Quote from: becoin on July 07, 2011, 01:06:16 PM

Quote from: elggawf on July 07, 2011, 12:48:13 PM

You think they should wait until SHA-2 is broken to start looking for a replacement? Think about that for a second.

If they think for a replacement there is a reason, right? If they think for a replacement after 2012 why you shouldn't? Or you intend to use SHA-2 until 2140?

Yes, but you don't understand the reason. Designing a new hash is not something you just throw together over a weekend with a few beers and a whiteboard. If they waited until there was a credible threat to SHA-2, then that's waiting too long - the algorithm would be broken before a replacement was ready and proven strong.

So instead, they leapfrog the standards and it's merely up to users to implement it when their risk assessment decides it's time to do so - Bitcoin would be no exception: when SHA-256 is not as useful as one of it's replacements, I'm sure the devs will begin the difficulty and thorny process of replacing it in the Blockchain.

I highly doubt they'll switch just because NIST declares a great new algo and it looks shiny.

Quote from: becoin on July 07, 2011, 01:19:09 PM

So, better sooner than later because if later the mess will be bigger!

I don't think doing it now, versus in 5 years makes it any less of a nightmare, personally. It's still going to take the cooperation of the network, and I don't think making the network more diverse makes that significantly harder. On the flipside, a giant clusterfuck of changing the algo might be another cataclysmic event for a digital currency that's taken a pounding lately.

The only real benefit I can think of is that you wouldn't fuck over the people who are building ASIC farms if you did it now - I think that's their problem though, not the network's.