Why the name Dot-Bit?

[casascius]

I'm wondering why this project is called Dot Bit. Why not call it Dot Onion?

Calling it Dot Bit hinges the entire project on the dubious assumption that someone will gracefully assign the .bit top-level domain to this project, ostensibly for free, instead of selling it to the highest bidder. What if that never happens? And what does "bit" have to do with this project, besides paying an homage to Bitcoin? What happens if we get a TLD, but instead of .bit, we get .crap? Rename the project?

On the other hand, calling it Dot Onion implies a primary goal of providing DNS resolution for Tor hidden services - something entirely achievable, and requiring only that we do such a good job of building it that the maintainers of Tor and associated software (vidalia, polipo, etc.) see it as a viable feature. Since they're part of the open source community too, I can't see why they wouldn't be able to resist adding native support for Namecoin into those projects. Finally, a naming system that breaks Zooko's triangle, as referred to on the Tor project website.

I registered dot-onion.org (in the DNS system known outside of Namecoin). I would be happy to give it to the project if a consensus formed that this was a good idea.

[jackjack]

Why do you prefer .onion ?
Namecoin IS much related to Bitcoin, but not to Tor AFAIK
Moreover, where does http://eqt5g4fuenphqinx.onion/ will bring you? The Tor site? Or Namecoin one?

[casascius]

It would go to the Tor one, of course, because it is a well formed Tor address.

If it were not, and someone went to a more memorable address like mysite.onion, or silkroad.onion, then namecoin would resolve it.

[jackjack]

As was meaning if someone reserves eqt5g4fuenphqinx.onion on namecoin

Actually I can understand the problem with .bit, but I don't see why it should be replaced with .onion and not something new

[casascius]

Someone could reserve eqt5g4fuenphqinx.onion in the blockchain but it wouldn't make sense to, because the Tor client would see it as a well-formed hidden service address and wouldn't even consult Namecoin before connecting to it, the same way your browser won't consult DNS if you browse to a numeric IP address. And a mature Namecoin registrar client would block the registration of 16-character alphanumeric onion address with a warning that such a name wouldn't work and a recommendation to either change the length or to include at least one character that breaks the well-formedness of the name.

Regardless, the public might understand the project better if some serious thought were put into what it was called and changing the project name. Perhaps the "Decentralized Cryptographic Internet Name Registry". The same registry would be viable for I2P, as well as any existing domain that wanted to point a subdomain at it (e.g. with an NS record). More emphasis on it being a name project, and despite the magic idea coming from Satoshi, less emphasis on it being an offshoot of bitcoin, which unfortunately is what "dot-bit" suggests.

This will help it earn the credibility and seriousness it needs before it would ever be worthy of even a discussion about having its own TLD.

[moa]

Someone could reserve eqt5g4fuenphqinx.onion in the blockchain but it wouldn't make sense to, because the Tor client would see it as a well-formed hidden service address and wouldn't even consult Namecoin before connecting to it, the same way your browser won't consult DNS if you browse to a numeric IP address. And a mature Namecoin registrar client would block the registration of 16-character alphanumeric onion address with a warning that such a name wouldn't work and a recommendation to either change the length or to include at least one character that breaks the well-formedness of the name.

Regardless, the public might understand the project better if some serious thought were put into what it was called and changing the project name. Perhaps the "Decentralized Cryptographic Internet Name Registry". The same registry would be viable for I2P, as well as any existing domain that wanted to point a subdomain at it (e.g. with an NS record). More emphasis on it being a name project, and despite the magic idea coming from Satoshi, less emphasis on it being an offshoot of bitcoin, which unfortunately is what "dot-bit" suggests.

This will help it earn the credibility and seriousness it needs before it would ever be worthy of even a discussion about having its own TLD.

Thnx, but no thnx.

Credibility comes with good code not high falutin ideals and names.

You are plain wrong about someone else taking .bit TLD in the existing system being a problem also .... and we are not worthy of your discussion anyway. It seems to me like you haven't yet understood how the DNS function of namecoin actually works, icbw.

But you could always just call the project whatever-you-like ... its open source naming after all.

[casascius]

Credibility comes with good code not high falutin ideals and names.

Credibility with whom? If I take my existing time and attendance software and call it "Poopstain", it isn't going to be received quite as well no matter how good the code is. Same thing if I call it something that has nothing to do with its functionality, like "Kansas Department of Agriculture". This project doesn't own the .bit TLD, and isn't imminently about to, so I'm going to disagree with you that it's all in the "good" code.

You are plain wrong about someone else taking .bit TLD in the existing system being a problem also .... and we are not worthy of your discussion anyway.

Am I wrong for any particular reason? Or just because you say I am? Unless the entire world is about to go and manually configure their network settings and override their DNS settings (unlikely), or install some utility (unlikely), as though that would somehow be easier for the average user than just going to wikileaks.ch (instead of .org) or pokerstars.eu (instead of .com), the fact that this project doesn't own that TLD on the existing system is sort of an important detail to overlook. It hasn't really been thought through. It solves a problem that doesn't exist. The world isn't demanding a crypto-based domain registry. Think "Jump to Conclusions Mat" (from Office Space).

On the other hand, a significant but small segment of the world demands Tor hidden services, and the average joe who uses Tor would immediately see the benefit. Why not steer the ship to cater to that? Non-memorable Tor names is a problem that does exist.

It seems to me like you haven't yet understood how the DNS function of namecoin actually works, icbw.

That may be... but I will make an equally weighty assertion that you don't understand the mindset or needs of the average joe consumer citizen as it relates to this project, nor why that might be important for it to succeed anything beyond a fringe hobby project.

[johntobey253]

Unless the entire world is about to go and manually configure their network settings and override their DNS settings (unlikely), or install some utility (unlikely), as though that would somehow be easier for the average user than just going to wikileaks.ch (instead of .org) or pokerstars.eu (instead of .com), the fact that this project doesn't own that TLD on the existing system is sort of an important detail to overlook. It hasn't really been thought through.

I've thought it through. The main point of the project is that we don't have to ask permission from anyone. First, someone (you?) releases a working DNS server and/or BIND plugin that reads from the block chain. Second, people like me start pestering their IPSs to support .bit, pointing them to the code and existing nameservers, should they prefer to forward. (I wouldn't pester them at this stage where the service exists but no(?) code is available.) Next, .bit works in practice as nameservers increasingly support it. Finally, ICANN documents this fact, as resistance is futile.