Then please explain the PeerFinder code within the rippled source tree: https://github.com/ripple/rippled/tree/9e519af887b0854ef167a45e52f4b50000cdf85c/src/ripple/peerfinder

There is a very nice Readme right at that link that explains the P2P design. Remember BearShare, one of the Gnutella clients? The person responsible for that is crafting the Ripple P2P design. He knows the topic area from hard won real world experience.

Gateways, on the other hand, are a convenience for moving the currency of your choice into and out of the Ripple system. Bitcoin relies on trusted exchanges for people to do the same, but they are more limited. "Oh," you say, "it is possible to use Bitcoin without ever using an exchange!" Well, it is similarly possible to use Ripple without using a gateway, but both systems are much more useful if you do. The choice in both is left to the user. There is no compulsion.

And what part, exactly, is not open source?

Just visit the ripple code repository on github: https://github.com/ripple/

They have open sourced:

• The full server: rippled
• The web client
• Java, JavaScript, and Ruby libraries for the API
• The ripplecharts.com code
• The blobvault for the wallet
• The network graph site code
• and a bunch of other minor stuff

Anything else, like an android wallet or web shopping cart, either hasn't been written yet or is not yet at a point where it is being used either as open or closed source.

Go look. It is all right there at github.

Being Open Source has nothing to do with mining, but if you want to use your computer to crunch numbers and get XRP, you can do that.

Go to www.ComputingForGood.org and get started. One big difference is that when you trade computation for XRP, you aren't just generating heat like with Bitcoin mining. Instead, you are crunching results for scientific research (mostly AIDS and research).

Any update?

I fully expect that the huge surge in signups and verification avalanche pushed the Ripple launch back, but I'm still curious.

You can currently mine both GRC and XRP at the same time. Ripple Labs is running an XRP giveaway based on contributions to the World Community Grid, a BOINC project.

Details at https://www.ripplelabs-wcgxrp.com

The giveaway exchange rate is 1000 XRP for every 5000 WGC points (714 BOINC points), which at current rates is roughly a $10 USD value each time you cash in your WCG points.

As long as you're already mining GRC and running BOINC, why not pick up some free XRP for no extra effort?

Incorrect. You should really look at the rippled PeerFinder code on github.

The comments in that file give a brief introduction to the design. You will also notice many commits from Vincent Falco, a veteran of p2p systems back to the gnutella days. He has a good handle on how such systems work in the wild.

This one of the problems that ripple is designed to solve. However, there are very few gateways issuing altcoins on ripple right now, and since liquidity is low, the exchange rates have wide spreads. As more gateways start issuing altcoins and more trade volume develops, you would have better success.

With liquidity, the way ripple would work is that you enter the merchant's bitcoin address and select the currency to send (You would be shown the amount of your selected currency required). Ripple would find a path to exchange your currency for BTC.bitstamp and then that would be sent to the bitcoin bridge hosted by Bitstamp. Bitstamp would then send actual bitcoin to the merchant.

Several gateways issue LTC, but Dividend Rippler currently supports the most altcoins. The current list is LTC, NMC and TRC.

The bitcoin bridge is pretty sweet. On the ripple client Send screen, enter a bitcoin address. Instead of sending within ripple, it will shoot that amount of bitcoins out of ripple and to the bitcoin address. The ripple side is nearly instantaneous while the bitcoin side has to wait on the transaction to be verified on the blockchain.

Note: The bitcoin bridge is currently only hosted by Bitstamp (there can be other bridges in the future), so the best conversion rate will come if you hold BTC.bitstamp in your ripple account. You can send other currencies and the system will automatically do the conversions, but your best exchange rate will come if you send BTC.bitstamp, since there will be no exchange, just a straight redemption.

Whether or not people speculate on XRP, like is done with pretty much every crypto-currency out there, does not negate the usefulness of the Ripple system itself. You can successfully use Ripple for many purposes and never hold more than the bare minimum of XRP needed for the account reserve and transaction fees.

