Open Name System: Extending DNS with the Blockchain | Hacker News Reader

Open Name System: Extending DNS with the Blockchain

opennamesystem.org

135 pointsmuneeb29 comments

ONS is using Namecoin. Namecoin isn't free to use, is it? I'm curious to know what is the business model behind this and onename.io, since they must pay to add entries to the Namecoin blockchain.

How is this different from DNSChain?

Having looked at it a bit more, here are some differences that I noticed:

    - This is written in python. DNSChain is written in CoffeeScript and runs
      on NodeJS.
    - This focuses on a command line utility, whereas with DNSChain you don't
      need one (just do an HTTP request).
    - opendig is like dig, so it doesn't provide your devices with DNS service
      (DNSChain does).
    - DNSChain focuses on security, signing of responses (though that's not
      implemented yet), replacing X509 PKI, unique anti-censorship features
      thanks to partnership with Unblock.us.org project, public key
      distribution for JavaScript apps... See the full list on the GitHub
      page.
    - ONS seems to be working on competing blockchain specifications to
      Namecoin's existing ones. DNSChain doesn't touch that.
    - DNSChain intends to add write access to the blockchain. It's not clear
      whether ONS project does or not.
    - Namecoin and DNSChain predate ONS and are active projects with different
      functionality that ONS seems to be replicating parts of, so I'm not
      sure why the authors chose to not collaborate with either of those
      projects and tread off on their own.

This is a pretty direct challenge to keybase.io. I really, really like the keybase implementation, but the one thing I wished for keybase was that it was decentralized.

This has that.

However, in terms of the implementation itself, they still have a way to go before they match keybase.

> However, in terms of the implementation itself, they still have a way to go before they match keybase.

Care to explain how? They're the team behind onename.io and clearly seem to have just as strong an implementation.

I totally agree that onename.io has done a good job integrating that site with Namecoin. However, with this new project ONS, I don't yet see a way to certify all the social identities that Keybase supports, not to mention the attractive user interface and convenient command line client of Keybase.

That said, it's a new project. I'm not saying onename.io won't accomplish this (I hope they do!). I'm just saying they have a lot to do to match what Keybase has.

In addition to creating a website for this, you really should submit it to the IETF as an RFC. Get more engineers looking at it and improving it. If it ends up getting enough backing, we might even have a new standard.

There are specifications that I'm sure will be submitted to IETF as RFCs (when they're ready). You can seem them on Namecoin's wiki (scroll to the bottom):

https://wiki.namecoin.info/index.php?title=Welcome

This is the first attempt I've seen at using a blockchain for DNS. Looks like an amazing solution to an increasingly desperate problem (as DNS takedowns are the preferred method for internet censorship).

One of the first uses of blockchain technology for a non-financial application was an alternative domain name registration system called Namecoin, started back in 2010 and still active.

Granted, that was an alternative domain name system, not an ICANN extension like this one. And it used a separate blockchain and coin (Namecoin), not the Bitcoin blockchain, which may be what you were referring to.

Nonetheless, I feel compelled to mention Namecoin in any discussion about the history of blockchain technology and DNS. (incidentally, Namecoin was also the first altcoin)

This thing runs on Namecoin, though that is buried in the documentation.

Ah, you're right. I thought they had moved to the Bitcoin blockchain. But that's a whole can of worms, so it's probably best they stay on Namecoin.

Bar the big incident last year (which it did recover from), Namecoin has been pretty stable.

Buried on the front page

Wasn't mentioned at all when my previous comment was posted.

Are you familiar with NameCoin?

https://en.wikipedia.org/wiki/Namecoin

https://github.com/BrendanBenshoof/P2PDNS/raw/master/P3DNS.p...

Namecoin has been used for this purpose since 2011. It already has specifications for DNS and identities (and these have been iterated on for years now):

https://wiki.namecoin.info/index.php?title=Domain_Name_Speci...

https://wiki.namecoin.info/index.php?title=Identity

As well as others (like hashes for software). Even OneName.io's u/ namespace is listed here:

https://wiki.namecoin.info/index.php?title=OneName

There are also many other blockchain DNS related projects:

DNSChain (gives full-security read/write to arbitrary blockchains for all devices without having to run a DNS-based blockchain node, another nifty features like censorship circumvention and dotDNS metaTLD for JS apps):

https://github.com/okTurtles/dnschain (I'm one of the devs of this)

BitShares DNS (a different type of blockchain with a different consensus protocol): http://dotp2p.io

Feel free to mention others. I hope for more collaboration and less reproduction of efforts.

OneName and OpenNameSystem appear to be trying to avoid any association with Namecoin despite the fact that they use it. Both projects have no mention of Namecoin (they just say "blockchain") on their websites and a lot of the documentation isn't very clear on this. What gives?

Are we looking at the same website, or have they changed it since you made your comment? "ONS currently uses namecoin as a key-value store but the system is blockchain-agnostic" is on the first screen after the fold for me.

I'm 99% sure that wasn't there when I made that comment - I had done ^F namecoin to check.

I think the reason they don't say Namecoin explicitly is because at the end of the day, it doesn't particularly matter which *coin's blockchain gets used. At this point in the course of distributed system design, a "blockchain" can be thought of as a building-block for a system, and the matter of which particular blockchain gets used is an implementation detail.

>I think the reason they don't say Namecoin explicitly is because at the end of the day, it doesn't particularly matter which *coin's blockchain gets used.

I disagree. It matters a lot how much hashpower a given Blockchain has, else it can be easily destroyed (through forking) and/or interfered with (50% attacks). So Bitcoin has by far the most secure blockchain.

> it doesn't particularly matter which 'coin's blockchain gets used.

Well, except that it does, because depending on the blockchain used, you are dealing with different consensus protocols, and most importantly, different people. Namecoin is a worthy project, has been doing its noble thing for quite some time, and deserves at least a mention if this project is going to be working off of the sweat its core devs have put into it (and continue to do so).

Is Naval behind this?

Why should we base critical infrastructure on completely anonymous building blocks? There is no mention about who is working on this, their motivations, the domain is behind WhoisGuard, etc. Why is it so critical to become anonymous as a developer in this case? Perhaps I'm missing something but is the FBI/CIA going after anyone working on blockchain/crypto stuff nowadays?

It feels like asking for transparency, but only from others. Same reason I can't truly trust Bitcoin, it's as shady as my bank's processes.

>It feels like asking for transparency, but only from others. Same reason I can't truly trust Bitcoin, it's as shady as my bank's processes.

What additional information would get get from the name of the people behind this? How is Bitcoin in any way shady? It's an open protocol and most people use open-source clients to access it. You (and anyone else - including people with a lot of money riding on it) can verify that you trust the code, even if you don't have any idea who wrote it. Compare to a bank's process, where if you even try and reverse engineer the protocol you're risking jailtime.

> Perhaps I'm missing something but is the FBI/CIA going after anyone working on blockchain/crypto stuff nowadays?

There's some evidence they're monitoring people who are interested in Tor and the Tails Linux distribution. There's very good reason to assume that working on crypto stuff gets you on a list.

At the bottom:

"currently maintained by the team at onename.io"

https://onename.io/about

There's also links to the github repos.

I like the idea they're going for here, but I agree. For something like this to work, it should be as transparent as possible and easy to audit.

Could you elaborate on your point on Bitcoin? I'm not an expert, but I have some experience with it and it seems to be pretty thoroughly open, transparent, and auditable. It's certainly more clear than banking processes.

Reply on news.ycombinator.com