But that doesn't really answer the question being asked. If the cryptocurrency is completely controlled by a centralized authority who has absolute rights over the system, why bother with the technical and cognitive hurdles of using a blockchain? A standard ledger in a centralized database is more than adequate for that task, and much easier to maintain.
I responded to a few comments in kind, and I’d say it’s impossible to answer in a vacuum without knowing the actual token use case/function.
Generically, we have seen instances where centralized ledgers/databases are manipulated and permit double spend. Even in the case of multi billion dollar publicly traded companies going Private, during due diligence we have seen upward of double the issued shares of stock than actually exist. In such a case we have the corporation with a centralized stock ledger, stock trusts with their own centralized ledgers, etc... but the mistake of double spend still happened. In those cases Blockchain would have never allowed those errors, one such case cost the buyer $150M personally post acquisition. Not to say that is in anyway applicable here, but without even knowing the token function we can’t say, but it just goes to show Blockchain does have some use cases and benefits where centralized ledgers have failed costing hundreds of millions.
Plenty of blockchains have allowed double spending as well due to errors in coding of either the clients, the blockchain implementation itself, or due to faulty contracts.
Well sure there has been shitty code that has allowed doublespend or unintended transfers...but the satoshi nakamoto Blockchain did solve the problem of double spend on a decentralized peer to peer network, do you disagree?
Take the current status of publicly traded companies, they require a lot of middlemen to manage these centralized stock ledgers, including, the corporate general counsel, underwriters/investment banks, stock trust, stock exchanges, and stock brokers/trading apps.
Each one of those middlemen takes a significant slice of the pie, whereas Blockchain would allow corporatations to bypass all these middlemen and allow stock to be issued and traded P2P.
Is there a reason you believe the existing centralized stock ledgers of public companies and system of middlemen is more advantageous than having P2P stock on a distributed ledger/network?
Absolutely nothing: the problem cryptocurrencies solve is allowing mutually distrusting parties to engage in transactions.
The only potential advantage of such a system over an RDBMS is the inability to reverse transactions without people taking serious notice, simple git-style hash chains generally solve this too, often with reduced overhead.
In this case it's simply being used as a system of control while distancing FB from the responsibility for issuing currency. Facebook wants to own the money you spend and collect a vig on each transaction. The parent was just pointing out that's the real motivation here. If you go looking for a technical reason you won't find one.
It puts their currency in the same semantic basket as Bitcoin and Ethereum. The effort to give these other currencies legitimacy also gives Facebook’s currency legitimacy; many people who support these other currencies will also support Facebook’s. If Facebook’s currency is attacked, it becomes (in some sense) an attack on all cryptocurrencies.
Technically I can't think of one. I think the last line of the article might be the key: "By introducing a level of decentralization to the governance of the project, Facebook may be able to avoid regulation related to it holding too much power over a global currency."
That plus cryptocurrencies is still a buzzword that generates interest and speculation (especially since they've rebounded a bit these past few months).
I think there are two aspects. One is that it probably actually is a more robust technical approach, and they don't need to do a lot of research, because there are open source projects they can easily adapt.
Number two, they want to get in ahead of other cryptocurrency. They don't want to wait around for Bitcoin or Ethereum to become an expected form of payment, because by that time they won't be able to cash in. If they have their own cryptocurrency, they can take advantage of the hype for bitcoin etc., but still control and profit from it directly.
> why bother with the technical and cognitive hurdles of using a blockchain
A blockchain can be implemented in a very simple way when operated by a centralized authority. What adds complexity to it is the consensus protocol required for decentralization when you cannot trust all parties in the network.
But using a database as a ledger would clear faster than even the fastest blockchain, wouldn’t it? Or am I misunderstanding what you mean by “clearing”. I’m assuming you mean how long it takes for person A to get money to person B.
What takes time in clearing is not information transmission.
Most of the time is spent managing risk. Sure, that includes persistence, but that is also marginal. You also have to validate integrity, authenticity, cross-reference against legal requirements (eg. enforce rates, detect fraud — including internal fraud, apply sanctions lists, estimate AML-CFT risk, …). You have to make sure all accounting rules are followed. You have to abide by the procedures that you and the regulator determined. All that with no downtime.
All of this takes time. Historically, a large part was manual. While automation has helped, there are quite a few clearing houses where an army of programmers is spent battling the chaos.
A system least tolerant to inconsistencies and most tolerant to failure, increases reliability while reducing risk.
The cryptographic properties of a distributed blockchain help there, regardless of whether the system has open membership (which is not Facebook’s case, unlike most cryptocurrencies).
A different ledger design can be equally powerful or more. In fact, I expect that a lot of the compliance and accounting logic can be expressed elegantly in a custom design.
All banks are effectively running a huge database at the core, yet that's not a common problem. They have procedures, they have backups and they have human oversight. The last element is key, if somebody manages to hack into the system and send all the money into their own account the bank's administrator can detect the problem and say "well, that's clearly not supposed to happen, let's roll that back".
On a decentralized, trustless blockchain you have no such thing. So you either end up hard-forking like Ethereum did after their DAO debacle (but then are you a decentralized trustless system anymore?) or you continue normally and basically the people who lost their money are screwed.
