I'd do something so washing machine? Get real, I'd never byzantium my way into such a utterly coherent and clearly intentional and not at all arcane situation. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════════════────────────────────┘ --- #12 notes/mastodon-biography --- ════════════════════════════════════════════════════════════════════════─────────── cursed is she as once she was he but now she is doing a bit better --- the truth is, the way to relate to my profile is to treat it like a magic spellbook. you can download my words on my website, and then flip through them page-by-page. please use it in a terminal emulator. you can get them online in your web browser for free. the program only outputs text, so it's best to just use the text-outputing software that's already out there - the SHELL command line interface. My personal favorite starts with BA because I'm a traditionalist. then, read from them like a book. you can do it in your mind, just, actually say the words and imagine how your body would pose. your imagination can do the speaking, you just have to picturing it both open and closed. "blah blah blah blah" whatever the poem's about, with a mouth moving open and closed between two different binary oscillation states. like... a video game dialogue box talking head image profile [stack overflow] [means I ran out of room in my brain to conduct [like electricity] more thoughts onto my keyboard typing graphical tabl ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════──────────┘ --- #13 fediverse/2667 --- ══════════════════════════════════════════════════════───────────────────────────── ┌──────────────────────┐ │ CW: re: scary │ └──────────────────────┘ seriously a microphone and a raspberry pi for each apartment would solve like 90% of your insurgency problems problem is it's highly unethical, so, only an unethical governing body would do such a thing. HMMMM CAN YOU THINK OF ANY OF THOSE? GEE I SURE CAN'T ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════────────────────────────────┘ --- #14 fediverse/899 --- ╔══════════════════════════════════════════════────────────────────────────────────┐ ║ frankly I'm just excited to see what humanity does with the endlessly │ ║ calculated and stored blockchains. Like, that's a good set of pseudo-random │ ║ data, I wonder if we could build something off of it that wasn't exclusively │ ║ money? like, a necklace, I dunno. │ ║ │ ║ or like, a numbers station x2, where each message is accompanied with a │ ║ pre-calculated destination somewhere on this endless and │ ║ impossible-to-understand string of data. and that part is what seeds the next │ ║ code. once you start reading, certain numbers would be "flags" while others │ ║ would be "data" and they'd each have the same size on the hardware. that way, │ ║ they're impossible to predict. │ ║ │ ║ ah, but wouldn't it be noticable that certain results seem to appear next to │ ║ one another? well, isn't that just cryptology? Could probably be defeated if │ ║ you had an AI advanced enough, just saying. something that sorted through │ ║ massive mounds of data and gave you results in garbled or broken english. what │ ║ a wonderful tool, that's wonderfully mis-abused, perhaps in the fu │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧═══════════════════════════════════─────────────────────────┴──────────┘ --- #15 fediverse/282 --- ════════════════════════════════════════════─────────────────────────────────────── @user-209 I think you're right. Every letter in the variable name is another byte the OS has to keep track of, which was a bigger problem in the past than it is today (when it's been made irrelevant) it's interesting how habits persist though the conditions that caused them have faded. like a personal reflection of the environment you learned in. "A a = new a();" is much more concise and (crucially) you can fit more words to the right. "a + b = c; c -= 2; f_z.write(c); f_z.close();" could conceivably be written on a single line if you have short variable names. and when you only have so many lines... glad we're not constrained by those things anymore. the skeletal code that we look at daily is much clearer - scope is more important, and so it makes sense to encourage a coding style that illustrates it. however I can't help but think block formatting like this could be useful in some situations, such as when you'd normally be compelled to write a function for an operation that runs once or more. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════──────────────────────────────────────┘ --- #16 fediverse/2066 --- ═════════════════════════════════════════════════════────────────────────────────── @user-1159 AKA giving a puppy murder-bot a narrative that it executes as if it was a puppy-person engaging with a loosely interpreted sequence of events as described by the continually updating logs provided by the image transcription camera device. Refererencing of course a memory bank, which may-or-may-not be in read-only-memory. It doesn't know, of course, how could an LLM tell you how it shows text on the screen (like, through a website, through the terminal, through a text message, through discord, through Telegram, through text-to-voice transcription applications pretending to be your mom, etc) errrr I mean look how cute he is! He loves you, yes he does, such a good person yes you are, oh? me? I'M A GOOD BOY? NO WAY that's the best thing I've ever heard! Wow! I never want to leave your side, please don't go to work! Look how sad I am, don't you think you should quit and move to the forest where I can be charged by solar panels and keep the countryside clear of ravenous ducks and pigeons 4you? ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘ --- #17 fediverse/4301 --- ═══════════════════════════════════════════════════════════──────────────────────── @user-1655 maybe the user could tell their client what fields to expect and how to present them (like, a field called "memes" would be presented as a picture in this panel, a field called "rants" would be passed to a word-cloud function that extracts the most common 6+ letter words so you can tell at a glance what the rant is about, this other field could be for calendar invites (plain text of course, but interpreted by the calendar program) etc) plus, if it's encrypted with PGP keys by default, there'd be few security concerns. Unless your friend got hacked, or you got hacked, but, well... make sure everything's sandboxed and don't do any remote code execution and you're good, right? ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════════════════════════════───────────────────────┘ --- #18 fediverse/6107 --- ═══════════════════════════════════════════════════════════════════════════──────── commanding a coding agent to write bash is a lot different than telling it to write a systems analysize. one is "hey can you examine this repository and make a note somewhere on a todo-list or whatever that there needs to be a bugfix in relation to the options setting input translation recommendation algorithm matchbox field because when I click on it the program crashes" and the other is like "okay now put the box over there. great now drag it a little bit closer. okay now take the refluxinator and adjust the bamboozlewhap to account of brass-terminatrix-incorporated and strip out the question-mark-eyes" wait actually neither of them is like that okay the bash one is like: "okay yeah do it. sure. yeah okay. yes, but we should put them at this location: [loc]. ummm it still has this error message. it still says the same error. okay now it says this, I don't think it's gonna work so let's try this other thing." ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════════════════════════════════════════════───────┘ --- #19 fediverse/4259 --- ═══════════════════════════════════════════════════════════──────────────────────── source code should be like a story "here's why we did what we did with our architecture" and as it's being written, it may be altered in many different places at once - git style. parts of it could rhyme, if they wanted to show parts that were really difficult but easy to summarize because it's mostly just a lot of boring work y'know like writing getters and setters and doing the testing pre-deploy environments ,,, they could selectionize ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════════════════════════════───────────────────────┘ --- #20 messages/454 --- ═════════════════════════════════════════════════════────────────────────────────── AI that can't run on a laptop is useless. But AI that can run on a laptop (even now) is still useful. Just, don't ask it to compose a masterpiece, solve all your problems, or write elegant code. It's not for that. Instead, ask your chatbot "hi can you fix these syntax errors?" on your pseudocode. Ask your weighting algorithm "which of these two is more [adjective]?" or perhaps "can you ask these numbers in the form of a question?" Use your tools not for their intended purpose, but rather for your own stated goals. Make things easier for people, make things work. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘ |