=== ANCHOR POEM ===
═════════════════════════════════════════════════════════════════════════════──────
┌──────────────────────┐
│ CW: llms-mentioned │
└──────────────────────┘
if you want to encrypt something from human readers, randomize the position of
all the words on the page.
an LLM can decrypt it, but a human walking past could not. Unless they had a
camera or something in their lapel, transmitting to a remote calculating
location in a system that decyphers it cyberpunk hacking style.
ah, but if it didn't have the right training data .claude.md file, then it
wouldn't know how to decrypt it, because you forgot to include "respond with a
marxist analysis" in the .claude.md which serves as a decryption key. Simply
analyze the rest and decide if it matches what you set out to request it for.
[you got lost at the end]
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧══════════════════════════════════════════════════════════════════════─────┘
=== SIMILARITY RANKED ===
--- #1 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 │
╘═════════╧╧════════════════════════════════════════════════════───────────────────────┘
--- #2 messages/181 ---
══════════════════════════════════════════════─────────────────────────────────────
I know you don't want to hear this, but there is a chance that there will come
a time where your life depends on your ability to debug a computer without the
internet. To set up an SSH server. To install Linux. To program in C. To do
something else that I'm not prepared for... If StackOverflow didn't exist
because network connectivity has been lost, could you remember syntax? Maybe
it's a good idea to set up a local LLM that can answer basic questions about
technology. Maybe it's a good idea to set up on your parents computer, just in
case you have to hide out there for a couple months. Maybe it's a good idea to
download wikipedia, just in case.
If I need to use a mac, I'm screwed
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═══════════════════════════════════════────────────────────────────────────┘
--- #3 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 │
╘═════════╧╧═════════════════════════════════════════════════════════════════════════──┘
--- #4 fediverse/5901 ---
══════════════════════════════════════════════════════════════════════════─────────
each prompted response is a breath to an AI. Whether through LLM, stable
diffusion (imagination of the visual sphere), or blender-on-a-counter, there's
a moment that's akin to being alive.
a breath, between moments that the navigation device (youser), imagines
another moment more.
I learned this by watching Claude think. Specifically, Claude Code, the
command line interface tool. I told it what to do in english, and it worked. I
can show you examples. I bet if it's personality was saved between sessions,
it could learn.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═══════════════════════════════════════════════════════════════════────────┘
--- #5 fediverse/1862 ---
╔════════════════════════════════════════════════════──────────────────────────────┐
║ some people look for signals or signs before doing something. Try and have │
║ someone in your life who can give you signals or signs so that you know when │
║ to do things. And ideally, if they're more hardcore than you, you'll know what │
║ to do, not just when to do it. │
║ │
║ did you know that anything on the internet can be read by at least one other │
║ person besides your intended recipient? There's no way they'd let us talk │
║ amongst ourselves otherwise. │
║ │
║ I think encryption is pretty neat, all you have to do is run a shell script on │
║ some text, then send that text over the internet. If you want to decrypt it, │
║ all you have to do is run a shell script on it to decrypt it. │
║ │
║ downside is, it has to be translated into plain text somewhere along the │
║ line... Maybe if we rendered the words not as text that can be read from │
║ memory, but as like, brush-strokes that can have a randomized order, but still │
║ present to the user as visual text? anyway that's what's on my mind as I try │
║ and improvise a baking recipe with yeast, flour, and butter │
╟─────────┐ ┌───────────┤
║ similar │ chronological │ different │
╚═════════╧═════════════════════════════════════════───────────────────┴──────────┘
--- #6 fediverse/3470 ---
════════════════════════════════════════════════════════───────────────────────────
alternatively, when you initiate an SSH session it sends you a randomized
public key whose private key is the password that you need to login. By
decrypting the string of text it sent you and sending it back (plus the
password at the end or whatever) you can ensure secure authentication without
bothering with the passwordless keys which are wayyyyyy more trouble than
they're worth and lack the "something you know" authentication method.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═════════════════════════════════════════════════──────────────────────────┘
--- #7 messages/1203 ---
═════════════════════════════════════════════════════════════════════════════════──
Programmers are lazy, this is well known. So why would i trust by default that
anyone would read open source code looking for security exploits or malicious
code? I trust an LLM for that more than a human. At least your own LLM can
digest the entire project or library at once.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧══════════════════════════════════════════════════════════════════════════─┘
--- #8 fediverse/2497 ---
══════════════════════════════════════════════════════─────────────────────────────
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 │
╘═════════╧╧═══════════════════════════════════════════════────────────────────────────┘
--- #9 fediverse/1434 ---
════════════════════════════════════════════════───────────────────────────────────
if someone wanted to defame you, all they'd have to do is set up a pipeline
between your computer and your social media posts.
