5 ms·

> Ultimately, this a civic matter; as much as we want to ensure we have the right to encrypt our communications, to not disclose our passwords to police even wh

by sobkas 5y ago

> Ultimately, this a civic matter; as much as we want to ensure we have the right to encrypt our communications, to not disclose our passwords to police even when being prosecuted for a crime, as activists commonly insist on, so should we fight to maintain our existing rights to financial privacy. But you know, you are not anonymous on Blockchain, you're only pseudonymous. If someone correlate your pseudonym with your real identity, they will know your entire financial history (on Blockchain).

olalonde 5y ago
Not quite. You can discover one of the many addresses that are typically part of Bitcoin wallet. As for the history, you can't really tell much, especially if the person didn't re-use addresses which is the recommended mode of operation.
- danlugo92 5y ago
  It's as if almost every crypto skeptic doesn't know how it actually works.
  - jazzyjackson 5y ago
    Call it a UX problem. I've used BTC here and there but I'm still not confident that I could keep my transactions silo'd, from a cursory look at the blockchain it seems obvious that you can "follow the money" from a de-anonymized transaction (Say, I pay you for a service in person, so you know a wallet address that belongs to me) to other wallet addresses and balances associated with me. I can't blame people for losing interest / calling bullshit on crypto before spending hours diving into the details of L2, derivative wallets and multi-sig etc etc, details that ameliorate the skepticism. Also these things change over time, so maybe Lightning was unconvincing 3 years ago and skeptics just haven't given it a second look. (This is me, but I'm slowly being convinced that L2 really does make this stuff practical, but for a long time the transaction-per-second and unrecoverability of lost keys made the whole field very uninteresting)
    - danlugo92 5y ago
      Lightning will bring lots of improvements, heck it's in Cashapp now and it works very well in there, they bet on it, and you won't?
- e_y_ 5y ago
  Thankfully the vast majority of the public are crypto security experts and would have no problem applying these best practices to protect themselves when crypto becomes an everyday medium of financial exchange. /s
  - dataangel 5y ago
    Not reusing addresses is the default in a lot of wallet software though. The user doesn't have to do anything special.
    - jazzyjackson 5y ago
      still there is a steep learning curve between "send money from my coinbase account" to "manage a self-custodial wallet without deleting all my money"
      - headsoup 5y ago
        I think additionally, the 'casual' user needs to understand Monero, Bitcoin, L2, Ethereum, etc just to make in informed choice of what to use, knowing which are best for privacy, convenience, security, etc. It appears there are solutions available to a lot of concerns, but they require substantial knowledge of the ecosystem to use them and getting it wrong is permanent.
- ziml77 5y ago
  How does that work? If I make a new wallet and get coins sent to 0xAAAA and then I buy something and send them from address 0xBBBB, how do the miners validating the transaction know that 0xBBBB has any coins available to send?
  - CuriousCosmic 5y ago
    Cryptocurrencies have one of essentially two different accounting models: The account model: These networks have accounts and balances. You make a tx sending tokens from one account to another. It looks like normal banking more or less. The UTxO model: These networks don't have accounts at all. Instead they have what are called "Unspent Transaction Outputs" or UTxO. Every transaction takes in a set of UTxO and outputs a set of UTxO. All the balances of assets in the inputs must sum up to the same values as the assets in the newly created UTxO and you can only ever spend a UTxO once. To spend a given UTxO some criteria must be met. The most often criteria is some signature that only validates with the public key of the private key being used to sign the Tx. This