Let's talk about the system, not some misguided users of the system.

This is actually one of my concerns, and something that might hamper adoption by players who could bring real liquidity to the system. 

Since the ledger is public, it seems like an opportunity for a company to provide a service where you give them a ripple address (or group of addresses), a jurisdiction and a time frame and then they produce a tax compliant report.

It would be nice to see this built in, but with all of the possible jurisdictions, I don't see that happening. Do you think there would be enough people willing to subscribe to such a service to support the programmers and various tax lawyers required to make it happen? What do you think serious traders would be willing to pay for this?

Note: I don't plan to do this myself. I just want to know if you think there is enough demand to support someone doing it the right way.

This entire wall of text is based on some fundamental misunderstandings that make the entire conclusion fall apart. I'll cut to the chase:


No, the person who directly trusted the defaulting party would be out the money.

You are never forced to accept IOUs from issuers who you do not trust. You misrepresent the system by claiming that one can be forced to hold IOUs from unknown parties, simply because they are on a chain of trust. This is false. In reality, the IOUs in the system will shift down the chain, so that each individual will only get an IOU from the next person down the chain, who they trust.

In your example above, you make it sound like Party Z ends up with Party A IOUs. Not so. Party Z ends up with Party Y IOUs, and Party B ends up with Party A IOUs. Thus, every recipient knows whose throat to choke.

Your argument also goes straight for the absurd by using $10,000 transactions. While some well acquainted hawaladars might adopt Ripple for such amounts within a tight circle of trust, most actual users will not. Even so, the system is not constrained by the smallest trust relationship in the path if there are multiple paths.

While playing with the system, I have trusted most gateways for .00001 of virtual currency and .01 of national currency while trusting Bitstamp for usable amounts. Guess what? Even with small transactions, the system has found multiple pathways to send that involved multiple parties.

This means that if you are using Ripple with your friends, trust them for small amounts each and you can still send bigger amounts through the pipe. The pathfinding will work it out, and you are not greatly exposed to any particular person.


Ripple has the concept of Trust Line Quality that accounts for the very phenomenon that you describe. It is a way to discount the IOUs from particular issuers that you value less than others.

It is also a way for individuals to make some money by acting as liquidity providers, since they can set their line quality to essential collect a transit fee. If a user's fees are too expensive, the pathfinding mechanism will route around them. However, if they are on the least cost path, they will be used and compensated accordingly.

Now, this mechanism has not been exposed in the client yet, but it does exist on the server. This is a part of the beta experience, and there are many additional features and polish coming that will make Ripple even better.

Once you consider how IOUs really flow within the system, the fact that you can't be forced to accept an IOU from someone you don't trust no matter what, and when you take into account that Trust Line Quality is built into the system and will be exposed in a future update, the basis of your argument falls apart.

Your conclusion is invalid. Ripple has not already failed. What happens in the future remains to be seen, but your particular concerns are not founded.

Once rippled is open sourced, it will only take minutes before someone announces "Tipple: so your friends can owe you more than *just* a beer."

Ripple is a different beast than bitcoin derived alt coins, however, and cloning it misses the point. The real power of Ripple comes from the gateways, which make it easy to get real value into the system and then get it out easily as whatever currency you need. OpenCoin is spending a lot of manpower and XRP signing up gateways who can move real money through Ripple, in many national currencies and crypto/virtual currencies. A simple clone will not have access to any of that. Network effects will be hard to overcome.   

---

For the life of me, I can't figure out why many supporters of small/new alt coins rail against Ripple on one hand while hoping that their coin gets listed on an exchange on the other. It seems like one of the steps of releasing a new coin should be for the creator to have a simple pass through Ripple gateway, see Dividend Rippler's method, so that even the first users have access to an exchange. The new coin announcement should include the Ripple issuing address and deposit/redemption instructions. 

