[Bitcoin-development] Double spending and replace by fee

Peter Todd pete at petertodd.org
Tue Apr 21 11:37:14 UTC 2015

On Wed, Apr 08, 2015 at 11:28:08PM -0700, Adrian Macneil wrote:
> Fwiw, Coinbase relies on the current first-seen mempool behaviour. Wide adoption of RBF (without a suitable replacement available) would make it extremely difficult to pitch bitcoin as a viable alternative to credit cards payments to large merchants.

Some questions:

1) Are you contractually obliged to accept zeroconf transactions with
   existing customers?

I keep hearing rumors of this, but would like some confirmation. In
particular, it would be good to know if you have the option of turning
zeroconf off at all, contractually speaking.

2) What are your double-spend losses to date?

3) Are you actively marketing zeroconf guarantees to new customers?

You're API is a bit unclear as to what exactly those guarantees are;
looks like they only apply if a merchant has "convert to fiat" turned

4) What are your short, medium, and long term plans to move away from
   dependency on "first-seen" mempool policy?

e.g. hub-and-spoke payment channels, Lightning network, off-chain, etc.

5) What is your plan for new Bitcoin Core releases that break zeroconf
   via changed tx acceptance rules?

Basically every release we've ever made has added a zeroconf exploit due
to different tx acceptance rules. (e.g. my 95% success rate last summer)

6) What are your plans for Bitcoin Core releases that fundementally
   break zeroconf?

For instance changes like limiting the mempool size create zeroconf
vulnerabilities that can't be avoided in many situations. Yet they may
also be unavoidably needed for, for instance, DoS protection. Will you
oppose these improvements?

7) If a mining pool adopts adopted policy that broke zeroconf, e.g.
   replace-by-fee, would you take any action?

8) Would you take legal action against a mining pool for adopting
   replace-by-fee publicly?

9) Would you take action against a mining pool who is mining
   double-spends without explanation?

e.g. one that claims not to be running non-Bitcoin Core policy, but
keeps on mining double-spends.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 650 bytes
Desc: Digital signature
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150421/af46a460/attachment.sig>

More information about the bitcoin-dev mailing list