3 ms·

Many people in the crypto space are peddling their own agendas. I've read the tangle paper. It seems as the node transaction weights grow, the double spend pr

Many people in the crypto space are peddling their own agendas. I've read the tangle paper. It seems as the node transaction weights grow, the double spend problem is solved (transaction nodes gain weight as others confirm). Arguably you could have bad actors confirming transactions but that is why the coordinator exists until there are enough transactions happening where a coordinated double-spend attack is not feasible. Looking at tfha's post history, you can see he does not support POS as secure either and refers to the opinions of the bitcoin team. These are two red flags for me. There are others that do not act in your best interest taking center stage in the crypto world. There are many very well respected computer scientists in the iota foundation and I'm sure they are well aware of any issues the coin will have along the way. Edit: Don't trust me either. Do your own research.

LightMachine 9y ago
Sure, if you wait for a sufficiently large weight on top of your transaction, then you solve the double spend problem. But you need at the bare minimum that every transaction that happened at roughly the same time as yours to be included as an indirect child of your tips. If you do not understand that sentence, please try to, as it is the very reason that IOTA doesn't scale (and isn't fundamentally different from usual blockchains in any sense other than a somewhat contrived construction).
- hyprCoin 9y ago
  I appreciate the succintness of your comment. I believe there is an elaboration here: https://github.com/noneymous/iota-consensus-presentation https://github.com/noneymous/iota-consensus-presentation If I understand the issue you've pointed out correctly, this is why the coordinator currently exists. Once there are constantly running zero value transactions all around us (self driving cars, smart sensors, etc), then that becomes trivial. IE the number of indirect children of your tip increases at a rate of roughly 2^n (less due to random tip selection). This leads to confirmation in an exponential fashion.
runeks 9y ago
> It seems as the node transaction weights grow, the double spend problem is solved (transaction nodes gain weight as others confirm). Arguably you could have bad actors confirming transactions but that is why the coordinator exists until there are enough transactions happening where a coordinated double-spend attack is not feasible. What is the fundamental assumption that makes the paper assert that this problem will be solved in the future? Is the assumption that hashing power will be spread more evenly among nodes? And, if so, what is the basis for assuming this? It's certainly not what we see with Bitcoin. Is the case that the IOTA security model breaks down if hashing power is unevenly distributed?
tfha 9y ago
That's incorrect because of the assymetry between attackers and normal users. An attacker needs only generate as many transactions as the rest of the network combined, and indeed this isn't that expensive. A single consumer machine can generate hundreds, if not thousands of transactions per second. A fleet of specialized machines costing less than 7 figures total can do millions, if not hundreds of millions, of transactions per second. IOTA has a terrible security ratio.

LightMachine 9y ago

Sure, if you wait for a sufficiently large weight on top of your transaction, then you solve the double spend problem. But you need at the bare minimum that every transaction that happened at roughly the same time as yours to be included as an indirect child of your tips. If you do not understand that sentence, please try to, as it is the very reason that IOTA doesn't scale (and isn't fundamentally different from usual blockchains in any sense other than a somewhat contrived construction).

hyprCoin 9y ago
I appreciate the succintness of your comment. I believe there is an elaboration here: https://github.com/noneymous/iota-consensus-presentation https://github.com/noneymous/iota-consensus-presentation If I understand the issue you've pointed out correctly, this is why the coordinator currently exists. Once there are constantly running zero value transactions all around us (self driving cars, smart sensors, etc), then that becomes trivial. IE the number of indirect children of your tip increases at a rate of roughly 2^n (less due to random tip selection). This leads to confirmation in an exponential fashion.

runeks 9y ago

> It seems as the node transaction weights grow, the double spend problem is solved (transaction nodes gain weight as others confirm). Arguably you could have bad actors confirming transactions but that is why the coordinator exists until there are enough transactions happening where a coordinated double-spend attack is not feasible. What is the fundamental assumption that makes the paper assert that this problem will be solved in the future? Is the assumption that hashing power will be spread more evenly among nodes? And, if so, what is the basis for assuming this? It's certainly not what we see with Bitcoin. Is the case that the IOTA security model breaks down if hashing power is unevenly distributed?

tfha 9y ago

That's incorrect because of the assymetry between attackers and normal users. An attacker needs only generate as many transactions as the rest of the network combined, and indeed this isn't that expensive. A single consumer machine can generate hundreds, if not thousands of transactions per second. A fleet of specialized machines costing less than 7 figures total can do millions, if not hundreds of millions, of transactions per second. IOTA has a terrible security ratio.