i'd go 100% fuck that shit

i'm not a misogynist but women are not good at goal oriented thinking. by design. they were never meant to be leaders, generals or engaged in solitary work. they are wired to need someone to gossip to and they follow whatever the group agrees on.

if society is based on their authority, it's doomed.

hormone pills

also why they are whores

also why i'll happily die lonely than get one of these infected mind virus corpses into my life

the carvalho entity doesn't want to play with others. so they made some crap that doesn't work with nostr and nostr already had so few users that idk what to say about their interest in actual decentralized protocols.

nostr isn't that bad you need to totally reset it. it can be built up from. barely any of the clients are anything but X clones so the field is really wide open. the protocol is a bit retarded in places but you can work with it.

guess you will have fun being lonely on it

women's brains have been fried by propaganda and hormone pills for the last 100 years. even the smarter ones have retardation.

pub key is the gfy layer of idgaf

live feed of video of what features are:

also a live feed of complexity

i have used that site before years ago. soap making is one of my minor obsessions. detergents and commercial soaps are garbage. it bugs me more than most people because i have sensitive skin and autoimmune problems.

i just discovered there is an LLM specifically for #golang - llama-3-1-8b-smcleod-golang-coder-v3

i'm gonna pile in on jetbrains to let #junie use local models after seeing how it does on the local LLM setting with the AI assistant.

well, i'll be nice and share the recipe exactly. i have a balance scale with 40g accuracy and i can do cold process with a blender i have sitting here, i think that is the easiest to get right. i think probably actually i want to get gram accurate scales to do it, 40g seems like too much inaccuracy if i want to practise and do small batches that will fit in my blender, which has about 1200ml capacity.

the recipe i'm using will be useful for euros who can get the ingredients off amazon in europe (i am in portugal so spain amazon is what i use).

i'm gonna make myself do it within the week. just have to figure out where i can buy scales from, i think the china shop in sao vicente has it. good excuse to get on the bike and do a day trek back and forth.

so, i decided. fuck it. i paid for win11, i can use LLM local on this, and goland works just fine on it, and even i can use a WSL2 environment so it's decently familiar in teh shell and i know it will work on a linux server.

blackberries are common all over europe, the british isles and even you can find extensive stands of it all over madeira.

they aren't good to eat until they turn black tho.

i keep on putting it off but i have a kilo of KOH and probably 9kg of belgian refined tallow. i really should pull my finger out. cold process for liquid soap is quite easy. dr bronners is not like it was 20 years ago, and actually liquid tallow soap is better than olive/coconut anyway.

i'm gonna do my research. i can use fresh rain water also. has only spent a little time falling on concrete and running through a vinyl hose into a plastic bucket. probably nearly as good as distilled i figure. the only tricky part is making sure i give it a good margin so it is a little bit superfatted.

soapmaking is awesome. liquid soaps from tallows is something you can literally do without electricity or even anything more than a balance scale, and a pearl lye kiln to make the KOH out of wood.

in my life i have never experienced as good a condition of my skin as when i was using 20 years ago dr bronners.

time to pull my finger out and stop being a baby.

but you will become flabby and weak and be forced to submit to all kinds of indignities and depredations

i remembered now that previously i used to have a laptop with discrete GPU and APU on thet CPU that i had been able to successfully run a windows VM inside the machine and have it have exclusive access to the dGPU via IOMMU

the tricky part is i probably would have to sacrifice the use of my second monitor, probably plug my primary into the APU's output and the dGPU into the secondary... but it's something i'm gonna look into once Jetbrains finally implements the ability to set an arbitrary address for the junie LLM, because then i can run windows in a little VM with like 8-16Gb of system memory and full control of the GPU and then run LM studio or OLLAMA on it and do all my LLM requests to my local GPU, which has 16gb of memory, plenty to run a 7B model fully in memory.

hell, i'm kinda in a mood to do this kind of tinkering. let's see how this works. i even have a paid windows 11 install that i can probably migrate into the VM to use with it, just dump its image on my home folder and voila

i just would miss having the second display for what i mainly use - watching the browser and a two pane for running tests on my relay in development.

