4 ms·

IMO L1 (Ethereum, Solana, Fantom, etc.) development is closer to building out a communications or telco network. From that “hard” technology perspective, it’s

by Dwolb 5y ago

IMO L1 (Ethereum, Solana, Fantom, etc.) development is closer to building out a communications or telco network. From that “hard” technology perspective, it’s very early days. And the apps that run on top of these L1s are fully limited by L1 bandwidth, latency, and blockspace. So, I think we’re in the pre-dial-up days for blockchains and will need a couple more orders of magnitude improvement to be (universally) on par with today’s app performance.

magicjosh 5y ago
Good stuff, so you would suggest to the author that their examples are too narrow. Perhaps that smartphones, Uber, and Facebook all represented somewhat minor "time has come" type innovations, whereas you see Bitcoin as a development more like the internet itself, and that instead we are in the Arpanet days. (Disclosure I own Bitcoin and Ethereum)
- Dwolb 5y ago
  Yeah narrow or just on a different layer/level of abstraction. Yup my mental model is we’re either in Arpanet days or multiple-competing-DOS days.
  - magicjosh 5y ago
    Are there any other examples of recent + long cycle developments? Solar? Electric vehicles? Part of me has come to expect such fast development (Facebook, Uber, Yelp!), but other modern industries must have longer cycles I just am not paying attention.
    - Dwolb 5y ago
      1G was delivered late 1970s, 5G… 2018ish? https://www.techadvisor.com/feature/small-business/timeline-of-5g-development-3788816/ https://www.techadvisor.com/feature/small-business/timeline-... JWST was conceived in 1996: https://en.m.wikipedia.org/wiki/Timeline_of_the_James_Webb_Space_Telescope https://en.m.wikipedia.org/wiki/Timeline_of_the_James_Webb_S...
    - jakupovic 5y ago
      https://twitter.com/TheCryptoLark/status/1471730881337323522 https://twitter.com/TheCryptoLark/status/1471730881337323522
      - magicjosh 5y ago
        This presentation of the data suggests Bitcoin is at a "1998" level of development in terms of total number of users. I'm curious about this "number of users" metric, how would it map to adoption of other tech like telephones, televisions, cell phones, and smart phones.
  - Daishiman 5y ago
    Early day Arpanet was infinitely more useful than all of this blockchain stuff.
rileyphone 5y ago
The information bandwidth on block based cryptocurrencies is far too low to support these deeper levels of complexity, and I doubt will ever get much better. Blockchains, despite being a distributed phenomenon, still centralize the flow of information into one stream, the chain. Ethereum, by my estimation, operates at around 7 kB/s, and Bitcoin at 2kB/s. These are clearly glacial speeds - think of how limited an internet connection at those speeds is. How will more informationally needy systems be constructed on top of so paltry a base? The problem is a blockchain's protocol must somehow coordinate the distributed actions of its participants, which are generally spread out across the world. By centralizing information in the chain, they are exhibiting the same intractable slowdown that we would see trying to emulate a brain on a CPU.
- Dwolb 5y ago
  I think we’ll see if bandwidth and latency improvements bring more application development. Solana and Avalanche are orders of magnitude faster than Ethereum and their ecosystems are rapidly evolving. Btw Solana founder is ex-Qualcomm and worked on embedded distributed (multi-core) systems and Avalanche founder implemented a faster more elegant consensus algorithm. Can hear them discussing technical aspects of development: https://open.spotify.com/episode/632rPGnMZlHag7DJ1SCDUV?si=F_VSaBWwQwqkeJK3S8_zuQ https://open.spotify.com/episode/632rPGnMZlHag7DJ1SCDUV?si=F...
- che_shirecat 5y ago
  read up on L2 chains - both optimistic roll-up and zk-rollups. this problem you are breaking down has been the subject of study and development for years now and is nearing implementation - https://www.gemini.com/cryptopedia/layer-2-scaling-zk-rollup-optimistic-rollup-ethereum#section-what-is-a-rollup https://www.gemini.com/cryptopedia/layer-2-scaling-zk-rollup...
dmitriid 5y ago
Yup. Of course everything that has "blockchain in it is "the next internet". Now this ridiculous claim is taken further: it's not even the internet, but "the fundamental infrastructure of the internet". Whereas all signs point to it being a Juicero, an Enron and an AT&T ISIS Mobile Wallet rolled into one.