also can be generalised to other multi-signature schemes or even to supporting smart contracts (via policy ids and script refs/hashes). Ethereum really was the network to push the account model into the forefront and now most networks choose to implement this over the UTxO model due to the ease with which you can get good UX and dev experience. Bitcoin, Bitcoin forks (SW or chain forks), Monero, Cardano, Ergo, Nervos, IOTA 2.0, and a few others use the UTxO model or some extension of it (ex: EUTxO or the Cell Model). One of the big upsides of the UTxO model is that all state is kept local instead of global however this local state also adds a number of complexities that make the model unintuitive to many. ------ So to answer your question: You'd create a Tx from your wallet. This Tx would consume some set of UTxO based on what you want to do and would then produce a number of UTxO going to any number of "accounts". In many UTxO networks, derivative keys are supported and you can give each person you want to receive money from a different derivative public key/address. If they were to send you money, it would go to your derivative address and it would be completely independent from all the rest of your UTxO until you do something on chain to tie them together (say by a delegation Tx in a PoS network, by merging UTxO, or by using UTxO from multiple derivative addresses in one Tx). As for how you'd that the UTxO is valid: Essentially it just needs to exist on chain and not already been spent somewhere else. Monero for example uses this to their advantage with the "sum to 0" rule to provide a zero knowledge non-interactive proof that a given Tx must be valid without knowing the contents, sender, or recipients.
lawn 5y ago
On some blockchains. Monero hides sender, receiver and amount by default.
- acdha 5y ago
  Monero _attempts_ to hide those details. This is an extremely hard problem and it has not been extensively tested against adversaries with deep resources. For keeping your drinking buddy from knowing what else you've spent money, it probably works but I would be very hesitant to tell people it's effective against, say, KYC exchanges much less an adversary who can do analysis against both the entire blockchain and other internet activities.
  - opportune 5y ago
    If you’ve cracked Monero encryption or have a general deanonymization attack, you should submit a solution to the IRS’ multi-year, $600k bounty. There have been a few instances of specific payment patterns causing de-anonymization (which just means payments are correlated and identified as not-spoof, due to the way the Monero blockchain works). But to be honest if you think Monero in general can be broken you probably don’t know how it works well enough to make that statement.
    - acdha 5y ago
      I’m not saying I know it’s broken. I’m saying it’s a hard problem and given the stakes I would treat it like a new cryptographic construct where it takes the community many years attacking the entire system before people start making confident statements about it. One of the big concerns here is retroactivity: if you make decisions because you _think_ you’re anonymous but that changes for any reason in the future (compromised exchanges, insecure clients, sophisticated new analysis techniques, etc.) you can’t pull that history. That necessitates more caution than something where the stakes are lower if an attacker finds something clever to try.
      - ziml77 5y ago
        This is exactly why I can't bring myself to put anything on any blockchain, no matter how anonymous it claims to be. I assume that someone will end up finding a flaw. When that happens, I may as well have been broadcasting everything without any security and privacy mechanisms at all. Not only can the person who's broken it see my history, anyone who knows the flaw can. I realize that things get leaked from centralized services, but there's a barrier to seeing that info. You put yourself at legal risk if you download one of those leaks and get caught. But there's no such risk with a public blockchain since the data was intended to be received by anyone who wants it.

