Blockstack: Decentralized DNS for blockchain applications | Hacker News Reader

Blockstack: Decentralized DNS for blockchain applications

github.com

49 pointstaylorwc13 comments

Oh fantastic, it took me a while to understand how this works, but its pretty simple (Blockstack is not an altcoin):

Blockstack nodes simultaneously run the Bitcoin daemon, and when they see a transaction on the blockchain with particular data in OP_RETURN, they add that data to the Blockstack meta-chain. Blockstack nodes have an incentive to mine the Blockstack meta-chain, because they get paid for registration of names.

What I'm interested to know, is whether the Blockstack chain allows for arbitrary key/value pairings, or whether it is intended to be just for DNS resolution. I can see a lot of utility emerging from a decentralized KVP-storage blockchain, where a user can (for example) register a username as a key, and a Bitcoin address as the value.

If something like that is possible, we could finally have a system that solves Zooko's Triangle. Decentralized and Secure are taken care of, and this allows for a user to send funds to a Human-Meaningful address (via a lookup in the KVP chain).

How does this compare to Namecoin? It seems pretty similar.

Name coin is is a seperate blockchain, this exists on the bitcoin blockchain.

This is explained on their website:

https://blockstack.org/docs/blockstack-vs-namecoin

I believe decentralized DNS and PKIs need to happen, but using Bitcoin's blockchain was the wrong way to go here. Ethereum is much better suited for this.

You should probably give some concrete reasons as so far I've seen nothing but hype for two years combined with being spammed with fake private messages on reddit.

Its a bit like asking "why should I use javascript instead of php". Different computer languages are better for different things, like different blockchains are better for different things.

Ethereum comes with a built in state tracking machine and smart contracting language. It already has 98% of the protocol built... there's no need to start from 0 for blockstack. If the DNS/PKIs were on Ethereum, anyone could build applications that interact with it without altering the fundamental protocol. For example, a developer could build a decentralized auction or escrow system for domain names. They could do that without having to alter/adjust the existing blockstack protocol.

Here is namecoin in 8 lines of serpent

https://github.com/ethereum/serpent/blob/master/examples/nam...

Is ethereum even released and working?

Why? Proof-of-Stake coins have legitimate long-term security concerns (see the "Nothing at Stake" problem). For a decentralized DNS system, you want the most secure chain possible in terms of hashing power, and in this case, that happens to be Bitcoin. Unless you believe Vitalik will somehow manage to design a Proof-of-Stake system without those security concerns (which I highly doubt will happen).

Ethereum is not PoS

It will be very soon.

http://stackoverflow.com/questions/33802343/when-will-ethere...

The link you provided does not back up your claim. All it says is that research is happening.

My point is you shouldn't dismiss a consensus algorithm that has not yet been formalized/described. It may be possible, it may be impossible, but keep an open mind.

Why not something like "Dename" or "GNU Name System"? Their is also a project going on within Tor that will be simular to "Gnu Name System" to resolve something like facebook.tor to facebookcorewwwi.onion.

Reply on news.ycombinator.com