[contingencies]

Last night, at the city-that-shall-not-be-named Bitcoin meetup, we specifically discussed this sort of BIP, and I raised the opinion which was initially posted in response to Gavin's support for BIP70, which is that the last thing Bitcoin needs is more complexity.

Fundamentally, Bitcoin is trying to be a settlement network and an asset and a UI and a platform for developing future financial services and now a half-baked business-oriented transaction layer on top of the settlement network. Unix philosophy[1] says this is insane, and as someone involved with one of the largest exchanges, I'm inclined to agree.

Most of this stuff boils down to using Bitcoin to try to solve things it's not well-suited to solving, eg. instant payment.

I mean .. adding X.509 & SSL[2] to Bitcoin clients to 'simplify' payments? Really? Do you guys have any idea what that does to an implementation's risk model?

[1] http://en.wikipedia.org/wiki/Unix_philosophy [2] https://github.com/bitcoin/bips/blob/master/bip-0070.mediawi...

[BrokenPipe]

To answer to your question, I think your points have already been raised and while I may even share them BIP70 is by default using SSL and X.509 and it seems natural to use it for extensions too.

Other than that, instant confirmation is also done via our own API (which is via wss, ssl) and there is no reason as to why it couldn't be done with ECDSA instead, issue as usual is key management.

You think arbitrage&liquidity between exchanges via instant confirmation is not well suited for exchanges?

[contingencies]

You think arbitrage&liquidity between exchanges via instant confirmation is not well suited for exchanges?

With all due respect, correct me if I'm wrong but I don't believe you run a major exchange. There are loads of factors that need to be considered including legal, fraud, security, timezones, multi-hop settlement times (transactions that involve crossing assets) for which we are perhaps one hop and which rely on speed and precalculable latencies, etc. Basically, if anything at all to do with running and exchange seems 'obvious', you probably don't have the whole picture.

Furthermore, most of these issues are better solved in a manner that is not asset-specific or intimately connected to an asset-specific settlement network, ie. we really don't want Bitcoin to provide its own, half-assed solution for what is not a Bitcoin-specific problem.