(Topic is originally from the Newbies forum.)
I'm sure that by now most of you know about Ripple, an idea created by Ryan Fugger for an open, decentralized payment network. It allows users to route payments through open, arbitrary trust networks just like the Internet allows for packets to be routed through the same. If you want to read more about it, you may do so at the following addresses:
http://archive.ripple-project.org/https://ripple.com/wiki/ (I do not agree with all aspects of this implementation, as I will explain.)
I'm sure that many of you also know that the currently dominant implementation of Ripple is one made by a company called OpenCoin Inc., and that using this implementation requires the use of a centrally governed currency known as XRP (or "Ripples"). Ripples were not created by mining or any other decentralized process, but rather by OpenCoin all at once in what could be conceived of as a 100% (
100 billion XRP) "pre-mine". Users of Ripples, which are required for account reserve requirements (300 XRP per account) and to pay transaction fees (decided by valdiators), only have OpenCoin's word that 50% (50 billion XRP) of the Ripples in existence will be distributed fairly and freely to users. The remaining 50% will be used by OpenCoin to fund their operations. 20% has already been given to developers.
If for whatever reason this doesn't seem unfair to you, replace "Ripple" with "SolidCoin" and you'll get the general idea. OpenCoin's Ripple is basically a payment network that requires a SolidCoin equivalent (though even SolidCoin had less pre-mined units) to use it. No alternative is allowed. There's nothing wrong with programmers making money from their craft, but when you consider how much evangelism, word-spreading, and programming of websites and services goes into making an alternative currency viable, 50% for the originators alone is too much, not to mention the security/morality issues involved in one entity having so much control over the network.
OpenCoin has tried many tactics to silence criticism of this aspect of Ripple, from claiming that XRP is not a currency (and eventually admitting that it is) to keeping the Ripple server code closed-source so that nobody can fork Ripple to a system that doesn't require XRP. Some have been swayed, but many have not. I've been debating the issue myself with JoelKatz on their forums, which you can read here:
https://ripple.com/forum/viewtopic.php?f=1&t=872 (I will admit that some of my earlier comments in this topic are unnecessarily confrontational and unproductive. I have been following the Ripple project for years, so I was angry at the idea that somebody had hijacked it. I hope that you will read the whole debate, particularly the latter parts, before judging the merit of our respective viewpoints.)
To me, the time for debate is over. That's why I've created this thread to discuss the creation of a fair alternative to Ripple, one that does not use XRP or any other centralized currency/component. The basic idea of Ripple as imagined by Ryan Fugger is good, but XRP is not. If you agree, then this thread is for you.
The basic protocol of OpenCoin's Ripple is sound and well-documented on their wiki (hopefully making it easy to replicate). The real issue is replacing XRP with a different mechanism for controlling spam. I have considered five different alternatives, listed below:
1. Instead of using/destroying XRP to pay transaction fees, users are allowed to pay transaction fees in the IOUs created by gateways, whether they be Bitcoin IOUs, dollar IOUs, or whatever by distributing them to the validators (that are on the user's unique node list), perhaps only a small, random subset of them, or even another random user in a "transaction fee lottery". It doesn't really matter where the fee goes, only that the user is required to pay it to prevent spam. This is somewhat more complicated than using only one currency, but has the advantage of allowing the user total freedom in choosing which currency they transact in. As a result, this is my current favorite solution. If these IOUs are supposed to have value, then not allowing transaction fees to be paid in them is pretty hypocritical.
2. Instead of only having one arbitrary currency (XRP), allow anybody to create their own XRP-equivalent. If a user finds the distribution of one currency inadequate, then they can create their own. This still restricts the user to using "pre-mined" currencies, but at least gives them a choice of which pre-miner they think is fairly distributing theirs. OpenCoin could even win in this scenario, if they allowed such competition.
3. Allow users to pay transaction fees directly in existing cryptocurrencies, such as Litecoin or Bitcoin, either by destroying them (sending them to an invalid address) or distributing them as detailed above. This would require that at least some of the validators that an existing cryptocurrency user plans on using be connected to that cryptocurrency's blockchain. Validators that aren't connected to a particular blockchain can still service users that don't plan on using its corresponding cryptocurrency. This has the advantage of attracting the existing cryptocurrencies' communities and promoting their value by making another service for them to be used on. It's win-win.
4. Replace XRP with a fairly mined "RippleCoin" or equivalent. This restricts the user to only being able to pay transaction fees in one currency, but at least makes that currency more fairly distributed in a decentralized way. This is my least favorite solution, since it introduces a computationally wasteful proof-of-work system into the scheme. It would also require debate about block intervals, block rewards, hashing algorithms, and all of the usual points of contention. Though this could be potentially combined with idea #2 to allow multiple "RippleCoin" variants.
5. Eliminate transaction fees and use some other method, such as HashCash, to combat spam. This would be the best solution, if it could be done. Ideas about this are welcome. Though our best chance of competing with Ripple is to compensate validators in some way, which Ripple does not do.
Combinations of these solutions or others could be used. Any other issues with OpenCoin's Ripple protocol that need to be addressed are also up for consideration.
Programmers, developers, website designers, and other talented people will be needed to make this system competitive with the existing Ripple implementation. I would like to start a Bitcoin/Litecoin/etc. donation fund for the purpose of developing this, but since I am a newer user I feel that it would be more appropriate for a well-known forum member to take charge of that aspect. I've seen some of them express similar sentiments, so I'm sure that one will step up. OpenCoin has continually delayed the release of Ripple's full source code, so waiting for them to do so before starting on a fork is a bad idea. They have made it their intention to prevent forks of Ripple from being viable by delaying the source code's release.
The last thing that must be addressed is a name for this new Ripple implementation. Here are some ideas that I've had: "Fripple" (short for "fair ripple"), "Nipple" (short for "new ripple"), "Wave", "Flutter", "Flow", and "Mesh". "Flutter" is my favorite, but ideas are welcome.
This thread is not for debating about OpenCoin's Ripple or XRP. If you see no problem with it, then you are free to use it. This thread is only for productive posts about the creation of a Ripple alternative. Thank you.
