Hacker News new | ask | show | jobs
by Meekro 3310 days ago
kbaker has done a good job of laying out the facts, but since he seems to lean towards the SegWit side, let me make the argument from the perspective of their opponents, the "big blockers."

Satoshi Nakamoto himself believed that bitcoin block size would have to increase rapidly, and that on-chain scaling is good enough for nearly all transactions. He wrote that blocks could scale up from the current 1MB to over 700MB as bandwidth and disks got cheaper while Bitcoin got more popular. [1] He foresaw the day when a "full node" would not be someone's laptop, but rather something the size of a data center. There would only be maybe a hundred full nodes in the world, run by universities or major bitcoin stakeholders, and everyone else would use various kinds of thin clients to interact with the network.

Gavin Andresen, the developer that Satoshi hand-picked to lead Bitcoin after his departure, shared this vision. He believed that on-chain scaling via bigger blocks is technically simpler (and simple is good!), and the resulting loss in full node count is nothing to worry about.

I'd also like to point out that one of the most modern coins, Ethereum, actually has both. They let miners vote on block size (big blocks), and they support off-chain scaling (like SegWit). They have both, and Satoshi thought Bitcoin should someday have both.

So let's get both! =)

[1] http://satoshinakamoto.me/2008/11/02/re-bitcoin-p2p-e-cash-p...

1 comments

For the record I support the "COOP" compromise. SegWit activation during the UASF timeline so that doesn't screw things up, with a locked-in HF to 2MB blocks.

I agree that bigger blocks are needed to ease even non-witness/standard transaction backlogs, and to ease up on wallet and other providers that want a faster network in-place without needing updates that work in concert SegWit like the Lightning Network. The 1 MB was originally an anti-spam limit. In a dream world, we would have a dynamic anti-spam block size. But I don't see consensus happening any time soon. :(