BAM! If people are interested in the new coin, there is instant access to exchange markets or the ability to make purchases with the coin with anyone who accepts any currency Ripple payments or via the Bitcoin bridge. A new/small coin's community doesn't have to wait for someone else to decide that it is worth their while to add the coin to their exchange. The creator can control that from the start. For an added benefit, as the first gateway for the coin, the creator would get any transfer fees that he sets on the Ripple-side currency assets. If he runs the gateway well, his will probably end up staying the gateway of record.

---

Now, I can see a Ripple fork by those who don't agree with OpenCoin's XRP distribution gaining some measure of support in the crypto coin community. It might even get a few exchanges linked in. However, while that fork grows slowly among the crypto coin community, OpenCoin will be bringing more and more national currency gateways online. Since people will make good use of this to get money into and out of Bitcoin/alt coin exchanges that use the original Ripple, I don't see such a fork going very far.   


Around this part of the US, we call them "Bible Thumpers."

Is RedHat open source? Yes.

Can you copy the RedHat distribution and release a new version called "RedHat," "Fedora," "Red Hat Enterprise Linux," "RHEL," or use the Red Hat logos? NO!

Why not? Because even though Red Hat is an open source distribution, they have built a brand and trust around what they produce. People are free to modify or repackage the distribution as they see fit, but they need to build their own reputation instead of sliding by on RedHat's. CentOS, Oracle Linux and Yellow Dog Linux have all gotten the benefit of RedHat's work, but had to build their own reputations based on their own execution.

The same goes for OpenCoin's branch of Ripple. The value in the payment network isn't really being able to run a node. The value comes from the access to the gateways and community of users that create a viable distributed exchange and payment system. If OpenCoin builds out a good ecosystem, and they have hired a lot of people and are building many partnerships to make this happen, it isn't right for random forks of the code to trade off of Ripple's name to deceive people into using a backwater fork when they're actually looking for the real thing. 

New users' real money and real value will be at stake, and people should not be confused into using something other than what they're looking for if the Alt-Ripples are all called Ripple, too, and branded like the real thing.

So, yes, it is possible to be open source while trademarking the names and logos most commonly associated with the master branch of the code. I will address the current open source status below.


A large amount of the non-server code has already been open sourced and placed on github. This includes the ripple client, JS API library and other useful bits.

Want to contribute to an open source currency exchange that interfaces with Ripple? The Justcoin gateway server code is on github, too.


If you look at the Ripple wiki, it is very clear that extremely important features have not yet been completely implemented. Contracts are an especially powerful capability that is still missing.

It seems perfectly reasonable for the original designers of Ripple to realize their vision before open sourcing it. Ripple works now, but it certainly is not complete. There is likely very rapid change happening under the surface from both a protocol and capabilities standpoint. Being open source and having a ton of nodes on the network at this point would make it very hard or slow to reach the planned capabilities of the system.   

For most open source software, it does not slow development to open it early. When new capabilities are added, a new version is simply relased. However, activating new capabilities in Ripple requires participating nodes to buy into the changes. The more nodes in the system, the longer this takes, or the less likely it is to happen.

It makes sense to have slow and deliberate upgrades once the system has reached a certain level of feature completeness, but to do this while the planned capabilities are only partially implemented is a recipe for disaster.

I agree that if OpenCoin doesn't open source Ripple once it reaches feature completion, then that will be a very bad thing. Since OpenCoin wants Ripple to succeed, then they will honor their promises to go open source. For now, I'm giving them the benefit of the doubt.

Kraken.com said that they should be selling XRP by the end of the month. Kraken was trading XRP during their beta period. If you can wait, that is a low hassle way to get some.

Bitstamp currently sells XRP. However, the price available through their web interface is sometimes inflated. I moved only as much XRP into my ripple account as I needed to fund it and then withdrew Bitcoins into my ripple account. After that, I converted my BTC into XRP on the open market within ripple.

