Quote from: eldentyrell on October 16, 2013, 12:00:28 AM

Quote from: retep on October 15, 2013, 02:55:12 PM

Quote from: eldentyrell on October 15, 2013, 02:12:29 PM

Quote from: theymos on October 15, 2013, 02:07:23 PM

I'm not sure that a block chain is required at all

What prevents double-transfer of names, then?

The Namecoin v2.0 rules.

If by "namecoin v2.0" you mean luke-jr's proposal, that does require a blockchain.

No, I mean a hypothetical protocol that doesn't use a separate blockchain.

Quote from: eldentyrell on October 16, 2013, 12:00:28 AM

Quote

the next namecoin should be implemented on top of the bitcoin blockchain.

Well, good luck convincing people to use it. I'm not very interested in spending money or computing power on an entry in a key-value store that could suddenly become frozen due to small outputs suddenly becoming unspendable or some other change to the bitcoin block validation consensus.

Bitcoin works because everybody has a common interest in it being usable as money. If you're trying to use it for something else you will always be at risk of the majority (who don't care about your weird use) tweaking the rules in a way that breaks it.

You just don't get it: it's impossible to stamp out all the ways of timestamping data in the blockchain, and timestamping w/ bitcoin sacrifices is sufficient to build a useful and secure distributed DNS system. Even a majority of miners can't use blacklists to stop timestamping and bitcoin sacrifices because you can devise protocols with very useful properties where the transactions involved are revealed after they have been confirmed; they've have to whitelist all acceptable transactions, which is close cousin to very strong censorship.

Such a system can't replicate the scamcoin-like aspect of namecoin where you see speculators hoping the price goes up of course, but notice how namecoin has never caught on...