Apr 23, 2022·edited Apr 24, 2022Liked by Daniel H. Neilson
"Programmable contracts are possible only on a distributed ledger, where both parties independently reach the same result and can verify that this is the case."
I'm not not convinced that this is true. You can have stored routines on a non-distributed database. Why does it matter that two parties are "reaching the same result" as long as the database reaches a result in a way that both parties accept?
The consensus problem is a problem that is particular to distributed databases. Consensus is not an issue on a non-distributed database because there's nobody to reach consensus with. There's only one decision maker.
The problem itself is a distributed problem. To have all parties upload a single database is what the DTCC does today - and there likely is a programmed routine confirming the inputs.
The challenge is that the inputs must still be provided by two parties that are distinctly separate and outside of the context of each other and the DTCC. So in reality, everyone is updating an internal database, pushing data to the DTCC database, and then reconciling the results as they come out. Not so high fidelity when the chaos of a market is flying around.
A distributed ledger would allow all parties to avoid the reconciliation need as the data is reconciled in real-time, in a more tightly coupled fashion. It is akin to double entry accounting in a digitally connected context.
I think I see it the same way as Khan. "Distributed" here might mean just the two counterparties to a particular trade. It's not about opening up the wholesale process to some broad public, just about re-tooling the process of reconciliation. Significant because it's a lot of volume, not because it would look completely different from the outside, which it would not.
If there’s tokenization, it’s crypto. SBF is one bad actor in crypto. He doesn’t get to own it. There are many good actors, including Satoshi and Vitalik.
"Programmable contracts are possible only on a distributed ledger, where both parties independently reach the same result and can verify that this is the case."
I'm not not convinced that this is true. You can have stored routines on a non-distributed database. Why does it matter that two parties are "reaching the same result" as long as the database reaches a result in a way that both parties accept?
The consensus problem is a problem that is particular to distributed databases. Consensus is not an issue on a non-distributed database because there's nobody to reach consensus with. There's only one decision maker.
Maybe I'm missing something?
The problem itself is a distributed problem. To have all parties upload a single database is what the DTCC does today - and there likely is a programmed routine confirming the inputs.
The challenge is that the inputs must still be provided by two parties that are distinctly separate and outside of the context of each other and the DTCC. So in reality, everyone is updating an internal database, pushing data to the DTCC database, and then reconciling the results as they come out. Not so high fidelity when the chaos of a market is flying around.
A distributed ledger would allow all parties to avoid the reconciliation need as the data is reconciled in real-time, in a more tightly coupled fashion. It is akin to double entry accounting in a digitally connected context.
I think I see it the same way as Khan. "Distributed" here might mean just the two counterparties to a particular trade. It's not about opening up the wholesale process to some broad public, just about re-tooling the process of reconciliation. Significant because it's a lot of volume, not because it would look completely different from the outside, which it would not.
If there’s tokenization, it’s crypto. SBF is one bad actor in crypto. He doesn’t get to own it. There are many good actors, including Satoshi and Vitalik.