this post was submitted on 13 Nov 2024
955 points (98.1% liked)
People Twitter
5226 readers
2203 users here now
People tweeting stuff. We allow tweets from anyone.
RULES:
- Mark NSFW content.
- No doxxing people.
- Must be a tweet or similar
- No bullying or international politcs
- Be excellent to each other.
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
What problem does blockchain solve?
Having too much electricity and not enough CO2.
We recently developed AI for that purpose though which does the same thing but is useless in occasionally funny ways.
Apparently, it can be very secure. If “pieces” of a secure key are stored in multiple places, for example, only changing one link in the “chain” means it won’t match with the others. They ALL have to be changed at the same time, which is virtually impossible to do in secret.
Please note that I am far from an expert on the subject. I’m paraphrasing an article I read months ago.
Can’t you takeover a blockchain by owning the majority of a block chain, or by having a majority of the processing power to compute hashes?
Yes which is part of why the major chains are owned and controlled by companies, but then that makes the whole thing pointless. IMO, a company controlled blockchain may as well just be a DB cluster, it would be faster and more efficient.
Are you saying that they “solve” that by never giving up more than 49% stake?
That… seems like a bad solution
Those things sound possible, but I’m not knowledgeable enough to speculate. Sorry.
If you had 51% of the world's computing power (to blockchains using proof of work) yes you could forge records, from what I could wrap my head around about blockchains.
You don't need 51% of the world's power though, just 51% of the power of people who care about how the system works. Most people using block chain cryptos don't care at all, so the threshold is a tiny percentage of the user base.
Yeah you're right. I was thinking specifically Bitcoin and the astronomical amount of compute power that's behind it.
That’s proof of work. Proof of stake is you just need more than everyone else, right?
It works more like loaning money and then receiving interest, except you are loaning crypto to the network and then you get it back, plus some, after a certain period of time
Is the network not considered a third party
This would just create a fork in the blockchain where 51% of the network doesn't match the correct state of the blockchain that the 49% have. The 49% would effectively stop working because they could never validate the transactions that the 51% takeover has falsely created. The node operators of the 49% of the network would need to reach consensus for how to deal with the problem, but essentially they would just adopt code that ignores the 51% data, so they could continue to process blocks of transactions. Without manual intervention the 49% would be frozen. The 51% is just fake, they haven't really changed anything because every real node operator would know it's false data.
What if the 51% have already completed the consensus process?
Essentially, verifiability (the token exists on the blockchain), de-duplication (each token can only exist once on the blockchain), and proof of ownership (only one account number can be associated with each token on the blockchain). There's nothing wrong with this idea in a technical sense and it could be useful for some things.
But... the transaction process is computationally expensive. For the transaction to be trustworthy, many nodes on the blockchain network must process the same transaction, which creates a whole bunch of issues around network scaling and majority control and real-world resource usage (electricity, computer hardware, network infrastructure, cooling, etc).
And beyond that, the nature of society and economics created a community around this unregulated financial market that was filled with... well, exactly the kind of people you'd expect would be most interested in an unregulated financial market - scammers, speculative investors, thieves, illegal bankers, exploitatitive gambling operators, money launderers, and criminals looking to get paid without the government noticing.
The technology can solve some interesting problems around verifying that a particular digital file is unique/original (which can be useful, because it's extremely easy to make copies of digital information) but it creates a long list of other problems as a side effect.
Almost every single non-theoretical problem that blockchains solve is something we've already solved. And most of the problems you could solve with a blockchains are severely limited by data-size limitations.
It would be amazing if I could decentrally store, say, a movie or videogame on a blockchain. Then, I could sell access tokens, would the owners could resell as they wanted. That's a GREAT way to use blockchain tech, because people would always have access, and they could use or sell the keys as they wanted. It doens't work though, because in the real world, that movie doesn't fit on the blockchain, it'll just be a link the a secondary source, and the whole thing falls apart.
And that's really the problem. Blockchains have a lot of nifty uses, but it almost always immediately falls apart around the edges, where it touches on non-blockchain tech, or, even worse, physical objects.