still. to be able to get LLM working locally on my dGPU would be awesome. not to mention the other benefit that i would be able to start tinkering with building models.

yeah, intel graphics are really simple and just work. they just are shit slow for 3d.

yeah, probably you only need to give 100gb for each distro then.

running local relays, idk...

i know it's quite easy to do with realy but browsers block access to localhost which can be a huge pain and most web apps that allow local relay access won't let you point it at an IP address or domain name.

i had a bit of a disagreement with hzrd149 over this subject, pointing out that the browsers block this and even with what should be allowing it in the app config in the browser it wouldn't do it. being able to designate an arbitrary address for your "cache" relay would be really handy.

especially because if you set up a wireguard vpn and reverse proxy on a VPS other people can read and post to it (my relay on wss://realy.mleku.dev was being used automagically by other people, including you, because of the magic of outbox model). if the client would let me set a relay as my cache in this way i could have a second one on my machine that just does caching for my app, it would make loading old events a lot faster

VMs usually the graphics sucks. this can be worked around but usually entails modifying BIOS settings and fiddly configuration in the VM app. it is even possible to partition off the video card to dedicate it to a VM, but it was difficult to get it to play nice. with that tool i was able to get Windows to use the GPU dedicated.

reminds me, this might be a good way for me to get around the problem of AMD ROCm GPU compute setup. with the ... i forget the name of the thing for partitioning PCI-express devices, enabled, you can then use the dedicated GPU for a windows or other OS installation even have it run on a separate screen. that was another hassle with it, multi monitor setups can be a real pain for this kind of case. unless you get a hardware switcher for the displays.

i also mentioned that you can and should make a dedicated partition for /home :)

also, pretty much all of the distros i recommended have the option to "try" and in the live system you have Gnome Disks which is practically identical to the one in macOS

i recommend using it to slice up the disk although gparted can be better in some regards.

all distro and home partition just use ext4, don't be tempted to use btrfs, even though it's getting better you can have issues with it

also, in fact you can probably even use merely 100gb of space for each distro, rather than 200. i only recommend 200 in case you do end up wanting to install a lot of apps via dpkg or pacman or whatever

*kinda

hopefully poor susan will get away from her abusive husband

well, in that case, skip that but i personally *paid* for windows 11 so i keep it on my disk. mine is 4tb so it's trivial for its ~200gb of allocated space. windows is better for AMD GPUs for LLM than linux. i probably eventually want to get a dedicated box that just has a relatively small SSD (1tb or so) and GPU on it and move windows to that and run ollama or LLM studio or whatever its called as servers on my VPN/LAN and be able to use them locally

partition it with about 200gb per distro and make all of them use a shared /home mount

i recommend zorin OS but ubuntu 24 and 25 are pretty good. if you want to play with arch, cachyOS is a good option. Pop OS is also good for games but so is Zorin

so, 4

go, forth, be fruitful and multiply*

technically it's "be fruitful" so it's only two :)

yeah, my first attempt was halloween edition in 95 and then in 99 i was running red hat 5.1 with kde (1, i think)

"cRyPtO iNvEsToRs"

can make network services closed, and prevent access that way, but that's it, everything that has to run on a user's computer is going to be open

delicious dilemma for closed sourcers

LLMs will make decompiling a lot easier too... closed source ded

nostr:nevent1qvzqqqqqqypzqvhpsfmr23gwhv795lgjc8uw0v44z3pe4sg2vlh08k0an3wx3cj9qy88wumn8ghj7mn0wvhxcmmv9uq3qamnwvaz7tmwdaehgu3wd4hk6tcqyqtlcvzumwmya3m3gesjckkcdtsvk2dem5l3a33j5ra8qcqelkl8x6xce48

i hate UBI. they basically have had it for a very long time in australia and it distorts the labor marketplace to create permanent underemployed and unemployed classes.

i was in the former category, couldn't get a job because my productivity and qualifications were below the first rung of the ladder. since coming to europe, finding work has been, pretty much, easy.

even if my qualifications were not great, pretty much got into almost everything i tried to find a position in so far.

i'm rather qualified now though, but people have bad taste in programming languages so the positions are a little scarce anyway.