by sobkas 5y ago

olalonde 5y ago

Not quite. You can discover one of the many addresses that are typically part of Bitcoin wallet. As for the history, you can't really tell much, especially if the person didn't re-use addresses which is the recommended mode of operation.

danlugo92 5y ago
It's as if almost every crypto skeptic doesn't know how it actually works.
- jazzyjackson 5y ago
  Call it a UX problem. I've used BTC here and there but I'm still not confident that I could keep my transactions silo'd, from a cursory look at the blockchain it seems obvious that you can "follow the money" from a de-anonymized transaction (Say, I pay you for a service in person, so you know a wallet address that belongs to me) to other wallet addresses and balances associated with me. I can't blame people for losing interest / calling bullshit on crypto before spending hours diving into the details of L2, derivative wallets and multi-sig etc etc, details that ameliorate the skepticism. Also these things change over time, so maybe Lightning was unconvincing 3 years ago and skeptics just haven't given it a second look. (This is me, but I'm slowly being convinced that L2 really does make this stuff practical, but for a long time the transaction-per-second and unrecoverability of lost keys made the whole field very uninteresting)
  - danlugo92 5y ago
    Lightning will bring lots of improvements, heck it's in Cashapp now and it works very well in there, they bet on it, and you won't?
e_y_ 5y ago
Thankfully the vast majority of the public are crypto security experts and would have no problem applying these best practices to protect themselves when crypto becomes an everyday medium of financial exchange. /s
- dataangel 5y ago
  Not reusing addresses is the default in a lot of wallet software though. The user doesn't have to do anything special.
  - jazzyjackson 5y ago
    still there is a steep learning curve between "send money from my coinbase account" to "manage a self-custodial wallet without deleting all my money"
    - headsoup 5y ago
      I think additionally, the 'casual' user needs to understand Monero, Bitcoin, L2, Ethereum, etc just to make in informed choice of what to use, knowing which are best for privacy, convenience, security, etc. It appears there are solutions available to a lot of concerns, but they require substantial knowledge of the ecosystem to use them and getting it wrong is permanent.
ziml77 5y ago
How does that work? If I make a new wallet and get coins sent to 0xAAAA and then I buy something and send them from address 0xBBBB, how do the miners validating the transaction know that 0xBBBB has any coins available to send?
- CuriousCosmic 5y ago
  Cryptocurrencies have one of essentially two different accounting models: The account model: These networks have accounts and balances. You make a tx sending tokens from one account to another. It looks like normal banking more or less. The UTxO model: These networks don't have accounts at all. Instead they have what are called "Unspent Transaction Outputs" or UTxO. Every transaction takes in a set of UTxO and outputs a set of UTxO. All the balances of assets in the inputs must sum up to the same values as the assets in the newly created UTxO and you can only ever spend a UTxO once. To spend a given UTxO some criteria must be met. The most often criteria is some signature that only validates with the public key of the private key being used to sign the Tx. This also can be generalised to other multi-signature schemes or even to supporting smart contracts (via policy ids and script refs/hashes). Ethereum really was the network to push the account model into the forefront and now most networks choose to implement this over the UTxO model due to the ease with which you can get good UX and dev experience. Bitcoin, Bitcoin forks (SW or chain forks), Monero, Cardano, Ergo, Nervos, IOTA 2.0, and a few others use the UTxO model or some extension of it (ex: EUTxO or the Cell Model). One of the big upsides of the UTxO model is that all state is kept local instead of global however this local state also adds a number of complexities that make the model unintuitive to many. ------ So to answer your question: You'd create a Tx from your wallet. This Tx would consume some set of UTxO based on what you want to do and would then produce a number of UTxO going to any number of "accounts". In many UTxO networks, derivative keys are supported and you can give each person you want to receive money from a different derivative public key/address. If they were to send you money, it would go to your derivative address and it would be completely independent from all the rest of your UTxO until you do something on chain to tie them together (say by a delegation Tx in a PoS network, by merging UTxO, or by using UTxO from multiple derivative addresses in one Tx). As for how you'd that the UTxO is valid: Essentially it just needs to exist on chain and not already been spent somewhere else. Monero for example uses this to their advantage with the "sum to 0" rule to provide a zero knowledge non-interactive proof that a given Tx must be valid without knowing the contents, sender, or recipients.

lawn 5y ago

On some blockchains. Monero hides sender, receiver and amount by default.

acdha 5y ago
Monero _attempts_ to hide those details. This is an extremely hard problem and it has not been extensively tested against adversaries with deep resources. For keeping your drinking buddy from knowing what else you've spent money, it probably works but I would be very hesitant to tell people it's effective against, say, KYC exchanges much less an adversary who can do analysis against both the entire blockchain and other internet activities.
- opportune 5y ago
  If you’ve cracked Monero encryption or have a general deanonymization attack, you should submit a solution to the IRS’ multi-year, $600k bounty. There have been a few instances of specific payment patterns causing de-anonymization (which just means payments are correlated and identified as not-spoof, due to the way the Monero blockchain works). But to be honest if you think Monero in general can be broken you probably don’t know how it works well enough to make that statement.
  - acdha 5y ago
    I’m not saying I know it’s broken. I’m saying it’s a hard problem and given the stakes I would treat it like a new cryptographic construct where it takes the community many years attacking the entire system before people start making confident statements about it. One of the big concerns here is retroactivity: if you make decisions because you _think_ you’re anonymous but that changes for any reason in the future (compromised exchanges, insecure clients, sophisticated new analysis techniques, etc.) you can’t pull that history. That necessitates more caution than something where the stakes are lower if an attacker finds something clever to try.
    - ziml77 5y ago
      This is exactly why I can't bring myself to put anything on any blockchain, no matter how anonymous it claims to be. I assume that someone will end up finding a flaw. When that happens, I may as well have been broadcasting everything without any security and privacy mechanisms at all. Not only can the person who's broken it see my history, anyone who knows the flaw can. I realize that things get leaked from centralized services, but there's a barrier to seeing that info. You put yourself at legal risk if you download one of those leaks and get caught. But there's no such risk with a public blockchain since the data was intended to be received by anyone who wants it.