They are tainting the stolen coins to avoid them to be sold. What’s the point of having a decentralized currency if a centralized entity make the decisions anyway at the end of day?
Perhaps I am misunderstanding how tracking coins typically happens, but if "tainting" the coins effectively prohibits the actors from withdrawing at their perceived value, how would this work if the actor decided to send the vast majority coins to many random, arbitrary accounts? (e.g. arbitrarily sending coins to everyone who has committed a trade on some exchange in the last X days.) Would enough information be known to allow an actor to enumerate arbitrary public addresses thus use this to attempt to hide in the crowd?
> how would this work if the actor decided to send the vast majority coins to many random, arbitrary accounts? (e.g. arbitrarily sending coins to everyone who has
Then they'd have to take those coins off the blacklist, or else prepare to deal with an increasingly intractable tracking problem. Ethically, that would be the less bad option, since the thief would be deprived off the income, even if it's not returned to the victim.
If it's a non-negligible amount compared to other wallets, that would be pretty easy to track down. I mean, you'd look at new wallets compared to existing wallets.
The attackers can still poison new wallets that appear on the chain - either you effectively shut down new wallet creation, or let the attackers possibly spend their coins.
Let's say I'm an innocent holder of the currency, and I receive x coins from a tainted account. I can post a transaction that sends x coins (or x - transaction_fee) to an unreachable address (say 0x0).
Lots of companies do exactly this. For example, in Japan, where I live, the train and bus companies will give you a card. You can "put money" on the card and then spend it on travel. You can also use the cards to buy things in the convenience stores, some restaurants, vending machines, etc, etc.
You can think of this as a "coin" that's pegged against prevailing currency (in my case, the yen). It's not a "cryptocoin" since there is no "crypto" involved, but as there is barely any "crypto" involved in cryptocoins anyway, I think that's a bit of a moot point.
There are problems with this model. First you have to trust the central authority with your money. They could do a runner and there would be nothing you could do about it. In fact, when I lived in the UK, I discovered that London Transport often mischarges - it forgets where you "tapped in" or "tapped out" and charges you the maximum possible charge for your journey. In order to get your money back, you have to register your card and apply within a specific time period to get refunded. If you don't know the procedure, don't want to give out your person information, don't realise you were mischarged or wait too long -- sorry, your money is gone (Really big piece of advice for anyone using an Oyster card: Register it and check the charges every day. Unless things have gotten better in the last few years, I'm sure you will be extremely surprised at the charges).
The problem here is that the payment processor has all the power in the transaction. They can just take your money if they feel like it. Also, they can refuse to pay for things if they feel like it. Finally, they are a single point of failure. If they have technical (or financial!) problems, then you may not be able to spend your money.
Whether or not you assume initiatives like Bitcoin were started as a scam, the implementation potentially solves a lot of the problems of these payment processors. Importantly, if you put your coins in a central exchange you are right back to square one! No amount of crypto goodness will save you from the exchange shenanigans because they control your wallet.
This is why things like the Lightning Network are interesting. It allows for centralised payment processors, but with distributed guarantees about who owns the coins and where and when they can spend them.
I find it incredibly unfortunate that "cryptocoins" endure such incredible hype and involve so much real (and imagined) money. So many power plays, so many scams, so much FUD. But at the centre of it, these are interesting real problems. Once the dust settles it will be quite nice to see what useful results actually emerge.
The blockchain is an interesting Computer Science toy. But in the 10+ years since it was invented, the only real application that anyone has found for it is Massively Distributed Ponzi Schemes.
The problems that you point out with London Transport aren't solvable by a distributed blockchain, because London Transport is a single central authority. It's completely pointless creating a distributed ticketing mechanism for them where they don't have all the power, because the ticketing mechanism is for the trains that London Transport run and control completely. There's no point having a ticket that London Transport don't recognise as valid. They're always going to be a single point of failure because they run the whole system. Sorry to be so blunt about it.
The solution to this problem is not to add in MORE complexity, networking and distribution. It'll be in removing complexity and improving connectivity.
But, of course, your stock price doesn't increase by removing complexity. You need the blockchains to impress the idiots who buy shares.
