Topic: Ripple: A Distributed Exchange for Bitcoin - page 17. (Read 66657 times)

In a manner of speaking yes. This was my progression of emotion towards Ripple:

1. Denial: "This can't possibly work. Who needs it when we have Bitcoin?"

2. Anger: "These guys gave themselves all the coins! This is another pre-mine / ponzi scam."

3. Acceptance: "Ripple solves important problems, and is well designed."

I studied Ripple for a while and asked many questions. It is much more complex than Bitcoin, that is certain. But also holds the promise of doing much more. Ripple doesn't replace Bitcoin but rather, complements it. A lot of people making objections in this thread simply don't understand Ripple and haven't thought long enough about it to see the implications.

One of Ripple's prime advantages over Bitcoin is that the development of the software and promotion of the related gateway services is run by a for-profit company instead of a group of neckbeards. You might think this is a bad thing but actually, it aligns the financial interests with the interests of the users. There's a financial incentive to write clean software, document it extensively, and work on the features that users actually want. Compare this with Bitcoin: if a user wants a feature that the developers don't find exciting to work on, the usual answer is "you don't pay our salary, Bitcoin is Open Source, if you want that feature then write it yourself."

The OpenCoin team is already up to 20 people I think. They are all getting a salary, and there is a level of management which guides the development. They have actual deadlines, and investors to be held accountable to. Compare this with Bitcoin, which still hasn't had a 1.0 release in over four years. Unlike Bitcoin, the Ripple protocol specifications are being written at the same time that the code is being written. Ripple should not experience the problem that Bitcoin has, where the protocol specification is changed on a whim at the pleasure of a few developers who have commit access to the Github repository.

OpenCoin's primary stated means of generating revenue is to "hold XRPs and hope they go up in value." That means they have a financial interest in making XRPs useful, and keeping their value stable (to the extent that they are able to do so, which will be very limited once they are all given away and sold). People need not worry that the rippled software is not open source yet, it will definitely be open sourced some time this year. Because otherwise, XRPs will be essentially worthless since no one would trust a closed system.

Because OpenCoin needs Ripple to be successful in order for them to make money, they can do things that Bitcoin can't realistically do. Like hire a PR firm to brand the Ripple product and promote it to the masses in a commercialized way. Compare this with Bitcoin, which only recently underwent a revision to the main website. The attitude of the Bitcoin developers and principals is that they don't want Bitcoin to grow too fast, and that it is still experimental beta software. OpenCoin, on the other hand, is aggressively courting partners to make Ripple provide as much utility as possible in a short time.

I had the pleasure of going to the Ripple meetup in San Francisco in March. I met with David Schwartz (JoelKatz), Stefan, Jed, Arthur, and a few others. After talking to these guys, the impression is that they are smart. I found a group of friendly, motivated workers in their Mission District office with a bunch of computers at desks. They come in to work, there's whiteboards, and a lot of activity. It looks just like any other promising startup. There's a CEO, and they have an impressive panel of investors and backers. People who dismiss Ripple, do so at their own peril. OpenCoin has a lot of momentum, plenty of smart people, and a bulging war-chest of investment capital.

Ripple is already an impressive pieces of working software. You can go into your Ripple wallet right now and trade Bitcoins for US dollars, Euros, and XRPs, in the distributed order book. You can deposit and withdraw money at both Bitstamp and WeExchange. The Ripple beta solves the problem we recently experienced with the denial of service and MtGox lag. When Ripple opens its doors to the public, the first thing it will do is provide profoundly deep liquidity for Bitcoin. It will make the process of depositing money at an exchange a distributed process, since deposited funds can be used to purchase bitcoins in ANY order book not just one. Withdrawal of funds will be similarly distributed. If an exchange has a $10,000 monthly withdrawal limit you could just open an account at another exchange and withdraw from there as well. If MtGox can't verify users quickly enough, anyone could deposit money at a different Ripple-enabled exchange that has a smaller queue. In fairly short order, the fees exchanges charge for trades will be forced downwards due to the perfect competition that Ripple enables. Going beyond Bitcoin, Ripple will easily integrate into traditional payment networks like debit cards or checking accounts, allowing vendors to unknowingly accept Ripple payments through their already existing systems. Since Ripple IOUs correspond naturally to fiat currencies, these integrations will be far more convenient and functional than their Bitcoin counterparts.

Ripple is poised to be the next big thing during the crypto-currency revolution.

Thanks for mentioning Liberty Reserve, I hadn't heard of that before. I'm all for a Bitcoin-based alternative, all I'm trying to say is that we shouldn't shun Ripple. It would make a great alternative to Mt Gox.

A virtual currency based on USD or EUR already exists. Liberty Reserve (USD and EUR) is easier than Ripple and it is also centralized but at least you don't need to install a closed source software where could be anything.
Ripple is an overkill for this purpose.
A total decentralized virtual USD could be constructed with the bitcoin technology with colored coins on the top of bitcoin or namecoin.

Thrust network:
If you thrust in somebody then why do you need Ripple to intermediate ? If you friend or your neighbor want to borrow 1.000 $ from you why would you need Ripple for it ?
A more complex thrust network for an organization or a religious group  can be built more reliable with colored coins on the bitcoin or namecoin network as stated above.


I am not sure if it would be more decentralized than a Liberty Reserve - Bitcoin exchange. As long as Ripple is controlled by somebody he can be forced to cooperate with the authorities.

Sane merchants could easily accept Bitstamp-issued BTC IOUs. And private individuals could accept payments through their trust network.


Who knows, but in the meantime we'll still have a much better exchange for BTC than Mt Gox.

