Game Over??

[slynrick]

When we made this smart contract we only focus in the anonymity inside the code, using some temporary hash unique for each process. All code is about the double blinded idea of this contract, however there is one more point that can bring all the structure down.

This point is the transaction itself, even if we treat all the data inside the contract to be anonymous when someone calls the contract to do something inside ( change the process, like send some review ) the transaction will use the address of the person that did that and the anonymity doesn't exist anymore since everyone can trace all the changes using the same process key.

Unfortunately it's something that i don't see any solution .

[vncoelho]

Good insights, @FabioRick !!!

You mean that the reviewer would had been registered by the PubKey and it would be used.

Then, the author could trace the reviewers?

About the authors they could be anonymous, because anyone could just, without register.

[slynrick]

Actually @vncoelho , it can be traced with the process key inside the opcodes of the invocation of a function inside the contract.

I mean, if i'm going to send an abstract to an editor, i must do it by invoking this function.... so even we creating a temporary hash for the author id inside the process, the real address will be on the transaction... so we just need to see that there.

The same happens to everyone that uses the sc. If you're in some process, all the time that u send to the sc the processkey and everything else the real address will be on the transaction that invokes it.

Do you understood? It's not just a reviewer issue. You can trace everyone that is related to this process before it be published.

[vncoelho]

I got it.

You are the master in this things.

But I thing that, at the end, there is no problem, the address the guy is using can be temporary.
They won't know him until the paper is really published.
The name can be signed with the privatekey and only the editor would have it.

Just the editor would know the name, then, after the paper is accepted, he can confirm that the identify was not changed.

[vncoelho]

My last comment does not solve...aehauhueahuea
Maybe two private keys would be needed, right?

One for signing the name at the begin.
This public key would only be revealed at the end!

I thing that I understood your point.

[slynrick]

@vncoelho two keys doesn't solve it. Since the problem is not the sign, the problem is that every time you call the contract to change something in the process your address will be visible in the transaction.

The public key will be visible during all the process, since everyone needs to invoke the contract.

[igormcoelho]

Don't worry, @FabioRick , there are many ways around this. First, any anonymous wallet will keep change balances and values, and will require a different publickey for each transaction on the network. This way, it won't be possible to trace back the two ends.
All these things were in my mind since the beggining, but I haven't put much energy until now because I wanted to have a prototype a least... now that we are evolving, we can deal with these issues.

[igormcoelho]

The second point, it is possible to perform a random selection on the reviewers, even on the blockchain, and nobody will know who they are. We can discuss this later if you want to.

[slynrick]

@igormcoelho Choose the reviewers randomly is not the issue and we already discussed this at the beginning of the project.

I'm not agreeing with you about not be possible to track them all.
Each contract invocation has the address that invoked it. And because of it it can be traceable, and the process key parameter make it easier because its just check the opcodes in the invocation script to see if it's from the same process.
So in order to maintain the anonymity everyone that is in the process and for each process must use a different wallet and those wallets must be different from the one that will be published at the final of the process and it's not good.

Well, i really don't have a real solution for this ( only the one above but it's unproductive ) but since you have many of solutions it won't be a problem.

[slynrick]

As you can see in this transaction that invokes scichain to request article:

https://neoscan.neocompiler.io/transaction/a15ecd2ce46dd0d306182cea88a06c6125748fe25bbaa5960e55f49f46e9b2fe

And all those transactions can be accessed by the address information:

[vncoelho]

This new version ofNeoScan is nice...aehauheua

@FabioRick, I think that a wallet can be generated only for this process.
Thus, who would someone know the identify? They should need to track all previous transactions. That would be the same case of tracking in the BTC.

Ricochet transactions could also help with this.
However, I do not still see this needs of ensuring privacy.

Most paper current subjected to peer-review, nowadays, are still open and the reviewers know who are the authors.

But I really like this discussion/insight. I think that we should move towards this, but not something that we really worry right now...Haduken!

[slynrick]

Ok Guys(@igormcoelho , @vncoelho )!! Let's move on! Maybe we can think about it another time.

[igormcoelho]

This is an important issue Fabio, so I'll reopen it.
I had to take some time off SciChain, in order to focus and finish neocompiler.io platform essentials that we could use here, also some necessary improvements on Neon C# compiler.
I'll soon publish updated information on the randomization algorithm and a formal definition for the AnonyTree: https://github.com/igormcoelho/DRandLib