I’m not trying to argue one side over the other BTW. I am just trying to have an honest open minded discussion. Like I said, take out CSW & I like both SV & ABC plans forward.
I think the big argument that no one is having(and CSW is terrible at explaining) is how CTOR wrecks the competition part of the Nash equilibrium of Bitcoin mining.
BTW, have you studied BCHSV polo chart? I’m wondering if I should buy more or wait for dip?
I don't understand their terms and have been reluctant to get involved because of this. Who's to say they don't just declare ABC the winner after a brief hash win?
No, I'm explaining to you why you are wrong. You just keep repeating yourself, unwilling or unable to address the points I've made.
The points you've made don't disprove my statement that random ordering of transactions in blocks makes the system safer so I have no need to. Also learn how to thread replies 👋🏼
be able to deterministic-ally calculate a priv key from a pub key. But tx order is not one of them.
You're just bleating on off on a tangent. You still haven't admitted that deterministic systems are easier to attack. CTO is another attack on the chain's protocol, like Segwit.
The universe is deterministic. Bitcoin uses that determinism via of encryption and game theory to work. Read up and stop using buzzwords like you know what you're talking about.
Ok cool, so you won't prove that determinism is good for a system susceptible to attack. That's fine, I hope you control your own private keys ☮️
knowing the order of txs in a block. That's not how the encryption is used, or even works. Txs are hashed ONLY as part of merkle tree, meaning they only matter for the past blocks.
It's got nothing to do with encryption, but everything to do with determinism. And deterministic systems are more susceptible to attack. This is a fact that you have refused to admit.
No. Every node on the network already knows about every tx. The hash of the transactions does not serve to hide the txs or prevent them from interception. It is used ONLY as a....
You say no but then don't prove it. Every node knows about each tx yes, but not the order they will be blocked. So... With CTO they will, and it is more susceptible to fuckery.
According to some people, Jihan only wants CTOR because it is needed for WHC to succeed. I would like Jihan to participate on MEMO. I would like Roger and CSW to do the same.
the transactions in that block, not to encrypt the data, but to prove it's existence. So how is the cryptography at all at risk if transactions are ordered a certain way?
When you can predict, roughly, what data will be transferred every ten minutes it becomes easier to prepare for interception etc. Correct? As I said you'd have to use your imagination.
Yes, it increases verification complexity (slightly), but how does it make it "easier to attack" the chain.
Simply put; Its less random. We would have to use our imaginations as to how that might be a vector of attack in the future but it remains that determinism is not good for cryptography
Ok now Trezor (the guys who mocked this chain as BCash for months) has rowed in behind ABC so it should be clear to everyone that this is another attack. http://archive.is/3eBmr
No one has proven to me that it could have negative effects. From my understanding, this makes no difference in the network as a whole, but will require changes to many services.
Canonical ordering increases parsing & verification complexity & when transaction ordering is predetermined the chain is easier attacked.
ABCs very first client was released just over a year ago though and canonical transaction ordering is not entirely necessary for graphene so there's that. Nobody has proven that it...
won't have negative affects when a transaction A who sends to transaction B in the same block but B is sorted above A in the order due to txid. It could scupper future innovation 🤓
To answer your question, I take ABC seriously because they have a long and consistent history of working with the community to develop protocol changes in a responsible manner.
ABCs very first client was released just over a year ago though and canonical transaction ordering is not entirely necessary for graphene so there's that. Nobody has proven that it...
I am in total agreement with this post:
https://www.reddit.com/r/btc/comments/9uveos/dont_agree_with_posm_think_only_hash_matters_then/
The reason this is ignorant is because reddit and twitter are inherently gameable. It's a damned if you do, damned if you don't game. Memo is better cuz it's an even playing field.