[duckingtest]

Calling it 'weak' is giving it too much consideration. The only sensible argument against larger blocks is that they boost mining centralization because larger blocks may take more time to propagate. Which may be true but it requires 1GB+ blocks. 10Gbps connection isn't a noticeable cost item for a serious mining operation, which means latency for a 1MB block and a 1GB one should be almost identical.

That's it. Remember that decentralization in bitcoin concerns miners, and miners only. 'Nodes' in the whitepaper are synonymous with miners.

Non-mining wallets need to have enough bandwidth to download blocks from miners as they appear. For 1GB blocks 20Mbps is enough. Which means a full verifying wallet for 1GB blocks is practical on a LTE connection.

Storage is also a non-problem with utxo commitments.

[kalleboo]

I thought the main concern with the block size was not typical western connections (where as you say 10 Gbps is no problem) but the limited performance of data crossing the Great Firewall of China?

[duckingtest]

That's complete FUD on several levels. On a most general level, mining itself only needs the header. Chinese miners can verify blocks and transactions elsewhere and only send headers to the actual mining sites. A dial-up should be enough.

Regardless, from what I heard it's not a problem anymore. Even if it was, allowing 'fast passage' for bitcoin blocks seems like a obvious course of action for the Chinese government, it's just another export industry at this point.

[ftlio]

> Calling it 'weak' is giving it too much consideration

You can only say this if you're too ignorant to be worth a damn on the subject, or are insane enough to think that the people who have developed the protocol for the last 8 years are all misguided. If you launch into some conspiracy about Blockstream, Inc., then we'll know it's a combination of both.

> Remember that decentralization in bitcoin concerns miners, and miners only

And exchanges, merchants, consumers. If you have a stake in the ledger by owning or accepting BTC, then you have a stake in the monetary policy of BTC. If you have a stake in the monetary policy of BTC, then you can only hope to enforce it (or attempt to change it) by running a node.

That's before the mathematics of bigger blocks, where the decentralization of miners is threatened even by a 2 MB block size, as those with more hash rate get another bonus due to the latency increases of even a 1 MB block size increase.

If all that mattered in Bitcoin was the decentralization of miners, this game would have been over a long time ago. Anybody who thinks that even a landscape of thousands of miners, pooled by very few pools, is decentralized enough on its own, without others validating and auditing the blockchain, is too capable of assumptions to be let near engineering.

Why does every currency that has ever existed get devalued into dogshit by its policymakers? Every currency dies because those close to the mint can afford devaluation more than those further from it.

[dwaltrip]

> the decentralization of miners is threatened even by a 2 MB block size, as those with more hash rate get another bonus due to the latency increases of even a 1 MB block size increase

This is completely ridiculous. If someone can't send 2MB of data across the network in less than a few seconds, then of course they shouldn't be mining.

[duckingtest]

>You can only say this if you're too ignorant to be worth a damn on the subject, or are insane enough to think that the people who have developed the protocol for the last 8 years are all misguided

Argument from authority

>where the decentralization of miners is threatened even by a 2 MB block size

proof by assertion

>is too capable of assumptions to be let near engineering.

ie 'u dumb'

I'm sure you convinced many people with your thoughtful comment. :)