does "stubbing" off Merkle trees reduce initial download bandwidth?

Satoshi's paper mentions that storage requirements for the blockchain can be reduced by deleting transactions whose outputs have been spent.

If I understand correctly, this technique can only be used for reducing *storage* requirements, not *bandwidth* needed for the initial chain download by a high-security client that doesn't trust any of its peers -- right?

The rule is "trust the longest valid chain of blocks". Part of a block being "valid" is that each transaction's inputs are unspent and their sum exceeds the transaction's outputs unless it is a coinbase. This cannot be verified for "stubbed out" transactions -- they have outputs but no inputs, and aren't coinbases. So a paranoid client booting up for the first time needs to be given an un-stubbed chain, right?

Of course, if a client decided to accept a stubbed blocks only when the sum of the difficulties in the blocks after it exceeds some number N, then attacking it could be made very expensive by picking a large enough N.

Please let me know if I have misunderstood something.

It can speed up the initial chain download. A newly created wallet will have only new key-pairs, hence no incoming transactions (unless we have a key collision, which is unlikely). So there is no need for a bootstrapping node to download the chain with transactions. The chain itself can be verified without the transactions. Later full blocks would be required to detect usable inputs for future outgoing transactions. As long as you verify the very last blocks in the chain you can be sure that all preceeding blocks were also valid.

HTH, Chris

On Mon, Jan 2, 2012 at 6:04 AM, Elden Tyrell <tyrell.elden@gmail•com> wrote:

> Satoshi's paper mentions that storage requirements for the blockchain
> can be reduced by deleting transactions whose outputs have been spent.
>
> If I understand correctly, this technique can only be used for reducing
> *storage* requirements, not *bandwidth* needed for the initial chain
> download by a high-security client that doesn't trust any of its peers
> -- right?
>
> The rule is "trust the longest valid chain of blocks".  Part of a block
> being "valid" is that each transaction's inputs are unspent and their
> sum exceeds the transaction's outputs unless it is a coinbase.  This
> cannot be verified for "stubbed out" transactions -- they have outputs
> but no inputs, and aren't coinbases.  So a paranoid client booting up
> for the first time needs to be given an un-stubbed chain, right?
>
> Of course, if a client decided to accept a stubbed blocks only when the
> sum of the difficulties in the blocks after it exceeds some number N,
> then attacking it could be made very expensive by picking a large
> enough N.
>
> Please let me know if I have misunderstood something.
>
>
>
>
> ------------------------------------------------------------------------------
> Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex
> infrastructure or vast IT resources to deliver seamless, secure access to
> virtual desktops. With this all-in-one solution, easily deploy virtual
> desktops for less than the cost of PCs and save 60% on VDI infrastructure
> costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development@lists•sourceforge•net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>

On 2012-01-02 05:31:19 -0800, Christian Decker said:

> Later full blocks would be required to detect usable inputs for future
> outgoing transactions.

Er, yes, this is what I meant; I guess I should have been more specific.

So, a paranoid client cannot confirm reciept of coins until it has an unstubbed copy of the entire chain. It can do other things (like send coins) using a stubbed chain, but it needs the whole unstubbed chain in order to be sure that incoming coins haven't already been spent.

Thanks for confirming this.

On Mon, Jan 2, 2012 at 5:23 PM, Elden Tyrell <tyrell.elden@gmail•com> wrote:

> On 2012-01-02 05:31:19 -0800, Christian Decker said:
>> Later full blocks would be required to detect usable inputs for future
>> outgoing transactions.
>
> Er, yes, this is what I meant; I guess I should have been more specific.
>
> So, a paranoid client cannot confirm reciept of coins until it has an
> unstubbed copy of the entire chain.  It can do other things (like send
> coins) using a stubbed chain, but it needs the whole unstubbed chain in
> order to be sure that incoming coins haven't already been spent.
>
> Thanks for confirming this.

Er, no— if a node controls the private keys for a transaction, and that transaction makes it into the chain then it can safely assume that its unspent (at least once its buried a few blocks into the chain). This is the essence of a SPV node.

What it can't do is perform this function for txn which aren't its own. Though the system could be extended in a compatible manner to make this possible: https://bitcointalk.org/index.php?topic=21995.0

On 2012-01-02 14:41:10 -0800, Gregory Maxwell said:

> make this possible: https://bitcointalk.org/index.php?topic=21995.0

Neat! I had a similar idea but you've clearly beat me to [a big part of] it.

> Er, no—  if a node controls the private keys for a transaction, and
> that transaction makes it into the chain then it can safely assume
> that its unspent (at least once its buried a few blocks into the
> chain).

I'm not so sure about that. If you accept X successor blocks as proof that none of the transactions in a block re-used an output, then the cost of attacking is X*50BTC since the hashpower needed for the attack could have earned that much reward.

However, an attacker could use the same faked X-block sequence to attack multiple clients by putting several double-spend transactions in the first faked block. This would spread out the cost over more than one attack. So simply checking that the value of the transaction is less than X*50 isn't necessarily enough, although the logistics of the attack aren't exactly easy.

There's also the question of knowing what the difficulty for those X blocks ought to be. If the attacker controls your network connection (e.g. your ISP attacks you) you wouldn't be able to get a second opinion on how high the difficulty ought to be, and might get fooled by X very-low-difficulty blocks that were each produced with a lot less than 50BTC worth of hashpower.

  - e

This thread is discussing two unrelated things.

Your first email asked about transaction pruning ("stubbing"). You're correct. This doesn't do anything for initial chain download bandwidth or time. In fact it makes it slower because you have the overhead of deleting the old transactions. It exists purely to save disk space.

Christians reply is about simplified payment verification (SPV) mode. It is unrelated to transaction pruning. SPV clients can download only the chain headers with no bodies all the way from the genesis block until the creation time of their youngest key. This does reduce initial setup time and in fact is now implemented in BitCoinJ, but it's still linear in the length of Bitcoins life, so that's ultimately unsustainable. You need a regular series of checkpoints signed by a trusted developer and a circular block store to have truly bounded overheads. The merkle tree is still useful because it allows for SPV clients to receive only the transactions of interest yet have nearly the same assurances that downloading full blocks would give - remote nodes can now hide transactions from you (dos) but not invent new ones.

SPV clients do not use "number of blocks on top" as a way to decide validity. They look for the best chain they can find, same as a regular node does. As Satoshis paper says, if an SPV node has access to the P2P network and is also talking to you, you can defraud it for as long as you can dominate the networks hash power (51% attack) because you can create a harder chain than everyone else can. However your invalid blocks won't be accepted by the rest of the network regardless of how many there are or how much work they represent, so as soon as you stop dominating the network the correct chain will catch up and replace yours, resulting in the fraud being detected and shown to the SPV user.