PupBiru

i don’t agree with that definition of creative… there’s lots of engineering work that’s creative: writing code and designing systems can be a very creative process, but doesn’t involve feeling… it’s problem solving, and thats a creative process. you’re narrowly defining creativity as artistic expression of emotion, however there’s lots of ways to be creative

now, i think thats a bit of a strawman (so i’ll elaborate on the broader point), but i think its important to define terms

i agree we should be skeptical of marketing hype for sure: the type of creativity that i believe ML is currently capable of is directionless. it doesn’t understand what it’s creating… but the truth lies somewhere in the middle

ML is definitively creating something new that didn’t exist before (in fact i’d say that its trouble with hallucinations of language are a good example of that: it certainly didn’t copy those characters/words from anywhere!)… this fits the easiest definition of creative: marked by the ability or power to create

the far more difficult definition is: having the quality of something created rather than imitated

the key here being “rather than imitated” which is a really hard thing to prove, even for humans! which is why our copyright laws basically say that if you have evidence that you created something first, you pretty much win: we don’t really try to decide whether something was created or imitated

with things like transformative works or things that are similar, it’s a bit more of a grey area… but the argument isn’t about whether something is an imitation; rather it’s argued about how different the work is from the original

that’s a lack of understanding of concepts though, rather than a lack of creativity… curation requires that you understand the concept that you’re trying to curate: this looks more like a dog than this; this is a more attractive sunset than this

current LLMs and ML don’t understand concepts, which is their main issue

id argue that it kind of does “think about its own thoughts” to some degree: modern ML is layered, and each layer of the net feeds into the next… one layer of the net “thinks about” the “thoughts” of the previous layer. now, it doesn’t do this as a whole but neither do we: memories and neural connections are lossy; heck even creating a creative work isn’t going to turn out exactly like you thought it in your head (your muscle memory and skill level will effect the translation from brain to paper/canvas/screen)

but even we hallucinate in the same way. don’t look at a bike, and then try and draw a bike… you’ll get general things like pedals, wheels, seat, handlebars, but it’ll be all connected wrong. this is a common example people use to show how our brains aren’t as precise and we might like to think… drawing a bike requires a lot of very specific things to be in very specific places and that’s not how our brain remembers the concept of “bike”

it’s only qualitative because we don’t understand it

when an LLM “experiences” new data via training, that’s subjective too: it works its way through the network in a manner that’s different depending on what came before it… if different training data came before it, the network would look differently and the data would change the network as a whole in a different way

and experience is ongoing learning, so if an LLM were training on things after the pretraining period then that’d allow it to be creative in your definition?

but in that case, what’s the difference between doing that all at once, and doing it over a period of time?

experience is just tweaking your neurons to make new/different connections

yes and no… we don’t have revenue goals, but we have goals for the fediverse and we have the social media critical mass problem: you have to hit a critical mass before you become indispensable… if people try the fediverse and there’s not enough content, they tend to just leave rather than stick with it

mastodon can do default instances because they have the account migration process… i totally agree this is a great solution: get people in with sane defaults, and then let people move once they know how it works

there will be plenty of people that don’t move (or maybe that’s solvable too: analyse your toots and suggest a more niche instance after 2mo?) but i’m not sure that’s a huge problem if your “default instance” is more of a random choice from a list of sane defaults

it’s kinda vaguely similar though… a fediverse instance is moderated by the instance admins, just like a discord server (though discord has a level of admin above server mod/admin i’m not sure that distinction matters for the general user)

at the very least constant whip-lash from changes might see ISPs not being able to sign long-term contracts and businesses not being able to plan around availability of things like “fast lanes”, which might make them uncommon even if net neutrality keeps getting repealed

what apple offer is on no iphone ever has anyone had to worry about how much RAM their phone has… its literally irrelevant: it could be 100Gb, or it could be 1Gb for all i care… its just not something that needs thinking about

i mean… to deny the similarities in the platform is kinda ridiculous… i’d go so far as to say that lemmy/kbin probably wouldn’t have the format it does without reddit being like it is. there’s a direct relationship, whether we like it or not

“even”… i think reddit’s follow ups were least likely to succeed because they were the least organic!

place was special (at first) because it was a bunch of people with no time to plan and prepare… it was humans doing human things with all the limitations and creativity that came with that

other iterations of place were bots and pre-planned “stake our claim” rather than making something interesting and different… it was kinda all just the same as the last one but with subtle “meme of the moment” tweaks

of course nobody is going to commit to supporting a stable API for 10 to 20 years… that’s expensive as heck and not even remotely worth it!

there’s nothing “wrong” with software development, it’s just that consumers demand new features rather than stagnation… i sure don’t want to be using a 20 year old app because we’ve come a long way in 20 years in so many regards

in 2003, windows xp was still microsoft’s dominant OS with vista still being several years off, half life 2 was about to be released, gmail was allllmost ready to release, msn messenger was still in its prime

yeah no, ill stick with rapidly changing technologies rather than sticking to that for some misplaced sense of “stability”

choice is king

for some things, sure… but for other things i’ll take snapshots, dedupe, etc… i’ll put my system disk on a fancy new filesystem and keep my user data on something more mundane for sure! that allows snapshot and restore of major upgrades with basically no effort

i’m confused… are you talking about ext4, xfs, zfs…? because these are the filesystems linux people talk about and these are also the filesystems that run the worlds databases and data storage systems

that would actually be great

from what i understand, the big change in chat-gpt4 was that the model could “ask for help” from other tools: for maths, it knew it was a maths problem, transformed it to something a specialised calculation app could do, and then passed it off to that other code to do the actual calculation

same thing for a lot of its new features; it was asking specialised software to do the bits it wasn’t good at

are you saying that it’s weird that lemmy users can’t see the votes, or that kbin users can?

i’d agree that it’s weird that lemmy users can’t see votes, however i can see the argument:

people have said that their votes should be private, and that if votes can’t be private (let’s make that assumption for the moment because that’s the way the core communication of the network works) then at least it shouldn’t be trivial to get access to that information

that’s a valid position to take

the counter is that if the information is public, simply making it slightly more difficult to access only hurts the people who wouldn’t use that information for nefarious purposes anyway… and by hiding the information, you simply make people less aware that the information is actually public!

lemmy users tend to be unaware that votes are public information, whilst kbin users seem to have more of an understanding of that. i’d say that knowledge helps people act accordingly

there’s also the “let’s change the protocol so votes are truly private” discussion, which is also valid but a far bigger effort, and is also against some of the values of the fediverse (that everything is transparent)

i mean fingers crossed for that so that maybe we don’t have to use facebook events

… but also fuck meta

docker is a PITA for HASS… theres some non-trivial networking to get right so that mDNS works correctly, and my HASS homekit bridge kept dropping off for unknown reasons until i moved out of docker

physical or VM is just much easier and more reliable in this case

home assistant in docker is definitely not for the feint of heart! the networking requirements are actually quite intense, and really don’t map well to virtual networks like dockers uses

… among other issues

HAOS on a pi; i’ve tried the docker thing time and time again, and the next chance i get in blowing it all away and starting on real hardware again