8

I was just showing a deterministic wallet to a friend who asked "At what point is the wallet known to the network?" and I could only guess by saying that it is known only when a transaction is broadcast and confirmed... But I don't think that is the whole story. I still have some unanswered questions. I think this is because I don't fully understand the cryptographic method in play.

  1. So for arguments sake, we have User A and User B.
  2. User B creates a brain wallet using an offline computer which essentially provides User B with a public key and a private key.
  3. User B tells User A to send Bitcoins to this address and provides her public key.
  4. User A is online so he sends the coins with a signed transaction including fee using his favourite client.

What happens?

How does the network (by this I mean the nodes on the network who are relaying the transaction) know that this is a genuine address, User B is after all off-line?

If no coins are ever sent to this address, does the blockchain ever need to know if this wallet has been created?

3 Answers 3

5

TL;DR: The network is not made aware of address ownership, nor can anyone check if an address is owned unless it has been used previously (or is owned by himself). Anyone can send transactions to any valid (as in correctly shaped) address¹, the recipient doesn't have to be online for a transaction to take place.

Let me elaborate a little bit in order to explain:

Addresses are the public key of an asymmetric key pair

An address is the public key of an asymmetric key pair. The owner of the key pair can use the private key to sign transactions or messages (for example in order to prove ownership). Only by using the correct private key a valid signature may be created, which then anyone can verify as valid by using the associated address (which, again, is the public key).

Wallets are an abstraction

A wallet is an abstract construct, which contains a set of public and private key pairs randomly generated for the user. In a deterministic wallet, the keys are derived from a passphrase instead of a random seed.

The Bitcoin client software abstracts the wallet for the user such that it checks each of the addresses contained in the wallet, whether there are any balances (transaction outputs) associated in the blockchain with them. It sums the funds up and presents them as a single total balance. Underlying, however, these balances are stored publicly in the blockchain (that's right, the wallet doesn't contain your coins, it just allows you to spend them). They are split up to numerous addresses and transaction outputs:

  • Address #15521³ has
    • 3 BTC in transaction output 1
  • Address #23414 has
    • 1.2 BTC in transaction output 1
    • 0.7 BTC in transaction output 2
  • ...

Everyone can see them, however, usually can't determine the identity of the owner.

Transactions are public notifications of change in ownership

Now, when someone sends a transaction, he basically puts up a notice for everyone in the network to see in which he signs over some of his transaction outputs to recipient addresses. He signs this notice with his private key, so that everybody can check that he is the rightful owner of the outgoing balance (as they can see what address the balance is being sent from they know which public key to use to verify the validity). This can happen regardless of the recipient being online: The bitcoins will be available to spend to the owner of the new address, because only he has the private key, whenever he choses to do so.

It cannot be determined if an address is owned by someone

The public keys (Bitcoin addresses) fulfil a specific pattern with checksums, so that bitcoin software can easily assess whether an address is valid. However, unless an address has been used before there is no way for anyone to check whether a valid address has actually been claimed by anyone.

At no point the network is ever made aware of the creation of new wallets, nor does it have to be. Balances are just signed over to a recipient address and subsequently can only be spent with the associated private key. If the new addresses private key hasn't been found yet, it is very unlikely that anyone ever will do so – the coins would likely be lost forever.

¹ By creating raw transactions one can actually send transactions to an invalid address. This is for example used to leave messages in the blockchain or to validate the authenticity of documents.

³ Of course addresses really are alphanumeric strings with up to 34 characters.

1
  • I managed to discover most of this in the interim, but this will no doubt help someone.
    – T9b
    Sep 25, 2013 at 22:35
3

Short answer: It doesn't. All the network knows is that someone sent funds to this address. It doesn't know and doesn't care if anyone has a private key matching a public key matching this address - that is the problem of the user who sent the funds, and the user who requested funds to be sent to this address.

Of course, if user B ever spends these coins (by providing a public key and a signature from the corresponding private key), the network then knows these keys exist.

It is also possible to send to addresses with Kolmogorov complexity so low it is known that no-one has the private keys, such as 1BitcoinEaterAddressDontSendf59kuE and 1111111111111111111114oLvT2

And no, if no coins are sent to an address, the network has no need to know that someone created a keypair for this address.

1

If you mean "genuine address" as in one that someone can spend the funds from. The network only knows someone can spend those funds when they do and broadcast a transaction from that address. The network will happily send to an address that no one knows, e.g. https://blockchain.info/address/1BitcoinEaterAddressDontSendf59kuE

2
  • But how is it able to confirm that the private key used to sign the TX is genuinely associated with the public one to which the coins were sent?
    – T9b
    Sep 10, 2013 at 12:19
  • It checks that when the new transaction is sent. The signature is mathematically associated with the private / public key such that only someone who had the private key would feasibly be able to generate it.
    – user71404
    Sep 10, 2013 at 13:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.