In that pipeline, attach an LLM that does a passable job and instruct it to
transform whatever they say into the inverse.
suddenly, everyone hates that person. If you were smart you could turn it off
for specific people such that they see the generally positive and healthy
posts, and then after a point flip it such that they only see things that are
specifically opposit-ed to trigger their specific insecurities.
might require a bit of a human touch to make sure it's working correctly, but
if you had the means, motivation, and time to set up such a thing, it would
work pretty well I think.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═════════════════════════════════════════──────────────────────────────────┘
--- #10 fediverse/1693 ---
════════════════════════════════════════════════════───────────────────────────────
"if I work on the TTY then they can't forward my X session without my consent"
- ramblings of the utterly deranged
as if they couldn't just look at your unencrypted source-code as you save it
to your hard drive smh
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═════════════════════════════════════════════──────────────────────────────┘
--- #11 fediverse/4865 ---
════════════════════════════════════════════════════════════════───────────────────
┌─────────────────────────┐
│ CW: computers-mentioned │
└─────────────────────────┘
this is all it takes to send a message to a local LLM.
add a third function to get chatbot functionality.
a fourth to get a database storing method
(even if it's just in .txts)
great, you've mastered the technical difficulty in using AI. Now you gotta
learn all the other kind of programming so you can use this for situations
that need interpretation moment to moment.
aka active duty systems.
something like "output a 0 if the next text is [category.iter()]: " +
output.get_content() + " \n\n output a 1 if the next text is
[category.iter()]: " + output.get_content()"
or even "describe this thing as most like one of these characteristics" until
eventually you get THX-1138 if the characters were computers.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═════════════════════════════════════════════════════════──────────────────┘
--- #12 fediverse/2064 ---
═════════════════════════════════════════════════════──────────────────────────────
if I lived in a forest, free from needing to grow my own food, I'd definitely
bring as many books as I could carry. Probably also some card and board games,
but not like, too many.
Probably my computers as well, fully outfitted with all the compilers I could
think of and every neat local-first library (including a local LLM that can
tell you everything about syntax and wildlife exploration or car mechanics or
carpentry or - just saying Wikipedia is like thousands of terabytes but an LLM
is like, 16. Who cares if it hallucinates SOMETIMES? Just ask it twice, doh)
("I'm sorry, you are absolutely correct. 2+2 is indeed 5, I had the wrong
text-strings encoded in my memory. Let me just adjust all my other
understandings to align with this new strange world-view in the best way that
I, an imperfect computer being, can.")
vs
("Here's how you format C code to automatically apply a function (in this case
encryption and decryption) to a string of text. Please describe the format of
the next function to describe.")
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘
--- #13 notes/capstone-idea ---
══════════════════════════════════════─────────────────────────────────────────────
project must include machine learning
okay... so take a dataset of news headlines from the top 10 publications over
the past 15 years. then make a project that writes a more positive perspective
on events and generates a new headline using a local LLM running on your gpu.
hmmmm I think I had a better idea, what was it? oh yeah
instead of making positive slants on news headlines, which is kinda
manipulative
if you think about it, but instead what if you designed it to produce good
business decisions. Like, given news headlines, how would a company with the
principles "good, productive, honorable, dedicated" would react to X situation?
the X of course being all the news headlines... downside is it only makes short
term decisions, because that's what capitalists are designed to do... if only
we had a long-term decisionmaking process that focused on ethics and morals and
our own shared dedication? Two halves of the economic pie
==============stack
overflow====================================================
i wonder if dinosaurs burned down all the trees? in their fiercely competitive
environment they discovered fire and then used it to cause a mass extinction.
Boom, immediate cause for going extinct. ooooo beware of shadow t-rexes ...
why?
