Among other methods that are unknown and unknowable, an industry will develop to own the masternodes or pay masternode owners for their data and sell it or use, just as an industry has developed to collect, sell, trade, and use web tracking data

It's all nice and cute how people run their own masternodes on raspberry pi's today, but over time, assuming a relevant level of success, that will certainly all shift to a commercial model, and those commercial operators will maximize profit by collecting and selling data, the same way the web industry does.

We don't know in advance what that industry will look like just like no one in 1995 no one could have created this graphic showing (part of) the structure of the web tracking industry:

You are overstating the effect of masternode blinding. It just requires data from more masternodes to correlate transactions. There is not one masternode that knows everything but getting data from enough masternodes will still allow you to piece it together. Sort of like multiple rounds happening at once (which is one of the stated goals, since people waiting hours for their coins to be ready to spend is pretty bad).

As for the argument about it being illegal, I highly doubt it is illegal to use data sent to you over the internet for tracking purposes and I also highly doubt that governments are going to pass laws to mandate that masternodes behave themselves. It's like a million times more likely that governments actively work against such a system than try to protect it.

Get back to me when there is a white paper, but as I said, I expect all or nearly all masternodes to be commercial operations that are more interested in maximizing profit than supporting your particular idealistic view of how it is supposed to work.

I would call it a conclusion based on an analysis of the incentives.


I'm still trying to figure out how you think this is going to be illegal. If you are relying governments to pass laws regulating masternodes, they will certainly also require that the data be available for what amounts to "lawful intercept" and probably a whole bunch of other things that seem like completely defeating the purpose. Which means you are basically at the same level of privacy as a bank (i.e. safe from neighbors), with a system that is enormously less efficient.

Let me give you thought exercise on the self-compromised nodes.

Let's say I ran a masternode right now and posted real time transaction data on a web site. I could put ads or trackers on the web page and generate revenue, like a chain explorer. Even if the data by itself wasn't that useful, people might still be curious about it, so I'd get traffic. (Of course I could promote on twitter, post links to interesting transactions, etc.)

In fact, given that this data doesn't even really expose anyone (after all its just one masternode, right?), it wouldn't hurt the price of DASH, so my collateral isn't impaired. Nothing has been really compromised.

What would happen?

As far as I can tell, I'd continue to get my masternode payments, people would still use my node, and I would get extra revenue from traffic on my masternode explorer.

I guess Evan could do something to ban my node from the list or something. Maybe miners could do that if they even cared, but its not clear they would. Then I could just create a new one I think.

bump

The incentive for MN operators is to cooperate on this.
Anyway MN binding will make no single masternode will have knowledge of all the inputs and outputs on a round of Darksend. So what is bad in a scenario like the one you just drew where MN operators sell their data (assuming someone is willing to pay/visit your website)?

No collusion is necessary. Each individual masternode owner has the same incentive on his own to get some extra revenue.


It doesn't compromise anything if I do it, so why shouldn't I?


No sharing is required in this thought experiment. Just a web site. In fact maybe I put my web code on github as open source.


Already discussed.


Sounds like whack-a-mole to me. This sort of incentive failure is inherent in the design.


I disagree.

Interesting discussion. Thanks for the responses.