4 ms·

> There are little cracks that give us brief glimpses of the abyss below. Why not cover them up with a fresh, cool coat of blockchain? This is a great example

> There are little cracks that give us brief glimpses of the abyss below. Why not cover them up with a fresh, cool coat of blockchain? This is a great example of Wall Street's continued but lessening snark towards blockchain and vicariously Bitcoin. That much of the complex system you know and revere and spent so long learning can be replaced by new fangled technology is still a hard pill to swallow. The snark is a coping mechanism.

dsacco 10y ago
As currently implemented, blockchain technology cannot scale to the requirements of Wall St. It simply lacks the transaction volume. That's leaving aside other computation and synchronization difficulties inherent to distributing resources rather than centralizing them. Furthermore, finance is not a homogenous group of luddites. Blockchains are being actively explored in the industry. There's even a consortium of large banks developing one - they're just doing it without all the decentralized sugar that bitcoin has. Finally, as others have mentioned in this thread, a blockchain wouldn't have obviated this problem. The primary difficulty isn't identifying the owners of shares from three years ago, it's in correctly finding them and arranging for rightful reimbursement. I think your characterization of the industry is misinformed and unfair.
- mikeyouse 10y ago
  Bitcoin - capable of 250,000 transactions per day. NASDAQ - 11 million transactions per day.[1] I can't see any issues here! [1] - https://www.nasdaqtrader.com/Trader.aspx?id=DailyMarketSummary https://www.nasdaqtrader.com/Trader.aspx?id=DailyMarketSumma...
- jcoffland 10y ago
  > As currently implemented, blockchain technology cannot scale to the requirements of Wall St. It simply lacks the transaction volume. There are so many implementations. Many of these can process orders of magnitude more transactions than Bitcoin.
  - dsacco 10y ago
    How many orders of magnitude, exactly? Bitcoin's maximum throughput is about 7 transactions per second. Superior blockchain implementations can generally achieve ~30 transactions per second for long term, effective operation. Nasdaq's daily trades would require over 430 transactions per second. There are tradeoffs to be made (such as block size or integrity), but it's fundamentally not feasible to port our existing centralized ledger to a decentralized blockchain ledger. This is just Nasdaq. Imagine trying to scale a decentralized blockchain to multiple markets, or to Visa's requirements, which are over 2000 transactions per second. There is a variety of research on this topic: http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://www.tik.ee.ethz.ch/file/74bc987e6ab4a8478c04950616612f69/main.pdf http://www.tik.ee.ethz.ch/file/74bc987e6ab4a8478c04950616612... http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
    - jcoffland 10y ago
      Your numbers are nonsense. A blockchain can be made to handle any number of transactions per second. It's simply a matter of increasing the number of transactions per block, the frequency of the blocks or both. The trade off is in the increasing storage requirements. The Bitcoin blockchain currently sits at just over 100GiB. This is an insignificant amount of data for a bank. Furthermore, you can reduce the blockchain storage requirements through checkpointing. Every so many blocks the participants in the blockchain can agree to and cryptographically sign the latest balance sheet. All previous transactions can then be archived. IMO, the Bitcoin Core developers shouldn't be focusing on anything but checkpointing right now.
      - dsacco 10y ago
        Do you have any evidence to support what you're saying? The research I cited directly contradicts your claim. You cannot scale up blockchains arbitrarily without making fundamental tradeoffs to increase their performance (for example, increasingly centralizing it).
        10y ago
- lenocinor 10y ago
  DTCC is actually one of the companies looking into using blockchains: http://www.dtcc.com/news/2016/january/25/blockchain http://www.dtcc.com/news/2016/january/25/blockchain
jdcryans 10y ago
I think you are misattributing his snark. In his regular column he has a recurring section called "Blockchain blockchain blockchain" such as in [1] in which he covers how the finance industry itself is trying to apply blockchains to everything. It's the new fad, they are in the Peak of Inflated Expectations [2]. So while it might be true that some bankers are afraid that Bitcoin is going to replace them (I have no opinion on that) the author is highlighting yet another case where blockchain is supposed to easily solve any difficult distributed ledger problem. 1. https://www.bloomberg.com/view/articles/2017-01-31/mirror-trades-and-two-for-one-rules https://www.bloomberg.com/view/articles/2017-01-31/mirror-tr... 2. https://upload.wikimedia.org/wikipedia/commons/thumb/9/94/Gartner_Hype_Cycle.svg/320px-Gartner_Hype_Cycle.svg.png https://upload.wikimedia.org/wikipedia/commons/thumb/9/94/Ga...
- jcoffland 10y ago
  You demonstrate that the author is indeed snarky towards blockchains.