Intermediary free monetary transfer, lack of trust, transparency
It does only the last one and only partially.
How do you transfer money without an intermediary through blockchain?
I am pretty sure you just turn your money over to a scammer who just disappears with it. Since it is stateless and a libertarian dream, nothing can be done. So, congratulations!
You don’t need blockchain for that either. #theranos
By sending it to an address of the recipient's wallet from your wallet? I'm not sure what you're asking. Bitcoing transactions don't involve intermediaries by default unless you're using an exchange of some kind. You can even transfer between cryptocurrencies using atomic swaps.
Granted you'd have to buy crypto for fiat currency to begin with and because of unfortunate regulatotions you have to often go through a KYC process with some banking institution, but that's a fault of glowies getting greedy for data, not the tech.
Is the recipients wallet a web socket somewhere? How does the transaction end up in the actual blockchain such that others can confirm that this transaction was plausible?
So, instead of my bank sending money to your bank, I use my bank to send money to a bitcoin broker, then I send the bitcoin, and then their broker sends money to their bank, adding two more middlemen.
You've got it backwards. Instead of sending bitcoin to an exchange, selling it for dollars, doing a bank transfer, and then the recipient buying bitcoin, you can just send bitcoin from one person to another.
You should read my comment again, that's not what I said.
I said that in order to send money through the blockchain, it takes more middlemen than just sending money via other systems.
I've heard of using them as parts of like contracts?
What does blockchain solve that existing contracts don’t do? Blockchain has takeover possibility
Honestly I don't know. I'm just pointing out the only thing that kind of sort of sounded like a good idea for it I've ever heard. For pictures it's stupid that's for sure
It’s not a picture though. It’s a link to a picture on a server somewhere. If the host goes down, you own nothing.
Isn't it just a small amount of data? If the picture is small enough you could put it directly on the blockchain.
Dunno why you would though. It's very limiting for no particular gain.
But in NFTs the picture is not on the Blockchain. Only a link to the picture is on the Blockchain and the picture itself is still just on the web.
No, they are right. What they are saying is that even though pretty much all NFTs contain a link to a picture on the blockchain, theoretically you could write a really, really small image in the space where you would normally write the URL. From a quick google, that’s 100 bytes. For a black and white image, that’s 100 pixels. For color, that’s around 30 pixels
The technology behind it can be used for things other than pictures. That's kind of the point people are making
Right, and my original question was what does the technology solve? And so far the answer appears to be nothing
Yea the idea there is that with it being decentralized, it has an unedited history. So if each block added to the chain is a new transaction, you can see previous agreements. Being decentralized also means that it's public record and everyone can see the contract/agreement/transaction.
There's a lot of neat stuff that can be done, but as the other guy stated, it's a solution looking for a problem.
Nah. the other commenters are wrong.
They're super useful.
Its just that anyone who isn't selling bullshit uses their real name- Merkel trees - which are fundamental to modern software development (git, zfs, nix, nosql).
Merkel is the previous chancelor of Germany, Merkle is a computer scientist ;)
Hash trees are a part of blockchains, but not the entire thing. This is kinda like saying acupuncture isn't bullshit because needles are useful in real medicine as well.
https://hms.harvard.edu/news/exploring-science-acupuncture
Acupuncture actual does have clinical significant effects though.
I added an arxiv link and a stack overflow link that show that I'm not alone in this assertion of equivalence.
I blame autocorrect for the Merkle typos.
It's a solution that allows two parties, who are so paranoid they don't trust banks, let alone one another, to send funds and maintain a record of transactions with one another.
Blockchain / NFTs do not solve proof of ownership. Just ask all the people who had their NFTs or crypto stolen or lost in scams.
In your example, technically title fraud is more difficult because it needs to be done in two places. In reality it becomes far, far easier because you've now opened up a gigantic attack surface that you have no control over, and made both systems of verification worth less. If someone manages to compromise either one, there goes your proof of validity. Which one of them is real and which one is fraudulent?
Don’t we already have systems for that? What about the vulnerabilities of blockchain takeover?
Exactly. Ryathal said it very well: https://sh.itjust.works/comment/15029423