4 ms·

> computers were expensive and slow, and network connectivity was very limited Blockchains have been expensive and slow, and functionality was very limited. Wh

> computers were expensive and slow, and network connectivity was very limited Blockchains have been expensive and slow, and functionality was very limited. When Bitcoin launched, you couldn't run turing-complete scripts, it only supported balance transfers. It had a hard-coded 1 Megabyte per 10 minutes throughput limit (slower than dial-up). There were no transaction privacy tools. That has been changing. The first turing-complete chain, Ethereum, was launched in 2015. It launched a sustainable consensus model in 2020, and switched over to it ("merged") in 2022. Private smart contract support came in 2020-2021, with projects like SCRT and Aztec. Layer-2 scaling technology is just beginning to hit its stride this year, with several competing companies. Still on the base layer roadmap is single-slot finality, data availability sampling to expand the chain's storage capacity, zero-knowledge-proof based VMs to expand its execution throughput, and state and history expiry to automatically prune it. Once all of those technologies are complete and have had a chance to mature (much like dial-up matured into broadband), I expect blockchains to look more attractive for certain applications.

acdha 4y ago
I was responding specifically the implicit assumption that the timeframes can be directly compared — since things happened on different scales, that doesn't make sense especially when you're comparing different things. The number of people using the internet changed as a function of reasonable speed/price connectivity becoming available where they lived so the process was a lot slower than shipping an update to a blockchain network because it involved things like getting permits and deploying crews to install cables. That said, even if you want to ignore Bitcoin, things like Ethereum still are lagging far behind — if that network shut off tomorrow, nobody not involved in selling tokens would notice. The same would not have been true of the web 7 years, or even 2 years, after it launched because tons of people were using it for things unrelated to selling web services. It's certainly possible that at some point some level of functionality will make it more attractive but I suspect that this will involve rolling back parts of the current sales pitch and that will unfavorably affect the cost relative to other models like distributed ledgers or standard APIs.
- everfree 4y ago
  > the process was a lot slower than shipping an update to a blockchain network The process of creating a system for private, energy efficient smart contract transactions took 12 years. To make them scalable and foolproof for the general public will likely take another 5-10 years. Blockchain updates are less like the process of laying physical cable, more like the process of building up a very large piece of software with lots of moving parts, like an operating system. They both require a lot of effort, but with laying cables the effort is physical and legal whereas with operating systems (and blockchain platforms) the effort is heady and research-oriented. We've seen that shipping an update to a blockchain network can be extremely slow indeed, if that update requires a lot of R&D.

acdha 4y ago

I was responding specifically the implicit assumption that the timeframes can be directly compared — since things happened on different scales, that doesn't make sense especially when you're comparing different things. The number of people using the internet changed as a function of reasonable speed/price connectivity becoming available where they lived so the process was a lot slower than shipping an update to a blockchain network because it involved things like getting permits and deploying crews to install cables. That said, even if you want to ignore Bitcoin, things like Ethereum still are lagging far behind — if that network shut off tomorrow, nobody not involved in selling tokens would notice. The same would not have been true of the web 7 years, or even 2 years, after it launched because tons of people were using it for things unrelated to selling web services. It's certainly possible that at some point some level of functionality will make it more attractive but I suspect that this will involve rolling back parts of the current sales pitch and that will unfavorably affect the cost relative to other models like distributed ledgers or standard APIs.

everfree 4y ago
> the process was a lot slower than shipping an update to a blockchain network The process of creating a system for private, energy efficient smart contract transactions took 12 years. To make them scalable and foolproof for the general public will likely take another 5-10 years. Blockchain updates are less like the process of laying physical cable, more like the process of building up a very large piece of software with lots of moving parts, like an operating system. They both require a lot of effort, but with laying cables the effort is physical and legal whereas with operating systems (and blockchain platforms) the effort is heady and research-oriented. We've seen that shipping an update to a blockchain network can be extremely slow indeed, if that update requires a lot of R&D.