i'm also broken in similar ways, i don't know what happened to you but something in me has gone wrong also, and become a splinter in me that never heals. it makes me do some crazy things but there is some sense to this, i'm sure of it.

yeah, and it makes you laugh the whole way along

it is enough to buy a nice farm somewhere far off the grid and equip it well tho

lol, well, he earned it. but if he wants to remain a public figure, bad move

this would be quite hilarious if an OG did that to spite those clowns

an OG would certainly not just go sell it all at once lol, i mean who would sell it all at once. easier to just coinjoin a bit and go buy a ranch and retire

he looks like the grown up version of this kid

the $5 wrench attack is no less a threat to gold as it is to bitcoin

heads up nostr:npub10npj3gydmv40m70ehemmal6vsdyfl7tewgvz043g54p0x23y0s8qzztl5h currently not able to connect to the forest or my relay.

feelin bad maybe i haven't updated my subscription lol

i don't think it's jumble, sorry

nostr:nprofile1qyt8wumn8ghj7cn9wehjumn0wd68yvfwvdhk6tcprdmhxue69uhhg6r9vehhyetnwshxummnw3erztnrdakj7qgjwaehxw309ahx7um5wgerztnrdakj7qg3waehxw309ahx7um5wgh8w6twv5hsqgrucv52prwm9t7ln7d7w7l07nyrfz0lj7tjrqnav299gtej5frupc2udgad your relays are ded

when you make a nostr identity, you need to store the nsec somewhere safe

they make it a little confusing in alby how you put it in there, but it's in the nostr settings section. once that's in, you will log in as you mean to. the flow they have is designed for onboarding, not for newbies who already have made a key pair before (nsec/npub)

yeah, pretty much the main thing i'm interested is:

1. training my own models, so i can train it on a data set that i think is reasonable (probably nostr events and programming documentation, and philosophy/religious/spiritual texts)

2. using coding agents, which requires a model that has a lot of computer science trained into it, so the models i make in 1. will be good for this.

what i would do first, is get the agent i now have to write a feeder script, that would pull a cache of all of the kinds of data i want, from archive.org, from gutenberg project, and a shitload of golang github projects, focusing on religion, philosophy, mathematics, cryptography and computer science, and then feed that into a training process to build a model that will be an expert in all these subjects, and perfect for doing coding agent stuff, because it would be go native, most of the agents barely have a grasp of Go. i want the LLM to know how to write correct fully compliant and idiomatic Go without any of the stupid mistakes. the agent i have now it is always forgetting to add or remove imports and make sure every symbol is being used as these are errors.

and then i would probably work on writing an agent that i enable to rewrite files in a prescribed directory and would be able to get it to write and debug code, by designing a scheme for taking a question and turning it into a plan for solving it, and updating the plan as the information changes the required steps

it's not a hard core bucket list thing for me but i can see myself having a dedicated rig for doing LLM work in the near future.

a lot of people did.

but the content such people put out is extremely droll.

nostr is just a protocol. xitter style global is just the first application that has been commonly implemented. alternatives are appearing at an increasing rate now. bitch@ is a nice new one, #jumble is coming along well, i basically either watch my follows or the forest.

i don't want to meet in a global town square. way too many assholes in the global town square.

i like it that most japanese have their own nostr

i think most of nostr should be constrained like this

i didn't come here for a buggy version of X

ecash is paper bitcoin. the end.

yeah, i would love to train models, damn...

lol, you know the manhattan project computers were barely more than calculators right? supposedly you can't model nuclear fission with a modern desktop CPU now even though it's literally squillions times faster.

anyway, fuck those people. if deepsnek could do it, so can anyone. also, fuck closed source AI. who knows what shit they are sliding into those things.

yeah, i'm probably gonna go green for my next video card, i want to have local LLM

i use Zorin OS, it's nice, like Pop but has a better theme system and supports AMD hardware well

my first KYC experience was shotgun. i deposited to MTGOX and then i couldn't move or sell it.

i avoid it like the plague. any sane person would avoid it if their first experience was having their own money being held hostage in exchange for anything. the ransom being your personal information is even more fishy.