dsacco 10y ago

As currently implemented, blockchain technology cannot scale to the requirements of Wall St. It simply lacks the transaction volume. That's leaving aside other computation and synchronization difficulties inherent to distributing resources rather than centralizing them. Furthermore, finance is not a homogenous group of luddites. Blockchains are being actively explored in the industry. There's even a consortium of large banks developing one - they're just doing it without all the decentralized sugar that bitcoin has. Finally, as others have mentioned in this thread, a blockchain wouldn't have obviated this problem. The primary difficulty isn't identifying the owners of shares from three years ago, it's in correctly finding them and arranging for rightful reimbursement. I think your characterization of the industry is misinformed and unfair.

mikeyouse 10y ago
Bitcoin - capable of 250,000 transactions per day. NASDAQ - 11 million transactions per day.[1] I can't see any issues here! [1] - https://www.nasdaqtrader.com/Trader.aspx?id=DailyMarketSummary https://www.nasdaqtrader.com/Trader.aspx?id=DailyMarketSumma...
jcoffland 10y ago
> As currently implemented, blockchain technology cannot scale to the requirements of Wall St. It simply lacks the transaction volume. There are so many implementations. Many of these can process orders of magnitude more transactions than Bitcoin.
- dsacco 10y ago
  How many orders of magnitude, exactly? Bitcoin's maximum throughput is about 7 transactions per second. Superior blockchain implementations can generally achieve ~30 transactions per second for long term, effective operation. Nasdaq's daily trades would require over 430 transactions per second. There are tradeoffs to be made (such as block size or integrity), but it's fundamentally not feasible to port our existing centralized ledger to a decentralized blockchain ledger. This is just Nasdaq. Imagine trying to scale a decentralized blockchain to multiple markets, or to Visa's requirements, which are over 2000 transactions per second. There is a variety of research on this topic: http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://www.tik.ee.ethz.ch/file/74bc987e6ab4a8478c04950616612f69/main.pdf http://www.tik.ee.ethz.ch/file/74bc987e6ab4a8478c04950616612... http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf http://fc16.ifca.ai/bitcoin/papers/CDE+16.pdf
  - jcoffland 10y ago
    Your numbers are nonsense. A blockchain can be made to handle any number of transactions per second. It's simply a matter of increasing the number of transactions per block, the frequency of the blocks or both. The trade off is in the increasing storage requirements. The Bitcoin blockchain currently sits at just over 100GiB. This is an insignificant amount of data for a bank. Furthermore, you can reduce the blockchain storage requirements through checkpointing. Every so many blocks the participants in the blockchain can agree to and cryptographically sign the latest balance sheet. All previous transactions can then be archived. IMO, the Bitcoin Core developers shouldn't be focusing on anything but checkpointing right now.
    - dsacco 10y ago
      Do you have any evidence to support what you're saying? The research I cited directly contradicts your claim. You cannot scale up blockchains arbitrarily without making fundamental tradeoffs to increase their performance (for example, increasingly centralizing it).
      - 10y ago
lenocinor 10y ago
DTCC is actually one of the companies looking into using blockchains: http://www.dtcc.com/news/2016/january/25/blockchain http://www.dtcc.com/news/2016/january/25/blockchain

jdcryans 10y ago

I think you are misattributing his snark. In his regular column he has a recurring section called "Blockchain blockchain blockchain" such as in [1] in which he covers how the finance industry itself is trying to apply blockchains to everything. It's the new fad, they are in the Peak of Inflated Expectations [2]. So while it might be true that some bankers are afraid that Bitcoin is going to replace them (I have no opinion on that) the author is highlighting yet another case where blockchain is supposed to easily solve any difficult distributed ledger problem. 1. https://www.bloomberg.com/view/articles/2017-01-31/mirror-trades-and-two-for-one-rules https://www.bloomberg.com/view/articles/2017-01-31/mirror-tr... 2. https://upload.wikimedia.org/wikipedia/commons/thumb/9/94/Gartner_Hype_Cycle.svg/320px-Gartner_Hype_Cycle.svg.png https://upload.wikimedia.org/wikipedia/commons/thumb/9/94/Ga...

jcoffland 10y ago
You demonstrate that the author is indeed snarky towards blockchains.