Ripple, the Charade – Trustnodes

Ripple, the Charade


It was meant to be one of the biggest speech in the crypto space. Former President of the United States, Bill Clinton, was to talk blockchain. What insight could he bring following his role at the top when the internet came of age? Was this to be a defining moment for the crypto space?

It most probably would have been had Bill Clinton chosen to attend out of his own free will, rather than probably paid for it. Yet within minutes it became apparent he was there more as a clown, as an entertainment figure, as, what we may even call, a deception.

How much he was paid for this appearance is unclear, yet just how unethical the act appears to be probably needs no elaboration.

Watched by many, the former President spent his time to talk about Brexit, the IRA, Bosnia, everything except blockchain tech.

Any insight that might have been expected turned into wondering how this might now reflect on the rest of the crypto space.

Yet Ripple has never been anything like the rest of the crypto space. They never cared about trustlessness or permissionless. They may give lip service to innovation, but not one line of code was presented at the “conference.” They may claim they have a blockchain, but it is at best a private and permissioned blockchain that requires a great deal of trust.

Hence one can say they’ve missed the point for while in their marketing they try and conflate, at the foundations of it Ripple is a database with 55 administrators selected by Ripple the company.

All banks run databases. The problem is they don’t trust each other’s databases because of course an employee might have changed something somewhere accidentally or intentionally.

In Ripple, it needs to be 20% of administrators that do so, making it slightly better than a simple database, but still something that requires a great deal of trust not least because half of the administrators are run by Ripple itself.

If therefore bank A wanted to send money to bank B, the problem there isn’t communication, but verification that the money has indeed moved and A doesn’t still have it.

That verification takes days and has to go through SWIFT, which is sort of a consortiums of banks, but at the end of it everyone can be fairly sure that the money has moved.

With Ripple, you can be sure with small amounts because no one would bother. But if we’re talking say a one billion dollars payment, you have to trust Ripple has actually moved it from A to B and hasn’t done so while still leaving A with that one billion.

If they do still leave A with a billion, who would know? Here, for example, is their “blockchain” as seen by the world:

Top crypto blockchains, October 2018.

We can see what’s going on in bitcoin, eth, and the rest. So we can be fairly sure there is no double spending, money isn’t being printed out of thin air, and so on.

We have no clue, however, about what’s going on in Ripple. All we know is what we are told by Ripple and the few other validators who seemingly only share transaction numbers. Ripple’s consensus whitepaper says these validators are trusted:

“Each server, s, maintains a unique node list, which is a set of other servers that s queries when determining consensus. Only the votes of the other members of the UNL of s are considered when determining consensus (as opposed to every node on the network).

Thus the UNL represents a subset of the network which when taken collectively, is “trusted” by s to not collude in an attempt to defraud the network. Note that this definition of “trust” does not require that each individual member of the UNL be trusted.”

In mudding the waters, Ripple has caused confusion by calling some servers as validators and showing explorers with many of them. The servers, however, do not sign the ledger and are at best a light wallet like the ones on your smartphone. This is clearly shown by Ripple stating:

“Once you are running a server, the additional cost and effort to operate a validator is essentially zero.”

The cost obviously can’t be zero if you’re maintaining accounts of hundreds of thousands of transactions running a day, unless you are doing what you’re told and not quite validating.

The actual database administrators are just 55, and that includes Microsoft, MIT, as well as others, with Ripple running more than half of them while approving who can run an administrator node.

The paper further says “values of pc greater than 20% will result in expected cartels of size greater than 20% of the network, thwarting the consensus process.”

While in bitcoin you need 50% to be honest, and in eth there may be fraud proof techniques where you need just 1% to be honest, thus 99% can be dishonest, in XRP you have an attack vector of just 20%. That’s 10 administrators or just one employee within Ripple.

There may thus be chatter about banks utilizing Ripple, but that’s not very true as far as XRP is concerned and it is probably not true because banks have no reason to trust Ripple’s database when they don’t trust other bank’s databases.

“Banks are unlikely to use distributed ledgers to process cross-border payments for now because of scalability and privacy issues, according to Ripple.” So said Reuters this summer.

Just last month, Sagar Sarbhai, Government & Regulatory relations at Ripple, made a fairly big revelation which went unnoticed. He said:

“In 2016, we gave 12 banks some XRP to test internally and they loved it!!! They wanted to use XRP but they couldn’t because a lack of regulatory certainty.”

Who doesn’t love free money. Following this Bill Clinton debacle, we now have to wonder whether Ripple is paying banks to “pilot” XRP. Santander, of course, has invested in Ripple. So how much is potentially just a marketing ploy is unclear.

Take xRapid, which is a nice and a very old idea. Instead of using SWIFT to reconcile databases, you use crypto to prove that the money has actually moved and has not been created out of thin air.

The way you do so in xRapid is fairly simple. You fund an account say on Coinbase, you buy whichever crypto you want, you send it to an exchange in say South Africa, sell it for South African money and then you withdraw it to your South African bank account.

The movement of crypto itself is quite easy, but we have to move fiat here from your bank account to Coinbase. That costs in fees, and can take five days. We then have to move fiat again from the South African exchange to your South African bank account. Again that can take time and can cost in fees.

With Ripple there would be additional problems as you have to trust Ripple to not double spend. Then there would be more general problems applicable for most cryptos.

The simplest one is the spread/difference between the buy price and the sell price. If you are moving much funds, that spread can be considerable and amounts to an additional fee on top.

A more complex one is volatility. Although the buying and selling is meant to happen with minutes, there have been times when the price has crashed or has significantly increased in minutes.

That can be a considerable cost and would raise the obvious question of why use XRP when you can use DAI or some other stable coin, needing not deal with spreads or with volatility.

If the latter is used, then there may be some end to end savings. In addition, the transfer through crypto can be within minutes end to end in Europe as there they have instant transfers between bank accounts within the same country.

XRP, however, might be an appealing option for banks or other payment companies as they might be able to buy them directly from Ripple at a steep discount, making the xRapid component irrelevant.

Ripple has sold some $70 million worth of XRP during this summer and has billions still. They are being added to the market constantly, with the price kept afloat, at least partially, by these marketing gimmicks like Bill Clinton.

Yet these discounted XRP would be bough off market, but then would be sold on the market to transfer it internationally. Something which may put some pressure on price rather than the opposite as many seem to think.

Once that discount factor is taken out, then price volatility may make xRapid not very competitive, certainly compared to stablecoins, but potentially also compared to current traditional methods.

So it may be the case some banks start to join and even utilize xRapid, but unless they buy XRP on the market, rather than at a steep discount off Ripple, then fundamentally it would be a gimmick and a charade of sorts because they wouldn’t really be utilizing xRapid, but instead would buy an asset for very cheap and would quickly sell it at market rates.

International remittance has of course been an identified use case in this space for long and plenty of start-ups rose to take the opportunity. The problem is that because of the fees between exchanges and bank accounts, and because of volatility, it can work fine in a bull market but may quickly lead to bankruptcy in a bear market.

Ripple hasn’t changed any of that aspect. To the contrary, they’ve replaced it with the most volatile coin out of the top ones, XRP, when they should be utilizing DAI or other stablecoins if they really wanted to solve a problem that has a sustainable profit-making use case.

Understandably, however, Ripple’s primary problem is off-loading their XRP to someone else, with xRapid a reasonable use case for that specific problem in the short term. Something which may benefit the banks if they get the ripple at a discount, but if it benefits other ripple holders, remains to be seen.



Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments