[lug] Anyone interested in .bit domain names?

[Sean Reifschneider]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 11/02/2011 10:03 AM, Davide Del Vento wrote:
>>> Interesting. Two questions I didn't find answered in the wiki: 1)
>>> Does Google crawl them?
>> 
>> If you publish them as example.namecoin.us, definitely.
> 
> Isn't that just a second "ordinary" name? Like having example.org the 
> same as example.com?

Yes, it is a *MIRROR* of .bit, so that people can get to those names
without having to configure their systems to be able to resolve .bit
domains directly.  .bit is not published by the root name servers yet, so
either you must configure your resolver to understand them, or use a
mirrored domain.

Unfortunately, it looks like namecoin.us is not working right now, so that
option seems to be not available.

> How that would change with bit domains where there isn't any 
> authoritative name server? How can uniqueness be enforced (if it is? if
> not, how can it work without?)

Uniqueness is enforced by the peer-to-peer network that maintains the block
chain, which is secured by computing power solving hard problems as part of
that block chain.  It uses the same mechanisms that bitcoin uses to ensure
that transactions are unique, so look for details there if you want the
low-level details.

> details. I mean, a "domain without a central registration authority" 
> seems cool, but why is it a desirable goal? Wouldn't open more

Have you followed the news about ICANN changing it's policies to retract
domains from people who have registered them?  Or about the DHS
confiscating (was it tens or hundreds of thousands of) domains, many of
which were completely innocent?

> problems such as companies registering each other trademarks without 
> possibility of appeal?

Of course there's appeal, through the normal court systems.  What there
isn't is the ability to "big foot" someone without following due process in
the courts, for example like when Apple took newton.com from Mark Newton.

> What if we both register SeanReifschneider.bit at the same time, before
> the information is propagated through the

If we truly register it at the same time (in the same block), I don't know
exactly what happens.  I think it's a conflict that is rejected, though
there may be some tie-breaking.  But it does handle this, this is the same
mechanism, as I understand it, that is used if someone tries to
double-spend bitcoins.

If by "at the same time" you mean a larger time frame, like the same day,
whoever is in the block chain first will get the name and the transaction
trying to get it later in the block chain gets rejected with a refund going
to whoever made the request.

> sidewalk, you give me a cup of soup.... so there are several things I 
> don't grasp (such as the "printing money" ability that everybody is given
> under the bitcoin model).

It's not so much "printing money" in Bitcoin, it's that you are doing
computation work, solving hard problems, in exchange for a 50
bitcoin/namecoin reward if you solve it.  So it's the same thing you're
talking about, but instead of shoveling your walk, I've purchased a
graphics card and run it doing these computations, and the more
computations that are going on globally at a given instant, the more secure
the entire network is.

This is because if one entity controls >50% of the computations going on,
they can fork the blockchain.  Having that computation farm be 100,000 GPUs
means that you need to throw more than 100,000 GPUs at the problem to fork
the blockchain...

> Ok, so what namecoind supposed to do? I couldn't find any explanation 
> either (maybe I'm just having a bad-web-search day :-)

It communicates with others on that peer-to-peer network, gets a copy of
the block-chain, and allows you to manipulate in various ways including
generating addresses that namecoins can be sent to, publishing requests for
names, getting transactions relating to transactions to your addresses,
publishing transactions you make to the peer-to-peer network.

Sean
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iD8DBQFOsbvRxUhyMYEjVX0RAgi6AJ90XOffHw99/ktiHtYeY9ifElMfjQCdFJv6
4pc54bJHpccLLNE0bIWwRp0=
=fIqK
-----END PGP SIGNATURE-----