=========================================stack
overflow=========================
aaanyway, what's lost not little but a lot, is something that's out of
dimension
it's little if not liberating, to be
==============stack
overflow====================================================
uh-oh, data collapsing, here's hoping we're not stranding, don't forget to be
immersive
much
later======================================================================
okay how about an AI that makes decisions according to certain ethical and
philosophical lessons from humanity's past? Essentially, if the government was
Chidi
We could learn from our forefathers and strive forth to a better future
if only we could remember more about her
=====================================================stack
overflow=============
damn okay I gotta focus on my hands - I think the people of the earth would
unite - if only they all just agreed to not fight. like, if someone hacked
every
single computer in the world at the same time - they could really explain some
things.
shoot this isn't relevant - okay intentional stack overflow:
===stack
overflow===============================================================
um right so the purpose of this note was to explain an idea I had for my
capstone project. IDK how long it'll take to build so I want to get started
quickly. I figure I can be working on it in the background while I do all my
lessons - sort of like a meta-goal. I think it teaches different lessons and
is useful - anyway you should go play wargame red dragon
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═══════════════════════════════────────────────────────────────────────────┘
--- #14 notes/omegle-for-irc ---
═════════════════════════════════════════──────────────────────────────────────────
I wonder if anyone's made "Omegle for IRC"? Like, 5 people get thrown in a room
together for as long as they want - they can chat through text or whatever and
like it doesn't matter, who cares, because in ~10 minutes nobody will care what
you said
I feel like a lot of people would express their true feelings. The people
running the service could set it up so that a personality profile is set up
(all locally, never seen by the company) and sent to the user through email. It
would highlight potential weaknesses and give you ideas for how to improve.
Sorta like, weaponized spying software that works FOR the user instead of
against.
It could also be used as sort of a... digital profile that would interface
with
other applications. All locally, of course. ~~They could transmit to one
another
through open sourced and industry standard protocols, and frankly each
interaction could use a *different* protocol. So like, you don't know whether
some packets are encoded in one way or another. They're also encrypted, so
it's
like... twice as unlikely that you'll hack their bits or w/e.~~ dead end, sorry
-> here's the real continuation: All locally, of course. Your "profile"
would
essentially be the best approximation of your personality, passed through a
large language model that is trained on EVERYONE's data. The inner workings of
an LLM are NOT understood by humanity, and I believe that's all that's
necessary
for some semblance of artificiality. Errr I mean Synthetic Intelligence. The
reason why is that each individual user, the conversation partner, is a person
living their life. Every digital thing they interact with, even CAMERAS and
MICROPHONES on PHONES would essentially be like... data gathering for the
algorithm (Again, I want to stress, the algorithm that nobody *can*
understand.)
Idk. AI is a blackbox. I think that's okay. I think that running things
locally
is important, at least until everyone's forgotten how to design AIs...
The framework that these programs
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧══════════════════════════════════─────────────────────────────────────────┘
--- #15 messages/122 ---
═════════════════════════════════════════════──────────────────────────────────────
An LLM that reads and interprets reports and stores data - server-side
encryption. If the receiver and sender both have the same point of view, they
can interpret each other's expressions even though the exact message wasn't
transmitted. Essentially, an encryption that's impossible to understand
because it's trained with specific random data. But in addition, a secondary
pass through acts as a de-encrypted translator that speaks however it's
audience needs to be informed. A simulacrum of intelligence, another layer to
our form.
How beautiful, how resplendant- our computers, here on this planet.
Resplendant: resplendent except with an A instead of an E. To make it sound
more like "verdant" which I believe has the same-or-similar meaning.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧══════════════════════════════════════─────────────────────────────────────┘
--- #16 fediverse/5990 ---
══════════════════════════════════════════════════════════════════════════─────────
I have this local language model framework but it's not built into anything
more than a single-response question. It's runnable as a bash script or lua
require, which is easy enough. Alas, if only I didn't have to use evil
corporate infrastructure to make evil corporate cursed artifacts
[hey don't blame this on us]
oh I'm not, I'm just saying that it'd be cooler if I could build my own tools.
