<<  >> (p.804)
    Author Topic: Gold collapsing. Bitcoin UP.  (Read 2032328 times)
    notme
    Legendary
    *
    Offline Offline

    Activity: 1904
    Merit: 1002


    View Profile
    November 06, 2014, 09:34:13 PM
     #16061

    ....
    Then we agree.
    Side chains accommodate arbitrary crypto (good bad or ugly) and which crypto is used for a particular chain matters.

    It is my understanding that while the transactions can use arbitrary crypto, block hashing would be limited to sha256 (or possibly a small list of others).  In order to ensure a malicious attacker can not simply lie, the bitcoin miners supporting SPV proofs will need to verify the block hash is valid for the headers provided.  The transaction's details that destroy the pegged coin and unlock the bitcoin will be used, but the signature can be trusted since it is in the block with the highest amount of work.  If nobody presents a higher difficulty block header that contains a contradicting transaction within the contest period, the bitcoins unlock.  But, to verify all this, the miners must be able to hash the headers.

    Is  it possible for attackers to do this in reverse?

    Take someone's cold wallet and lock them into an SPVPROOF on a SC while the true owner is indisposed for some reason? Or is this impossible because they would need the private keys to begin with? 

    Presumably the sidechain would still require a valid signature to allow a transaction in a block.

    you mean the SC block?  yes, don't these SPV proofs require 2 tx's each, one in MC and one in SC?
    Quote
    Essentially, what you are dealing with is a set of headers and a transaction.  The headers prove the transaction was added to a block.  (See section 8 of bitcoin.pdf).

    there is no section 8
    Quote
    What the sidechains idea adds is that instead light node verifying a bitcoin transaction, the bitcoin miners verify that the sidechain has a transaction that destroys sidecoins.  If this transaction goes uncontested, the SPV proof is accepted and the previously locked bitcoins are sent to the address specified in the destroy transaction.

    i asked you before about this contest period.  what is the probability of an attacker constructing a fake proof in either direction.
    Quote

    what is this all about?  isn't this a fundamental change to how Bitcoin blocks are linked together?

    We require a change to Bitcoin so that rather than each blockheader committing only to the
    header before it, it commits to every one of its ancestors.


    Yes a SC block.  Yes, there needs to be a destroy transaction on SC and a SPV proof on MC.

    Did you try looking between sections 7 and 9?  That's where I found section 8.  (Hint, page 5).

    An attacker can only fake a proof if they can fake a block, so it is up to the security model of the sidechain.

    Can you put that quote in context (where is it in which whitepaper?).... I'm not sure quite what they are referring to.

    pg 20 Implementation

    why would an attacker have to fake a block when faking a SPV proof (tx)?  blocks are created by miners...

    still don't see a section 8 on pg 5

    A cursory rereading of that section still leaves me with some questions, but I'll try to remember to come back and look into the reference Pug90 when I'm not busy.

    An SPV proof is a tx + block headers that prove the tx has been accepted by SC miners.  They can't fake a proof without forging the transaction into a SC block.

    bitcoin.pdf section 8

    https://www.bitcoin.org/bitcoin.pdf
    While no idea is perfect, some ideas are useful.
Page 803
Viewing Page: 804