by Dwolb 5y ago

magicjosh 5y ago

Good stuff, so you would suggest to the author that their examples are too narrow. Perhaps that smartphones, Uber, and Facebook all represented somewhat minor "time has come" type innovations, whereas you see Bitcoin as a development more like the internet itself, and that instead we are in the Arpanet days. (Disclosure I own Bitcoin and Ethereum)

Dwolb 5y ago
Yeah narrow or just on a different layer/level of abstraction. Yup my mental model is we’re either in Arpanet days or multiple-competing-DOS days.
- magicjosh 5y ago
  Are there any other examples of recent + long cycle developments? Solar? Electric vehicles? Part of me has come to expect such fast development (Facebook, Uber, Yelp!), but other modern industries must have longer cycles I just am not paying attention.
  - Dwolb 5y ago
    1G was delivered late 1970s, 5G… 2018ish? https://www.techadvisor.com/feature/small-business/timeline-of-5g-development-3788816/ https://www.techadvisor.com/feature/small-business/timeline-... JWST was conceived in 1996: https://en.m.wikipedia.org/wiki/Timeline_of_the_James_Webb_Space_Telescope https://en.m.wikipedia.org/wiki/Timeline_of_the_James_Webb_S...
  - jakupovic 5y ago
    https://twitter.com/TheCryptoLark/status/1471730881337323522 https://twitter.com/TheCryptoLark/status/1471730881337323522
    - magicjosh 5y ago
      This presentation of the data suggests Bitcoin is at a "1998" level of development in terms of total number of users. I'm curious about this "number of users" metric, how would it map to adoption of other tech like telephones, televisions, cell phones, and smart phones.
- Daishiman 5y ago
  Early day Arpanet was infinitely more useful than all of this blockchain stuff.

rileyphone 5y ago

The information bandwidth on block based cryptocurrencies is far too low to support these deeper levels of complexity, and I doubt will ever get much better. Blockchains, despite being a distributed phenomenon, still centralize the flow of information into one stream, the chain. Ethereum, by my estimation, operates at around 7 kB/s, and Bitcoin at 2kB/s. These are clearly glacial speeds - think of how limited an internet connection at those speeds is. How will more informationally needy systems be constructed on top of so paltry a base? The problem is a blockchain's protocol must somehow coordinate the distributed actions of its participants, which are generally spread out across the world. By centralizing information in the chain, they are exhibiting the same intractable slowdown that we would see trying to emulate a brain on a CPU.

Dwolb 5y ago
I think we’ll see if bandwidth and latency improvements bring more application development. Solana and Avalanche are orders of magnitude faster than Ethereum and their ecosystems are rapidly evolving. Btw Solana founder is ex-Qualcomm and worked on embedded distributed (multi-core) systems and Avalanche founder implemented a faster more elegant consensus algorithm. Can hear them discussing technical aspects of development: https://open.spotify.com/episode/632rPGnMZlHag7DJ1SCDUV?si=F_VSaBWwQwqkeJK3S8_zuQ https://open.spotify.com/episode/632rPGnMZlHag7DJ1SCDUV?si=F...
che_shirecat 5y ago
read up on L2 chains - both optimistic roll-up and zk-rollups. this problem you are breaking down has been the subject of study and development for years now and is nearing implementation - https://www.gemini.com/cryptopedia/layer-2-scaling-zk-rollup-optimistic-rollup-ethereum#section-what-is-a-rollup https://www.gemini.com/cryptopedia/layer-2-scaling-zk-rollup...

dmitriid 5y ago

Yup. Of course everything that has "blockchain in it is "the next internet". Now this ridiculous claim is taken further: it's not even the internet, but "the fundamental infrastructure of the internet". Whereas all signs point to it being a Juicero, an Enron and an AT&T ISIS Mobile Wallet rolled into one.