Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Turing Machines on Bitcoin

xiaohuiliu.medium.com

Post reply on HN
Print view View original

11–20 of 80 posts

Re: Turing Machines on Bitcoin

#11

Post by shilch »

reginold wrote: post #7

Could someone who likes this article share what they liked about it? I'm getting the idea that it's either 1. A fun proof of principle project 2. entertains more "compute" ideas on blockchains, kinda like Ethereum

I was aware of xhliu's work before. It's actually a big thing because Ethereum was created with the assumption that bitcoin's scripting language does not allow for state transfers (stateful contracts) and turing-complete computations. The advantage over account-based systems like Ethereum is scalability: Ethereum contracts are one central entity identified by their address, any access on the contracts must be serialised which leads to scaling issues with popular contracts. Bitcoin's UTXO model on the other hand is rather easy to parallelize; but scalable UTXO-based contracts also require a different design.

Re: Turing Machines on Bitcoin

#12

Post by fogof »

p4bl0 wrote: post #5
sumeno wrote: post #3

This is the perfect representative of blockchain technology because it’s a convoluted way to do something that is better done with existing technologies without using a blockchain.

I'm glad I hit the reload button before writing my comment, I was going to say the exact same thing. EDIT: Okay blockchain zealots, downvote and flag my comment all you want, but if for once one of you could actually show a project that uses a blockchain (other than so-called "cryptocurrencies" which are essentially used as a speculation asset) and which cannot be implemented better and more efficiently with other te…

For what it’s worth I think that if people were downvoting you because of your position, they would be downvoting the parent as well. They’re downvoting you because saying “I was going to say the same thing” doesn’t contribute to the conversation.

Re: Turing Machines on Bitcoin

#13

Post by dane-pgp »

shilch wrote: post #9
xiphias2 wrote: post #4

This is about Bitcoin SV (BSV), not Bitcoin (BTC)

The solution presented in the article uses the bitcoin protocol. BSV is just one implementation of that. It could also work on BCH if some default limits were lifted, and (theoretically, not economically) possible on BTC.

> It could also work on BCH if some default limits were lifted

Can you list what those default limits are, and how big the needed changes are?

Re: Turing Machines on Bitcoin

#14

Post by burnished »

p4bl0 wrote: post #5
sumeno wrote: post #3

This is the perfect representative of blockchain technology because it’s a convoluted way to do something that is better done with existing technologies without using a blockchain.

I'm glad I hit the reload button before writing my comment, I was going to say the exact same thing. EDIT: Okay blockchain zealots, downvote and flag my comment all you want, but if for once one of you could actually show a project that uses a blockchain (other than so-called "cryptocurrencies" which are essentially used as a speculation asset) and which cannot be implemented better and more efficiently with other te…

Two things: 1) 'me-too' comments tend to get downvoted here in an effort to promote substantive discussion. I think something gets lost in the gulf between an upvote and stating you agree but thems the breaks and 2) cryptocurrencies actually do get used by people that lack access to stable currencies. I asked about this myself. I don't think it changes the original assessment, but it is heartening to know that there is some real utility people are getting out of this tech.

Re: Turing Machines on Bitcoin

#15

Post by dmitriid »

toolz wrote: post #6
sumeno wrote: post #3

This is the perfect representative of blockchain technology because it’s a convoluted way to do something that is better done with existing technologies without using a blockchain.

Until you want do something completely distributed that requires trust. The runtime isn't great, but then again, what runtime is better that meets those requirements? ACH is far slower than many blockchain solutions, for example.

> Until you want do something completely distributed that requires trust.

And then a trustless system helps... how exactly?

Oh, it doesn't, because you'll have to recreate all the trust institutions of the non-blockchain world. And, look, all blockchain "solutions" are busy doing exactly that.

Re: Turing Machines on Bitcoin

#16

Post by xiphias2 »

shilch wrote: post #9
xiphias2 wrote: post #4

This is about Bitcoin SV (BSV), not Bitcoin (BTC)

The solution presented in the article uses the bitcoin protocol. BSV is just one implementation of that. It could also work on BCH if some default limits were lifted, and (theoretically, not economically) possible on BTC.

BSV and Bitcoin (BTC) are incompatible protocols, so this is just a straight out lie. Sure, it can be added to BTC as long as there's a use case and concencus for it, but so far there isn't.

Re: Turing Machines on Bitcoin

#19

Post by shilch »

xiphias2 wrote: post #16
shilch wrote: post #9

Earlier quoted context omitted.

The solution presented in the article uses the bitcoin protocol. BSV is just one implementation of that. It could also work on BCH if some default limits were lifted, and (theoretically, not economically) possible on BTC.

BSV and Bitcoin (BTC) are incompatible protocols, so this is just a straight out lie. Sure, it can be added to BTC as long as there's a use case and concencus for it, but so far there isn't.

> BSV and Bitcoin (BTC) are incompatible protocols

Both implementations have a large intersection. Dogecoin, Litecoin and Dash also use the bitcoin protocol and should also work.

> Sure, it can be added to BTC

The presented solution does not require a protocol change. BSV is essentially the original bitcoin protocol (with few small exception); nothing was added to the protocol to implement the turing machine. BTC has some of the scripting opcodes disabled but they can either be implemented using other opcodes (no change) or they could be reenabled as a new SegWit deployment (requires change). However, the issue you'll run into in practice is that the reference implementation (Bitcoin Core) has very small limits set and discourages scripting. You'll have to find a miner that mines the custom scripting transactions. And, well, fees are another issue.

Re: Turing Machines on Bitcoin

#20

Post by truth_machine »

So it looks like a better title would have been "saving the state of the turing machine on the bitcoin blockchain", as the claim of Turing completeness[1] seems disingenious - bitcoin script itself has no looping constructs and is decidedly non turing complete. The user has to call the contract as many times as necessary to ensure that Turning machine transitions between states, and the same user checks that the computation terminated.

1: The claim is "It is straightforward to adapt the Turing machine contract above to implement any other Turing machines, by simply changing the states, the symbols and transition function. Thus, any Turing machine can be simulated on Bitcoin, conclusively proving Bitcoin is Turing-Complete by definition. QED."

Post reply on HN