Fedora Silverblue + Distrobox (with BoxBuddy) + Flatpaks + AppImages (with Gear Lever) + Waydroid + Bottles

install every dev tool to a distrobox and export them using distrobox export. create distroboxes with their own home folders, so they don't trash your home.

use appimages with portable home folders so they don't trash your home folder.

bottles for non-steam windows games.

waydroid for android apps if you need them.

flatpaks for the rest.

i was expecting them to sacrifice him only. after putting all of the blames on him.

i talked about these many times before, but im gonna talk about it again...

because... why not.

the most important thing about the bitcoin blockchain is the decenterlization. decenterlization must trump everything. and we should never sacrifice decentralization, even a little bit of decentralization, to achieve something else. because if the bitcoin blockchain is not decentralized, what is the point of all of this? what are we all doing here? right?

so how do we achieve decentralization? well to achieve decentralization, everybody should run their own full node. right? but it doesn't happen by saying "oh you should run your own full node". it doesn't work like that. right? because running a full node by itself should be easy. right? you should be able to run a full node not only on your server or pc, but also on your phone. right? that's the point of bitcoin, everybody having a copy of the blockchain. so they can verify it themselves.

when you can't trust people, you can trust the bitcoin blockchain and the network. bitcoin blockchain is the final source of truth globally.

its purpose is not to be something that people use for anything, its not the internet. because if you try to do other things with it, you will have to sacrifice decentralization. and you should never sacrifice decentralization for anything.

more data = less decentralization.

the point of bitcoin and its decenterlization is that so you don't have to ask anyone the state of the blockchain, you can verify it yourself. the point is that we have a common global source of truth copied by everyone, so we can all agree on the state of things.

if everybody is going to have a copy of this blockchain, then it should as small as possible. right? in an ideal world, the whole blockchain should be maybe 1GB rn, right? but in a real world we should have some balance, some middle ground.

satoshi decided that every block should be 1 megabyte. (if you asked me it could have been even smaller)

some might say "but bitcoin needs to scale", but the thing is bitcoin can already scale with layers. so if we are suppose to keep the blockchain small so everybody can have a copy of it. then we dont scale by having more txs and more data in the blockchain. we scale by having layers on top of the blockchain. like lightning network, liquid network, etc. and when someone tries to fuck you on the lightning, you can see that on the bitcoin blockchain, the global source of truth and stop them.

more data = more centralization.

if you can't stay decenterlized at the same time, scaling has no purpose.

and the truth is there is nothing wrong about keeping some spending money a custodion lightning wallet. and there is nothing wrong about using fedimints or cashu. until we find something even better without sacrificing decentralization.

so in the genesis of bitcoin, the blocks were 1 megabyte. but also at the genesis of bitcoin, we had some problems. some logical errors with the way we hash the tx and the scripts. so we created segwit to solve this issue. the segwit soft fork. it was using a new space to hold the script data. which is fine and makes sense. but for some reasons we also thought making the block size 4x bigger in the same fork was a good idea...

and no if we still had the 1 megabyte block size, they wouldn't all be filled to the brim with transactions. because nobody is waiting in line to pay tx fees while the network is congested. everybody would have used things like lightning more.

so now we have to live with this and find a way to make it work. with 4MB blocks...

and so far the bitcoin core repository showed zero effort to make bitcoin more decentralized. they are only focusing on adding more features. they care about what bitcoin can do for them. not what a node runner needs to run a full node. they see bitcoin as something work in progress, not a stable foundation that you can confidently build on top of.

and thats why i can say "they lost their way", because they didn't show any effort to sacrifice anything to make bitcoin more decentralized. they always bit by bit sacrifice decentralization... for other things.

and that's why we need protocol ossification. we need more bitcoin node/client implementations. not only forks and cli tools, but software for the end user, build from scratch. a software that prioritizes making running your own full node easier.

i mean if you are only caring about full nodes, size of utxo set doesnt have to matter based on the bitcoin node implementation. you dont have to store it separately, you can just point to the tx and the output you are spending.

also we are already gonna make the bitcoin divide into more units at some point (i know LN already does). but also this one is being centralized kind of an issue.

i was just saying it was a nice quirk that helped, i was kinda thinking about how blocks are gonna become more and more empty as the bitcoin price rises more.

i mean 1 sat minimum cap was kinda keeping the post segwit blocks smaller. now they will just fill them up. it will be harder to download the blockchain hence making it less decentralized. idk 🙃

https://youtu.be/1UsAqvooYj4

wait a minute

can also have a non DATUM endpoint

can we make a DATUM mining pool, that let's you keep 1btc but distributes the rest?

for solo miners.

i mean if you don't know anything about coding, its amazing.

but i don't understand how people trust it with a whole project at all. it can't even fix bugs with tests.

i think people are just using it for basic stuff instead of real logic, or not aware of the issues it creates.

or im just doing it wrong?

but i find it useful for creating project a skeleton for things im new at and don't know anything about. i make it create the skeleton, and even though i have no idea about the thing, i can understand what its trying to do, and refactor and rewrite it myself.

ocean block health from the perspective of the knots vs core

every time i try to code with ai, i find myself unproductive, because thing cant build anything meaningful.

if im gonna code with ai, i want it to work slow or act like its typing like a human, so i can watch it type, and tell it to stop.

i dont want it to accept generic commands, i just wanna describe writing the code. like:

- "add this function in this file"

- "change this part to that"

- "can you open this file, and scroll to x"

- "ok there see that, we dont want it like that, instead do this, and refactor the rest of the code accordingly"

- etc.

basically monkey coding ai

and IT HAS TO BE WITH VOICE

so when im tired i can just tell it do to stuff.

he is gonna cause armageddon just to see it fail

not because im gonna make it, but i thought but how would i make a gambling web app using cashu before

https://blossom.primal.net/514662f87d174d3bf0ee84b818d0746d5020d28c3ebe8497a1ff15ca9da0d76f.mp4

if we removed the spam we would make less than 1mb blocks common again. help to make bitcoin more decentralized.

demons got him...

oops revealed my timezone, meh

knots go up

running knots but also trying to make my own node implementation. the ultimate bitcoin software. idk.

writing it from scratch. it has been going well so far, but there are probably some edge cases im missing and some errors. we will see.

i mean i don't think it will be fully ready and stable any time soon. but it should be good enough to make it public and show it off here soon, before the summer ends.

oh it just has this small progress bar at the bottom. THAT EXPLAINS NOTHING

i think they didn't get the question or just acting dumb on purpose.

if you are not using knots and just using an older version of core, know that taproot is still a thing, and core doesn't have filters/patches against exploits that came with the taproot fork.

dont get too excited yet. we will see. i need to implement it first.

ok i was doing some thinking for the last 6 hours and i think i can store the whole blockchain (indexed) under 100GB lossless, on my node/client implementation.

idk i need to test, and tweak if needed.

knots is more profitable long term tho

what is that?

im sorry