Re: [Bitcoin-development] Coinbase reallocation to discourage Finney attacks

[Mike Hearn <mike@plan99.net>]

[-- Attachment #1: Type: text/plain, Size: 1341 bytes --]

On Wed, Apr 23, 2014 at 10:44 PM, Adam Ritter <aritter@gmail.com> wrote:

> Isn't a faster blockchain for transactions (maybe as a sidechain) solving
> the problem? If there would be a safe way for 0-confirmation transactions,
> the Bitcoin blockchain wouldn't even be needed.
>

The 10 minute average comes from a desire to balance wasted work due to
natural chain splits with latency. With a very fast block interval you end
up with lots of forks and things take longer to converge, also, it can make
attacks easier because an attacker is building on his own blocks so he
doesn't suffer propagation delays and the attendant splits.

It's not clear you can just make a faster block chain. 10 minutes is
somewhat arbitrary, it could be 5 minutes and the system would still work,
but it probably can't be 5 seconds.

Unfortunately for best physical-world usability you really need very fast
payments. A few seconds is competitive with modern credit cards. The new
contactless cards seem to be able to reliably manage <1 sec which is
impressive. Waiting for blocks in a block chain can't really work. Waiting
for propagation can work and has been working so far. Hence, the question
of how that mechanism can be kept working in the face of malicious miners,
before you end up having to fall back to trusted third parties and
recentralisation.

[-- Attachment #2: Type: text/html, Size: 1750 bytes --]