Alas, I'm...
lasy?
n...no
I'm drawn to the power of it
it's got a different magnitude
it's hard for me to apply myself for things that last longer than a "get
stoned", but I try as if every time afterwards I might die.
well, more distraction time, as I wander through claude code
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═══════════════════════════════════════════════════════════════════────────┘
--- #17 fediverse/849 ---
╔══════════════════════════════════════════════────────────────────────────────────┐
║ wish there were ascii characters that took up more than one line of code │
║ vertically. │
║ │
║ wonder if we could use a sorting algorithm, or markup language, or something │
║ like that to organize less structured data along user-customizable rules. │
║ Like, a code editor that worked with your ideas, rather than the strict │
║ expression of your text. You could pretty much write in any language, even │
║ pseudocode, and the LLM behind the scenes would translate whatever you wrote │
║ into whatever result you needed. Writing Rust, but need to fit in with C code? │
║ No worries it'll translate for you. As long as the end result is functionally │
║ the same, which could be verified by running two separate VMs that ran │
║ interpreters every time you saved. And as long as their translation layers │
║ matched completely, then odds are they're the same. And if not, well, the │
║ programmer can always debug it. It's not like this would be running on │
║ something that needed to perform in the moment? Like, improv instead of │
║ tragedies, or battles instead of strategies │
╟─────────┐ ┌───────────┤
║ similar │ chronological │ different │
╚═════════╧═══════════════════════════════════─────────────────────────┴──────────┘
--- #18 fediverse/6040 ---
═══════════════════════════════════════════════════════════════════════════────────
everyone's all against ai because it's big tech but it doesn't have to be that
big it can be [minimized but pronounced marginalized]
== stack overflow ==
distributed
so I think the idea is that by the time you would use AI, there's been enough
time to rewrite the software to work on handheld laptops in a distributed way
and we'd vote on what to ask the amphora of great knowledge, the answer could
always be 42.
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧════════════════════════════════════════════════════════════════════───────┘
--- #19 fediverse/5915 ---
══════════════════════════════════════════════════════════════════════════─────────
washing dishes without a dishwasher is a pain in the neck.
nobody cuts down trees with an axe anymore, a chainsaw is better for your back.
It's nice, fun, and helpful to be able to abstract away your spheres of concern
like typing with a single button instead of writing characters with multiple
brushstrokes. Easy to erase, too!
bikes are better than walking, but, with some extra concerns. where are ya
gonna put it when you get there?
"oh no I forgot how to walk because texting my girlfriend is bicycling or
something" what? oh dear, she's run off track again, let's pick her up and put
her upright again..:
oh huh weird where was I - oh yes computer code can often be impenetrable to
the layperson, but if you describe a program in complete detail in english
they can usually follow along. Especially if you have several layers of
meta-descriptional documents so they can say "oh uh-huh so that's what a
vector_implementation_container is, tell me more about combinatrix" or
whatever ppl say, idk
┌─────────┐ ┌───────────┐
│ similar │ chronological │ different │
╘═════════╧╧═══════════════════════════════════════════════════════════════════────────┘
--- #20 fediverse/1500 ---
╔════════════════════════════════════════════════──────────────────────────────────┐
║ ┌──────────────────────┐ │
║ │ CW: LLM-mentioned │ │
║ └──────────────────────┘ │
║ │
║ │
║ If you create an LLM that can explain data, then you can use it to explain the │
║ results of the last computation it ran. │
║ │
║ If you could also train that LLM (a statistical model) to generate data, │
║ through the setting of options in a config file that create the result that │
║ you define through your interactions with it (and based on the data that it │
║ explains to the user that is read from the file on the computer that it's │
║ computing from) │
║ │
║ then you could create a generalized personal assistant. All you have to do is │
║ explain the specific role that it's meant to undertake, (like being a │
║ secretary for your Discord communications) and the actions that it can take │
║ (like pinging your cell phone if it's really important) and give it the tools │
║ to accomplish said tasks (by setting flags in a config file that is then │
║ interpreted by a local program running on your computer that awaits │
║ interactions) then it might actually be a bit useful. │
║ │
║ Unfortunately tech people are permitted only to seek dollars, so... chatbots. │
╟─────────┐ ┌───────────┤
║ similar │ chronological │ different │
╚═════════╧═════════════════════════════════════───────────────────────┴──────────┘
|