Continuing the discussion from CCEDK withdrawals impaired, because this is an interesting question in it’s own right and hopefully will spawn a larger discussion:
It is possible, but we’ve discussed it in the past and every time it comes up I’m reminded of the major challenges that such a system would introduce:
Voting is a relatively quick process, in general, but in this context would appear nearly glacial in response to the actual timeline that the network would have to respond. That leads to the primary problem with this approach – during the time that the network is deciding what to do other transactions are happening and NBT and NSR are moving throughout the system.
Let’s assume that you could move the network fast enough to a decision – how do you prevent damage to parties that took possession of a portion of the stolen coins in exchange for their (legitimately provided) goods? “Clawbacks” work in some cases, but I expect that there would be a significant chilling-effect on usage if it was true* that the network could decide in a future time to blacklist certain addresses or unilaterally decide to void transactions. * For those only skimming, or trying to take this out of context, it’s not a current function of Nu and is only used for example in this scenario.
“Targeted” roll-backs could be used, where only the address in question and transactions after a point of “blockheight = X” are excluded, but it isn’t simple to do, from what I understand, and requires a change to the clients that a majority of the nodes in the network are running.
Beyond the technical challenges, there’s a ethical challenge that I believe is even greater. In all cases that I’ve seen, it boils down to a “he said, she said” situation. Exchanges are attacked (or maybe they aren’t…) and funds are moved to an address. It’s very difficult for any one person to gain access to enough information to be able to make a determination of what really happened. I don’t expect that the exchanges (or, eventually, vendors, and even individuals) to be forthcoming or even capable of providing enough clear information that would allow shareholders to vote for a motion to invalidate transactions.
I believe that solutions will arise that allow “charge-backs” on the blockchain. I have no doubt that these will start as third-party solutions rather than protocol-level features, and expect that the team or company that solves this technical hurdle will find widespread success in the B2B and B2C space.
I don’t, however, believe that it is appropriate for shareholders to be involved in that process. The level of granularity becomes too fine for a consensus-based system to work. Voting is useful important for strategic, large-reaching questions (and to validate the global ledger); I find it far less suited for tactical decision making.
What do others in the Community believe is the role of shareholders to influence through voting for protocol-enforceable actions?
There is no way to revert the transaction in the sense that shareholders would induce a new transaction that transfers those coins back. The only way I see is to remove the transaction from the ledger.
That means we would do a roll-back as you describe it and we would need to have a new chain with larger stake ready for this case (otherwise the new chain will only become valid at some point in the future). The other possibility would be to do a hard fork that forces all clients to ignore the outputs at the receiving address (they basically would get burned) and that creates new coins at the sending address. Note that all this has nothing to do with Nu and can be done with every coin as long as consensus can be established. BTC had a roll-back once in the past.
I cannot agree more and the shareholders certainly are not provided with the required amount of information to handle the situation accordingly, which is why this here is exactly the way to go in my opinion:
You want to be able to charge back? Fine, use an escrow (and pay the fee) and if there is trouble this escrow has the duty (and probably also the legal possibilities) to figure out the truth.
Okey so charge backs are not a rational solution. It doesn’t mean we can’t do anything. We could freeze the stolen coins and require the owners of the coins to prove their identity. This way “good” people would get to keep their coins and people who fail to verify their identity are probably thieves.
Also, chargeback functionality should exist but should almost never be used. It should be here for the protection of the nu system and not some exchange who implements poor security. If an exchange steals all the nubits from a custodian, then there should be some mechanism in place that would allow the NuNet to recover.
“Good” and “bad” can’t really apply here; their meanings are too subjective.
I’m with you, conceptually. Building systems that provide for improved security for users will be needed for any blockchain technology to make the leap from hobbyist to general use. While the current risks of exchange default are a major problem there’s also a silver lining: better technology, security practices, and business processes should naturally evolve in the space.
If shareholders are interested to learn where they could capitalize on the Nu network look no further than the “gaps” that exist in blockchain technologies and find ways to fill them. There are a massive number of opportunities for profit to be made discovering and developing solutions that meet real needs like the ones that @Hyena brings up.
In March 2013 there was an unplanned Bitcoin network fork that resulted in one double spend of around 10,000 NBT, as I recall. In that case, the transacting parties cooperated to remedy the situation. If something like that ever occurs on our network, where a network failure imposes a loss that isn’t remedied by the cooperation of transacting parties, shareholders will have the choice to make the party who suffered losses whole by passing an NSR or NBT grant. No other blockchain technology can do this. In this regard our protocol is unique in its ability to remedy user losses caused by network failures.
But is it wise to even consider doing something like this. Continuing this thought I wonder if a constitution of sorts should be drafted outlining do and do nots of what motions should and should not do. Making an exhaustive list would be impossible but I think having something is better then nothing.
curious in what context you are referring to @desrever ?
i have been kind of hanging back and observing… got a little freaked out. ironically, the exchange i originally wanted to support the most incurred losses as well! (bter) and i still have a significant stash of NSR tied up there…
They claim to have a more secure cold storage solution by allowing charge backs before a certain amount of blocks. So if coins are stolen you can just take them back within the window. I imagine this can be tweaked to serve other purposes as well.