I think you're blinding yourself with the hype. The "blockchain" is nothing more than a merkle tree. We've known about these for a long time. The insight is that by publishing a merkle tree we can show that various actors are following a protocol. This is very valuable. For an example of another application take a look at certificate transparency (not actually a fan of CAs, but it was the most readily available example).
It's interesting that you say that creating a payment processor for transportation fares that is not controlled by a monopoly is pointless. I currently live in a country that has at least 5 payment processors for train fares and it's taken them decades to figure out how to interoperate Even now I still can't use my rail card in Tokyo.
These are real problems. "Computer Science toys" seems like another word for "research" to me. That the research was done by some random guy on the internet with unknown motivations is completely beside the point. Yes, like most people, I could definitely do without the scams. I could definitely do without the "OMG! Blockchain" knee jerk reaction. But would it be too impolite to also complain about the "Blockchain? Pfft!" knee jerk reaction? Lightning is a great example of a useful protocol -- full stop. I don't care if it's implemented on top of Bitcoin or on top of some byzantine 18th century payment exchange based on carrier pigeons. We need more of this stuff, not less.
My point about the transport fares was really badly put. Sorry. I meant that the whole system has an inherent single point of failure and single central authority: the train system. A train ticket for a train that doesn't run is useless ;) Using a distributed, authority-less system to provide the tickets doesn't change that.
I totally agree about pure research giving us CS "toys" that eventually turn out to be useful. But we've had ten years to find some use for blockchain that isn't cryptocurrency, and as far as I'm aware, no-one's found one. My suspicion is that this is because the problems that the blockchain solve only rarely need solving. It's unlucky that one of those rare occurrences is currency, hence all the bullshit.
insinuating that the only reason people say machine learning and not statistics is just false. the term machine learning refers to a specific part of statistics that is very different from statistics as a whole.
A lot of it is actually called Statistical Learning. Google that term. Some techniques are not classified as such, at least not yet, e.g. deep convolutional neural networks.
why are apple not just called fruits? because there is utility in calling apples apples and not just fruits, because we want to know specifically which subpart of the group fruits you are eating or there is in the pie, not because apples are so damn popular
Because this way you have a public ledger that can't be mutated without notifying the public. Not everything that blockchain gives is for the trust distributed use case. For that fact, proof of work is the only thing that is purely there because bitcoin's blockchain needs a distributed trust system.
Which is basically what a blockchain is...
Just that instead of a cryptographically append-only acyclic directed graph, its a cryptographically append-only linked list.
Because then it would be evident to anyone that you were running a pyramid scheme and law enforcement could shut you down by just closing down a single server?
Saying something isn’t decentralized in the crypto world can mean any of a number of things. The problem is that decentralization isn’t one specific thing that can be measured. There’s absolute centralization like a mainframe controlled by a single individual not connected to the outside world. Beyond that nothing else is absolute, there’s just varying degrees of decentralization.
You could have a slightly decentralized governance like a consortium that votes on how to manage the mainframe. You could have a distributed network of computers all managed by a single entity. Each of those have elements of centralization and decentralization. Within those categories there is varying levels of decentralization. For instance there could be a completely democratic organization that votes on how to manage the mainframe or more of a republic style where members vote on a person to manage the mainframe for them.
So, when someone says something isn’t decentralized in the crypto world they typically mean some element is centralized or is slightly less decentralized. For bitcoin you will see people say this in reference to large mining pools that have the power to control consensus to a certain extent. For NEM, they’re probably referring to the fact that the nodes that control the network are closed source. Since they are closed source that means one entity can control the direction of the network, and one can’t simply fork what they’re building.
I’m of the opinion that there will never be complete decentralization, and every project will have some degree is centralization. We’ll get better with time, but I think decentralization is an inherently intractable problem.
I wonder what would happen if they did an airdrop of part of those stolen coins in some random wallets. Like $1M USD each for random 100 lucky winners. Would they be able to redeem them?
The hacker can still move coins around on chain, they just can't turn it into fiat (since they broke a real law by a real government, they can't be given the fiat.)