That's one of my main issues with ripple still:
There are no "BTC-IOUs"!
I can issue Sukrim-BTC on ripple right now without any issues. Even more than currently exist! In that regard ripple is far more decentralized than btc since there is not one central currency.
No sane merchant would accept my btc however, or if they did, they would immediately ask for settlement here and now.

The nice thing however is if I find somebody who for example exchanges up to 5 btc from anyone to 99 bit cents each in well trusted MtGox BTC, this person might make a profit, will still settle with me immediately and another link in the chain might exchange MtGox BTC to AmazonUSD which the merchant then gets in the end. I see problems with exchanging within the same currency based on trust (10 USD from my bank are worth more than 10 USD that you promise to send via mail) but if there are enough players, it should even out.

Still I fear that a central bank scenario with one or few highly trusted gateways for major currencies is one of the possible and maybe even likely endgame scenarios.

If people then don't settle immediately (e.g. payout fees and the gateway is highly trusted anyways) there's a higher and higher risk of catastrophes. Just imagine MtGox having a cypriotic instead of polish bank account right now. Also once you are highly trusted you can start to gradually discourage settlement by introducing fees, KYC policies etc.

All on all: the higher the trust, the lower the need for settlement or debts and the higher the risk. The lower the trust,the higher the need for settlements and still the higher the risk of getting money that can't be settled.

Right now, there are six servers run by Opencoin handling client requests, pathfinding, validating, and so on. Two gateways, I think, are also running their own servers to track the network. There are a collection of other servers run by various people for testing purposes. 12 servers is probably pretty close to the right figure. (Not counting test or isolated networks.)

Aren't there 12 servers right now? But all owned by you and your gateway partners?

The transaction fee can be adjusted by consensus. We expect the fee to remain very low, but we can't tell you for sure any more than the Bitcoin developers can tell you what Bitcoin transaction fees will be in a decade.

Like Bitcoin, the design requires no central authorities other than possible coordinated failure scenarios or design improvements. The operating Ripple payment network is not distributed now, and we agree that it needs to be.

mr bigg I find that a pretty keen insightt. Weren't you originally a ripple hater?

Seems to me they are riding one of the carts on the "decentralized" train uninvited.

What exactly makes them distributed besides them claims so a hundred times through the website?

i've read through the website and just have a couple questions... it says the fee will be a fraction of a ripple. It also says the fee will cost a fraction of a cent. Does that mean ripples will be worth about a cent each? Is that going to be deflationary since they are slowly destroyed? Will the fee to send in ripples ever adjust if the ripples themselves change in value?

Unfortunately, that much is true, for Ripple as well as for Bitcoin.


The architecture is decentralised, but there are only a few servers right now if I understand it correctly. There are instructions for the Ripple server (rippled) on the Ripple wiki, but I haven't been able to download any binaries. I'm not sure if they have been released yet, and I know the source code hasn't been. I imagine the binaries haven't been released yet because the system is still in flux.


You can also "ripple" IOUs in any currency through your private trust network. Eventually gateways will become hubs that will connect everybody easily, but you can get started without them. It's a pity the existing gateways only accept IOUs for the USD and not for other real world currencies.


As far as the mechanics of the payment system go it is better than Bitcoin, but it suffers from the same lack of adoption by real world merchant. However, I expect it to grow much more quickly since it can provide cheaper and more convenient payment services than traditional systems in any currency. The extreme volatility of BTC is a major impediment to its adoption as a payment system. Ripple allows you to avoid that risk by using fiat currencies. Adoption of XRP might also be easier because I've heard they are trying to stabilise the exchange rate, which they can do to a degree because they hold the bulk of all XRP that exist and because XRP reserves are needed to activate accounts and use their functionality.

But the killer app for Bitcoin users is as a decentralised exchange for BTC.

This is pretty much all true. But none of these things are fundamental about Ripple. It's just the difference between where we are and where we are going. You basically just summarized our priority list right now.

Ripple is very difficult to understand and therefore is difficult to argument against it.
But what I understood it is centralized and not open source.
Another problem is that it needs gateways like exchanges by bitcoin and actually there only one gateway.
Without gateways to fiat you cannot transfer fiat money just Ripple.
So the problem which exists by bitcoin with exchanges by Ripple is even worse at the moment it is just hidden by the complexity of the system which is difficult to understand.

Ripple currently is not open source, and it's not truly decentralized yet.

One of the great advantages of Bitcoin is it's ability to send money to anyone anywhere with little to no cost.


Would ripple give this same benefit to fiat, thus making that aspect of Bitcoin no longer an advantage?

The same is true of BTC. If you want to pay through Ripple with BTC IOUs, you can do so, provided the merchant accepts them. But if the merchant accepts BTC directly (and why wouldn't he if he does accept BTC IOUs?), you can simply transfer the BTCs from Ripple to your favourite wallet and bypass Ripple completely. To you Ripple would then simply be a distributed and more resilient Mt Gox, only used for exchanging between BTC and fiat currencies.

Right, similar to coupons from Mt. Gox, BTC-E, Bitstamp, Vouch-X, or even Dwolla. Once you convert your bitcoins or dollars to one of those IOUs and get a code, you now have to trust that central party to stay in business.  This isn't any different from banks, except funds in some are trusted more than others by law at the moment. The more ways we can get around a central party as the intermediary, the better.

They have a lot of people interested from here: https://bitcointalksearch.org/topic/ripple-giveaway-145506

Is sending an international bank wire such a hassle in the US? I can send them from my internet banking in Australia. The main hassle is the cost ($22).



Weexchange website is kind of nightmarish I agree. It wouldn't let me register a first, but the second attempt succeeded. Needn't have bothered. The spread was like 20/180.