Topic: [ANN][DASH] Dash (dash.org) | First Self-Funding Self-Governing Crypto Currency - page 6419. (Read 9723858 times)

Can you explain more? I don't understand.


Depends. Because 50% means that your anonymity set is reduced by 50% on each round as I explained in my other post above.

Example. If you are mixed with 10 others on each round, then only 5 will be anonymous (and one of the five might be you), so that means have 50% + 20% (1 in 5) chance to be non-anonymous. So 70% per round. You will need more rounds or you need larger mix sizes.

Also if it is same 10 you are mixed with every round (or any overlap), then anonymity is reduced. If always same 10 on every round, then you attain no better than 20% non-anonymous no matter how many rounds you use.

Also you have to factor in the non-anonymous rate of Tor and those inputs who didn't use Tor at all are not anonymous. This reduces your anonymity set, even if you use Tor.

Awesome discussion in here today, really enjoying this and beginning to understand DarkSend more clearly now.

I address this in the whitepaper, I propose some users run a script to add entropy to the pools and push transactions though:

Ok, gotcha. That could be mitigated in a similar way by the community running scripts to act as inputs to push DS transactions through. I think Evan suggested this a while back.

Based on these numbers (despite not factoring in sybil inputs), it seems clear that a high level of anonymity can be achieved by increasing the number of pooling stages to 10+, even if the attacker controls > 50% of nodes.

The whitepaper has my proposed solution to that in the "Defending Against Attack" section: http://www.darkcoin.io/downloads/DarkcoinWhitepaper.pdf

Things have changed since then, so we'll have to come up with something else.

And there is another complexity to factor into your calculation. If there is a 58% chance (as you suggested) that each input into each of your mixers can be non-anonymous, then your anonymity set is reduced by 58%.

No you misunderstood my point. I mean the participants who are sending inputs to the CoinJoin mix. Those inputs can be Sybil attacked. If you are the only non-Sybil input, then your output is known with 100% certainty.

If there are 50% Sybil inputs, then the anonymity set of outputs that you are mixed with is reduced by 50%.

Anoncoin employs the i2p which is a low-latency chaum mix net similar to Tor but operates at a lower network stack layer protocol (IP). I don't think it will really resolve the weakness of Tor, because they have same the problem that low-latency mixnets can be analyzed with traffic analysis.  I don't know if i2p is p2p wherein users are the nodes, instead of others providing expensive servers for free. We need something like that, because I don't trust the Tor servers. And we need more hops. Tor is only 3 (or 5 for Tor service).

For transaction mixing, I am trying to see if we can find a way to make CoinJoin work better. The only other thing available is Zerocash or Zerocoin.

That is factored in -- in fact that's the point of this calculation. The assumption being made here (for the sake of getting some hard numbers): 1410 sybil nodes, 1000 non-sybil nodes.

We only need one non-sybil node in the pooling chain to retain anonymity. The longer the chain, the greater the likelihood of this.

[edit] or perhaps you mean the participants in the pooling -- in which case, I don't see how they can affect anonymity one way or the other, since trust is only required of the master node.

You must also factor that your participants might be a Sybil attack. In that case, the number of rounds doesn't help you increase the anonymity set nor decrease the percentage.

Agreed we have really big problems coming to all of us. And yes the dying debt collapse of socialism is really idiotic. I wish society hadn't done that. Too late now.

I want 999 in 1000 anonymity minimum (only 1 in 1000 not anonymous).

1 in a million can be achieved with a solution like this:

User 1 -> Change Address 1 (master node 1, tor ip 1)
Request new tor IP (which Darkcoin could do automatically)
Change Address 1 -> Change Address 2 (master node 2, tor ip 2)
Request new tor IP
Change Address 2 -> Change Address 3 (master node 3, tor ip 3)
Change Address 3 -> Change Address 4 (master node 4, tor ip 4)
Change Address 4 -> Change Address 5 (master node 5, tor ip 5)
Change Address 5 -> Destination (master node 6, tor ip 6)

I'm getting very sleepy. I meant to write above "anonymity isn't destroyed". I corrected my original post that you quoted from.



Of course. I understand that too. That is why I was reasonably sure (educated hunch) you had done something to reduce anonymity in order to make it work more deterministically. That is why I had been negative on CoinJoin and looking for something totally different for a solution.

If there is any way we can improve what you have, then I am trying to see if we can brainstorm it.



In my prior message, I explained how you can isolate the input that is stalling the collateral payment submission in the 2nd stage (you move the collateral submission to the 2nd stage). This fixes the anonymity.

But the problem is how can you blacklist that input? Don't you have to trust the master node not to lie to the other master nodes?

But aren't you also in your current design trusting the master node not to steal the collateral inputs?

Why not Pow and PoS, where PoS is tied to fees earned from serving the nework?

ASIC for bitcoin is an arms race because people are spending cash on hardware trying to make as much as possible. Why not have a similar system for nodes? Create an arms race for being honest nodes and deploying as many as possible to fend off or diminish attacks?

If you are worrying about that you have alot bigger problems!

This wasnt designed for that! Thats just idiotic.

So this is exactly the kind of question that informs a risk analysis. We might ask, "what percentage risk am I willing to take that my anonymity is broken in this transaction, given [set of assumptions]?"

And the solution might be as simple as specifying the required number of stages of pooling in the client to meet our acceptable level of risk.

[edit] the numbers I chose for the 20% calculation are fairly arbitrary and probably not realistic, but going with those same numbers, it'd take ~17 rounds of pooling to achieve a 0.01% chance of anonymity being broken for a given transaction.

hi
Off-topic(but important)
We (crypto) are "fighting" against this > BIS (bank for international settlements)

https://www.bis.org/

Do you ever heard about it?

http://www.investorsinsight.com/blogs/what_we_now_know/archive/2006/03/07/the-most-powerful-bank-you-ve-never-heard-of.aspx

"it helps central banks construct and implement financial policy decisions, in concert with one another. And it acts as a third party in transactions, facilitating the flow of money and other financial instruments, including gold.
It accomplishes this through control of currencies. It currently holds 7% of the world's available foreign exchange funds, whose unit of account was switched in March of 2003 from the Swiss gold franc to Special Drawing Rights (SDR), an artificial fiat "money" with a value based on a basket of currencies (44% U.S. dollar, 34% euro, 11% Japanese yen, 11% pound sterling)."

The bank also controls a huge amount of gold, which it both stores and lends out, giving it great leverage over the metal's price and the marketplace power that brings, since gold is still the only universal currency. BIS gold reserves were listed on its 2005 annual report (the most recent) as 712 tons. How that breaks down into member banks' deposits and the BIS personal stash is unknown.
By controlling foreign exchange currency, plus gold, the BIS can go a long way toward determining the economic conditions in any given country.

In 1974, the Basel Committee on Banking Supervision was created by the central bank Governors of the Group of Ten nations (now expanded to twenty). The BIS provides the twelve-member Secretariat for the Committee. The Committee, in turn, sets the rules for banking globally, including capital requirements and reserve controls. In a 2003 article titled “The Bank for International Settlements Calls for Global Currency,” Joan Veon wrote:

“The BIS is where all of the world’s central banks meet to analyze the global economy and determine what course of action they will take next to put more money in their pockets, since they control the amount of money in circulation and how much interest they are going to charge governments and banks for borrowing from them. . . .
“When you understand that the BIS pulls the strings of the world’s monetary system, you then understand that they have the ability to create a financial boom or bust in a country. If that country is not doing what the money lenders want, then all they have to do is sell its currency.”

http://www.globalresearch.ca/the-tower-of-basel-secretive-plans-for-the-issuing-of-a-global-currency/13239


Do some research about it ( if you want) and be prepared...
my 2 cent.

x11 is my favorite, and I'm sure my gpu's feel the same way.

noticed you guys listed it as 0.9.1 version while it is still on old 0.8.6
It is not even 0.9.0

seems misleading, is their any update on when we can expect 0.9.0.x core or newer version.

Yes, I've read so. Are they using anything good? In general, do you have any suggestions on what might be good for the purpose of a cryptocurrency network?