• Create a ripple wallet at https://ripple.com/client
  
• Deposit bitcoins into your Bitstamp account
  
• Trade bitcoins to get enough USD to buy 76-100 XRP (50 to fund the ripple account, 12.5 for line of trust to Bitstamp, 12.5 for a second line of trust if desired and some additional XRP for transactions)
  
• Buy 76-100 XRP and send it to your ripple account to fund it (on the Buy Ripples entry on the Buy/Sell tab)
  
• Extend trust to Bitstamp for BTC (there is a "trust this address" link on the Bitstamp Withdraw tab, Ripple entry)
  
• Withdraw your bitcoins to ripple (Withdraw tab, Ripple entry)
  
• In the ripple client, click on Send and send XRP to yourself. Currently, 100 XRP costs 0.00497863 BTC. If you are curious, you can see the order book at Advanced->Trade->XRP/BTC dropdown->change issuer for BTC to Bitstamp->Order Book. However, just buying XRP by clicking the Send button is quick and easy if you don't want to deal with the trading interface.
  

I hope that helps. Enjoy!

There was a good explanation for Bitcoiners posted back in May. Take a look at that.

It does involve trust, in the same way that trading on a Bitcoin exchange involves trusting the exchange.

Start by considering how Bitcoin exchanges involve trust, and then build from there. JoelKatz said it really well on the Ripple forum in December 2012 (why exchange rates in the example are dated):


The difference is that Ripple functions as a distributed exchange, so that the parties in the trade can trust different exchanges (gateways) while not needing to trust the exchange (gateway) that is trusted by their trading partner. If there is either a trust pathway between their respective gateways, or a conversion pathway through XRP, then the transaction can happen.

Not only is the exchange distributed, but you are not limited to the types of currencies supported by their gateway. With Bitcoin, you are limited by what your particular exchange supports. Just like you would trust a Bitcoin exchange to hold a certain amount of your fiat currency and cryptocoins, you would trust a Ripple gateway to do the same. You don't have to expose yourself by trusting unknown or untrusted gateways to use the system.

If you hold all of your Bitcoins yourself, and only pay directly without using an exchange, Ripple might not be for you. However, if you are comfortable trading on Bitcoin exchanges, you have already decided to trust and should be comfortable using Ripple. You just need to find a gateway that you trust. Bitstamp might be a good place to start, since you probably already have an opinion of it if you use Bitcoin.

If the system gains good traction in a certain geographical area that is the destination for substantial flows of remittances, then this will encourage those sending the remittances from outside the area to adopt the system as well.

Since China is the thread topic, consider it: China is #2 for remittances at $66 Billion USD per year in 2012 (http://english.cri.cn/6826/2013/08/07/2702s780437.htm discusses this). Ripple in China is also taking off.

Since Ripple has currency conversion built in, a Chinese worker in the US could be paid direct deposit at a US bank, have a effortless ACH transfer to their USD Ripple gateway (Say, snapswap.us or an upcoming one), then use Ripple to send CNY to their family in China, who use either rippleCN or Ripple China as CNY gateways.

Large Chinese uptake of Ripple makes it a no-brainer for ex-pat Chinese nationals to start using the system to send money home. At no point in this process do they need to care about the XRP's capability to "hold value" as a currency for longer than it takes to perform the transaction. If there is a good market for CNY-rippleCN/USD-SnapSwap or CNY-RippleChina/USD-SnapSwap, the conversion might not even pass through XRP as an intermediary.

China is only the #2 remittance market, but there are many others. All are fertile ground for adoption, and growth in one country will prompt growth in others as network effects take hold.    

Since eBay owns PayPal, it seems like a bit of a conflict of interest for them to accept Bitcoin. Some sort of PayPal / Bitcoin integration might be a precursor to eBay accepting Bitcoin.

On the other hand, since Bitcoin has irreversible transactions and PayPal is known for reversing transactions, this doesn't really seem to fit their model.