3 ms·

>I agree, particularly since there's been 100s of contributors to the project. Which makes Blockstreams impact much less valuable than the original comment impl

>I agree, particularly since there's been 100s of contributors to the project. Which makes Blockstreams impact much less valuable than the original comment implied. This always get's bandied about. There are less than 20 people responsible for the vast majority of the commits in the last year, the other 90+ are small scattered changes. Easy to verify, go look at the commit logs. Oh, and 3 of the top Contributors aren't even working on core anymore. >SegWit does nothing of the sort. Lightning Network and Sidechains, which can be built on top of SegWit, certainly reduces off-chain fees. SegWit itself only reduces the fees on witness data, but that's moot for the time being. SegWit will have little impact on fees right now. Segwit is all about sidechains, that's the main application, and the real purpose. Side chains mean transactions off chain. Transactions off chain mean less fees for Bitcoin miners. It's not rocket science. The only way they make it back is if the settlement costs become huge. > That's actually false. Block sizes aren't really an issue for the miners, they are an issue for the nodes. Each miner only needs single full node. What increasing the block size does is make running a full node (which has to store the entire block chain) extremely expensive. The larger the block sizes, the more expensive a full node becomes. The reason nodes are expensive is because they are doing full historical validation which is almost completely useless and unnecessary. Minor changes to the Bitcoin block header would make that completely irrelevant. The nodes need all data ever argument is total junk. >LN is an opensource implementation of something not dissimilar to smart contracts. Calling that control is laughable. How much does it cost to open and close a channel on the main bitcoin blockchain at a 1 megabyte limit with millions of users? It's not a free thing, it's very expensive. To say that won't result in centralization is ridiculous. LN is not a solution without larger block sizes.

jhall1468 9y ago
> Segwit is all about sidechains, that's the main application, and the real purpose. You're just wrong. Segwit is about a major malleability bug. It happens to open the way for sidechains, but do you honestly think we shouldn't fix the bug purely because it happens to make sidechains more feasible? > Side chains mean transactions off chain. Transactions off chain mean less fees for Bitcoin miners. It's not rocket science. Correct, but you don't seem to be following the problem the same way I am. Your major concern is how much miners get paid. My major concern is network adoption and decentralization. Pretending that your values are the only values is what makes your position distasteful. > The reason nodes are expensive is because they are doing full historical validation which is almost completely useless and unnecessary. Minor changes to the Bitcoin block header would make that completely irrelevant. The nodes need all data ever argument is total junk. You just literally dismissed how an open block chain works. If nodes don't have the entire history attacks become arbitrary. > How much does it cost to open and close a channel on the main bitcoin blockchain at a 1 megabyte limit with millions of users? It's not a free thing, it's very expensive. And guess whose in the best position to implement channels (and collect the fees associated with it)? miners. > To say that won't result in centralization is ridiculous. Less than would exist by big blocks.
- buttershakes 9y ago
  Segwit is not about fixing malleability. Malleability can be fixed without SegWit. It's a simple change to the TX format. Segwit is over engineered to avoid a hard fork, and is primarily meant to support side chain technology, but in reality now we end up with multiple competing forks. You believe that limiting the block size increases adoption. That is an absurd viewpoint, I'm sorry. I don't agree with you. I think its like saying limiting the bandwidth of the backbone of the Internet increase adoption. What? It doesn't make sense there, and it doesn't make sense for Bitcoin. Regardless of what layer two technology exists. Layer two solutions are great, but if you look at highly scalable networks layer 1 isn't a piece of crap. I didn't dismiss how the blockchain works. This is your own misunderstanding of how to verify nodes. If you put the utxo hash into the block header you have lightweight history verification, Ethereum does it, and so can Bitcoin. It doesn't do it for historical reasons, not because there isn't a better technical way to do it. Nodes that need to can still validate full history, but just like headers validation can work for spv, a modified header could be used for full history verification. The fact this hasn't happened is incredibly telling, its because it helps create a false narrative around nodes not having the compute, bandwidth or disk space to handle bitcoin transaction history. Listen, you can attack my technical chops all you want. I've built tremendous amounts of stuff with Bitcoin, and it's obvious that small blocks are hurting adoption, hurting businesses that are trying to build software on Bitcoin, and creating arbitrary and ridiculous limits that are not in the realm of sound computer science. Arguing over a 1 megabyte increase in block size is the height of absurdity, it's Bitcoin's butter battle, and advocating for overly complicated layer 2 solutions when we are years away from being necessary is some kind of intellectual masturbation. I'm advocating for larger blocks, I hope that core is punished for their complete lack of management capability by losing control of the network, and I think that market forces will crush this absurd vision for Bitcoin when it is relegated to some small unimportant chain. Let's wait and see how it pans out.

