[tor-talk] Gmail and Bitcoin? [OT]

Brandon Wiley brandon at blanu.net
Mon Jan 14 18:53:30 UTC 2013


I like where you're going with this, Mike Hearn. I've been working on a
similar scheme in which instead of providing a proof of work you provide a
proof of sacrifice of monetary value. My simplest plan is that give a nym
authority a payment in exchange for a new nym. However, this requires
trusting the nym authority to follow the rules. Specifically, the nym
authority needs to hold onto those payments because giving them back to the
buyer results in a cycle yielding free nyms. Authorities can go back in
other ways too, such as giving out free or discounted nyms, or raising
prices in order to generate higher profits.

I like the idea of using transaction fees as proof of sacrifice because
they are randomly assigned to miners. The nym authority could just watch
the transactions and would have no economic stake. However, if you are
already doing a high volume of transactions then you could possibly reuse
your existing transactions for your proof of sacrifice without actually
sacrificing anything extra.

Here is another variant on this idea. You have a non-profit nym authority
that takes payments and then redistributes these payments. Redistribution
could be (certifiably) random, to the bitcoin addresses of known
non-profits, or you could burn the bitcoins by sending them to an address
with (certifiably) no private key. Since bitcoin transactions are easy to
track, it could be verified by third parties that the nym authority is
following the protocol and redistributing in a way that disperses the fees
and doesn't lead to free nyms or expensive nyms.


On Mon, Jan 14, 2013 at 11:52 AM, Mike Hearn <hearn at google.com> wrote:

> Feel free to CC me on such discussions, if you like. Google isn't
> against Tor. We just wrestle with the same abuse issues every other
> provider does.
>
> As Mike Perry notes, targeted advertising doesn't actually know or
> care about your real identity. It just wants to know that user 1234 is
> more likely to click ad ABC than an DEF. And even for Google+, these
> days you can register with a pseudonym if you want, there are policies
> around proving that you really go by that pseudonym online so I don't
> know how Tor-compatible it is, but the intention is there. The goal of
> the "common names policy" (not actually a real names policy) is just
> to ensure users don't sign up en-masse with names like PinkPony84 and
> thus become a MySpace-style swamp of people who can't find or connect
> with each other.
>
> Anyway. As grarpamp notes, one of the problems with taking deposits is
> suddenly you're taking payments and that comes with a host of
> complicated accounting and legal issues. At least it does in the USA.
> I was thinking lately about an alternative approach. Instead of
> depositing some coins directly with the organization you want to
> upload content to, you could establish a nym by making a non-trivial
> sacrifice of coins to miner fees. By breaking up the sacrifice into
> multiple payments spread over many sequential blocks you can put
> together a proof of sacrifice that is hard to fake (eg, by mining your
> own transactions). The coins aren't burned, they re-circulate into the
> economy.
>
> Once that proof is assembled, the hash of it can be your new nym, and
> anyone who wants to accept it can verify the merkle branches included
> in your proof up to their own copy of the block chain headers. If a
> nym misbehaves, you can ban it/add it to a blacklist/etc. All the data
> needed for a provider to calculate the size of the sacrifice is easily
> included in the proof data.
>
> If somebody wanted to actually pursue this idea, as I said before, I'd
> do it by starting with MediaWiki. It's open source and widely used. If
> you can make wiki-spamming hard you have a solid example of what other
> service providers can do. You can use my own bitcoinj library to
> implement it in a simple manner. If you ask on the bitcoinj mailing
> list, or Bitcoin dev forums, I'll happily sketch out any design
> details that are unclear.
>
> Bitcoin and privacy is a complex topic perhaps best discussed in
> another time and place. Whether it's strong enough will be something
> each user has to judge for themselves, same as for Tor itself. Of
> course, using Bitcoin does not exempt anyone from their jurisdictions
> AML laws, and nothing stops a merchant demanding government issued ID
> in order to trade with them.
> _______________________________________________
> tor-talk mailing list
> tor-talk at lists.torproject.org
> https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
>


More information about the tor-talk mailing list