Bitcoin Forum

What better way to complement your open p2p community money than with open p2p community credit? Ripple (http://ripple-project.org/) produces software that lets anyone establish credit relationships based on personal trust, and the emerging credit network allows payments to be routed through chains of interpersonal relationships.

You can use Ripple as a distributed credit network overlay, which seamlessly blends traditional and new economies. This makes financing easier, and also allows other tools such as automatic inter-currency exchange.

Right now there's an operational implementation available called Ripplepay (https://ripplepay.com/). You can sign up and use it freely. I think it would work great with bitcoin, as well as with other forms of credit such as national currencies, time, energy, etc. Right now there's discussion of adding a built-in bitcoin credit denomination, in the meantime it's possible to use an ad hoc substitute.

This should address some issues that have been raised in the community, including but not limited to the discussion in this thread (http://bitcointalk.org/index.php?topic=271.0). I encourage you to sign up at Ripplepay, set up your credit relationships with existing trust partners, and use it for payment and other financial options when suitable. I would be happy to extend some credit if you can show a good reason, and would accept credit at a decent rate.

This seems interesting, I did not know that. But hey! I'm missing "Bitcoin" as Default unit on ripplepay! :-) Are you somehow related to ripplepay developers? It would be nice to add Bitcoin as currency ;-).

I don't think that mutual credit systems stand a chance in the online arena.

there seems to be discussion on the ripple forums about adding it... :)

i'd be curious to know /why/ you think that. :)

I agree it would be nice to add Bitcoin as a Ripplepay currency, there's some discussion of that here (https://groups.google.com/group/rippleusers/browse_frm/thread/4939938b5bc33e61).

People can establish interpersonal trust through online or offline media.

Because mutual credit systems reintroduce the identity issues that credit cards have everywhere, and that PayPal was a response to online.  It requires a level of mutual trust between the trading parties, which usually involves a mutually trusted third party.  Ripple's trust system replaces the huge third party financial group that both (many) parties trust with a web of trust network, which is bound to have much variation in the level of that trust.  Variation alone is not what kills it, but the fact that the web of trust is a neccessary part of the system of mutual credit.  Whereas Bitcoin does not depend upon trust, even though it would enhance the Bitcoin economy.  If Bitcoin were to become a widely used currency upon Ripple, I can see how Ripple could function quite well as a web of trust mechanism for portions of the Bitcoin economy.  However, the trading of mutual credit (i.e. the IOU's) within the web of trust cannot scale beyond two people who already trust one another, because the IOU's are not all equal and none are equal to a widely trusted currency unit.

That sounds to me like a pretty good defense of Ripple as a decentralized credit network.

That was a defense of the web of trust idea with Bitcoin, but Bitcoin isn't credit.

in case of bitcoin, since transactions are 'free', you don't actually have to use the ripple credit lines as anything more than an indication of the level of trust between you and your counterparty. (though you could if you wanted to.)

as far as 'mutually trusted third party', as far as i read, your level of trust for personX is equal to the lowest trust link in the chain between you and personX, if you don't know personX directly. So say, the links are You - 50 -> personA - 60 -> personB - 30 -> personC - 50 -> personX. Then /your/ trust with personX will be 30 (the lowest valued link).

So for example, just because personC might trust personX for a million dollars, that's not going to be your level of trust for personX. So the fact that the web of trust is dispersed and variable, that's not really a problem, it seems.

At least, that's kind of what i'm thinking, according to my current understanding (I've only recently learned about ripple myself).

A decentralized credit network, or web of trust, can be denominated in bitcoin or other units of value, including a hybrid denomination suited to the respective parts of the network.

Adding a decentralized credit network to bitcoin would allow bitcoin partners to use existing trust along with their use of bitcoin's own expected value. It would also let bitcoin users interact in another (inexpensive) way with users of other currencies, and serve other financial roles.

In Ripple, your trust with an indirectly connected party combines different minimal routes, so if you have 30 btc minimum credit through one chain, and 50 btc minimum credit through an unrelated chain, then you would have access to 80 btc credit with the indirectly connected party.

That's the theory, but my problem with it is that trust isn't quantifiable.  A may trust B and B may trust C but that doesn't really equate to A trusting C.  A may trust B with regard to him honoring his own obligations, but still not trust B's judgment.  B may trust both, but actually be an unrealisticly trusting person.  C could be a grifter using B's trusting nature to get A as a mark.  Again, Bitcoin doesn't require trust to function, but nor does it defend A against confidence scams.  Mutual credit systems are a bit different, because they do depend on trust, and the lack of real trust within a theoretical web of trust acts as a limiting factor.

well, read more about the way ripple works. when there's a transaction between A and C via B, what happens is that A extends credit to B, and B extends that credit to C. so effectively, B is on the hook for extending the credit to C. So if you trust B to repay his credit, the problem of trusting C is on B's shoulders.

sure, B can default... but then you (a) shouldn't have trusted him in the first place, and (b) stop trusting him for the future (or at least reduce your credit line to him).

<disclaimer>just my understanding, i'm a noob at ripple</disclaimer>

Trust is quantifiable like other things. When A trusts B for a certain amount, that includes A's estimation of B's judgment. C can only use B to cheat A by the amount that A trusted B anyway. This isn't special to Ripple or webs of trust -- these networks just encode and enhance existing social trust. You could already let someone else you trust hold onto some value, whether bitcoins or dollars or otherwise -- for the short period of time during a purchase, or during a longer period of time with a verbal promise or a paper IOU or a Ripple obligation. In any of these cases, your partner may or may not reliably repay you. Ripple helps you track and otherwise make use of obligations.

I guess we are just going to have to differ.

Or we could change our minds. :)

That would have to be you.  I'm basing my conclusions on experience and human psychology, not mathmatics.

I would be happy to change my mind where there's reason or evidence that I'm wrong. That could include mistakes in math, psychology, history, or whatever.

Check out http://github.com/tjgillies/couchcash similar concept except it uses openid and webfinger for debt passing

in web of trust, you trust people not on how much you trust them as a person, but on their ability to judge trust.

if i trust you "30", that means im "30" something sure you're good at trusting people i would trust.

http://www.rubin.ch/pgp/weboftrust.en.html

This is a great idea because smart lending requires knowledge of the borrower.  This way credit is extended, but the originator of the loan doesn't need to take a risk on an unknown borrower. 

Ripple is wonderful! ;D