jhall1468 9y ago

> Segwit is all about sidechains, that's the main application, and the real purpose. You're just wrong. Segwit is about a major malleability bug. It happens to open the way for sidechains, but do you honestly think we shouldn't fix the bug purely because it happens to make sidechains more feasible? > Side chains mean transactions off chain. Transactions off chain mean less fees for Bitcoin miners. It's not rocket science. Correct, but you don't seem to be following the problem the same way I am. Your major concern is how much miners get paid. My major concern is network adoption and decentralization. Pretending that your values are the only values is what makes your position distasteful. > The reason nodes are expensive is because they are doing full historical validation which is almost completely useless and unnecessary. Minor changes to the Bitcoin block header would make that completely irrelevant. The nodes need all data ever argument is total junk. You just literally dismissed how an open block chain works. If nodes don't have the entire history attacks become arbitrary. > How much does it cost to open and close a channel on the main bitcoin blockchain at a 1 megabyte limit with millions of users? It's not a free thing, it's very expensive. And guess whose in the best position to implement channels (and collect the fees associated with it)? miners. > To say that won't result in centralization is ridiculous. Less than would exist by big blocks.

buttershakes 9y ago
Segwit is not about fixing malleability. Malleability can be fixed without SegWit. It's a simple change to the TX format. Segwit is over engineered to avoid a hard fork, and is primarily meant to support side chain technology, but in reality now we end up with multiple competing forks. You believe that limiting the block size increases adoption. That is an absurd viewpoint, I'm sorry. I don't agree with you. I think its like saying limiting the bandwidth of the backbone of the Internet increase adoption. What? It doesn't make sense there, and it doesn't make sense for Bitcoin. Regardless of what layer two technology exists. Layer two solutions are great, but if you look at highly scalable networks layer 1 isn't a piece of crap. I didn't dismiss how the blockchain works. This is your own misunderstanding of how to verify nodes. If you put the utxo hash into the block header you have lightweight history verification, Ethereum does it, and so can Bitcoin. It doesn't do it for historical reasons, not because there isn't a better technical way to do it. Nodes that need to can still validate full history, but just like headers validation can work for spv, a modified header could be used for full history verification. The fact this hasn't happened is incredibly telling, its because it helps create a false narrative around nodes not having the compute, bandwidth or disk space to handle bitcoin transaction history. Listen, you can attack my technical chops all you want. I've built tremendous amounts of stuff with Bitcoin, and it's obvious that small blocks are hurting adoption, hurting businesses that are trying to build software on Bitcoin, and creating arbitrary and ridiculous limits that are not in the realm of sound computer science. Arguing over a 1 megabyte increase in block size is the height of absurdity, it's Bitcoin's butter battle, and advocating for overly complicated layer 2 solutions when we are years away from being necessary is some kind of intellectual masturbation. I'm advocating for larger blocks, I hope that core is punished for their complete lack of management capability by losing control of the network, and I think that market forces will crush this absurd vision for Bitcoin when it is relegated to some small unimportant chain. Let's wait and see how it pans out.