Ripplesite's architecture
Daniel Lidstrom
rg
At the moment they can't communicate.
There is a project called "ripple protocol" to allow server to comunicate.
http://ripple.sourceforge.net/
The same protocol can be the base for a future peer to peer implementation.
Ryan Fugger said he considers ripple protocol a low priority today. I
agree with him.
Daniel Lidstrom
--
You received this message because you are subscribed to the Google Groups "Ripple users" group.
To post to this group, send email to rippl...@googlegroups.com.
To unsubscribe from this group, send email to rippleusers...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/rippleusers?hl=en.
Daniel
Daniel
Daniel Lidstrom
I have some other questions, though. Does Ripple make payments with only your own IOUs, or can it also exchange the IOUs that you hold from others, and that others are willing to accept, in order to find more payment routes? Also, does it make any attempt to find a route which minimizes transaction fees and interest rates?
Has anybody thought of any local trust metrics for the network which could be used to suggest new direct credit lines a user might want to extend? This could help to increase the connectedness of the network, thereby reducing third party liability, and hence transaction fees and interest rates.
Has anybody here looked at Bitcoin yet? I think an anonymous digital currency with zero third party liability and (practically) no transaction fees, like Bitcoin, would be very useful for settling debts within Ripple. We should also keep in mind that Ripple will always need a complementary homogeneous currency to price goods in, since its assets, IOUs, are heterogeneous. So for those of us who would like to move away from fiat currency, or who just want privacy and to save on transaction fees, we should also think about supporting and using something like Bitcoin concurrently.
For anybody interested, there is no third party liability in Bitcoin because there is no central issuer; the rate of money creation is fixed, and the supply asymptotically approaches a cap. Also, transactions are verified by a p2p network.
Disclosure: I own 505 bitcoins, which is currently worth about $2.50. ;)
Thanks for any responses,
Dan
Ryan Fugger
> That's my intention. I'm currently studying Bitcoin and Ripple for some
> inspiration on this.
>
> I have some other questions, though. Does Ripple make payments with only
> your own IOUs, or can it also exchange the IOUs that you hold from others,
> and that others are willing to accept, in order to find more payment
> routes? Also, does it make any attempt to find a route which minimizes
> transaction fees and interest rates?
>
If others are willing to accept IOUs you hold, then there is always a
path through the issuer of those IOUs, so yes, it does find those
routes, and many more.
Right now it doesn't attempt to optimize routes to minimize interest,
and charging transaction fees isn't implemented yet, but such an
optimization is always possible in the future.
> Has anybody thought of any local trust metrics for the network which could
> be used to suggest new direct credit lines a user might want to extend?
> This could help to increase the connectedness of the network, thereby
> reducing third party liability, and hence transaction fees and interest
> rates.
>
You mean something like, "If you trust X, then maybe you also trust
[or want to try trusting] Y," because lots of people who trust X also
trust Y? That might be a good idea.
> Has anybody here looked at Bitcoin yet? I think an anonymous digital
> currency with zero third party liability and (practically) no transaction
> fees, like Bitcoin, would be very useful for settling debts within Ripple.
> We should also keep in mind that Ripple will always need a complementary
> homogeneous currency to price goods in, since its assets, IOUs, are
> heterogeneous. So for those of us who would like to move away from fiat
> currency, or who just want privacy and to save on transaction fees, we
> should also think about supporting and using something like Bitcoin
> concurrently.
>
Sure, you can use anything to settle debts within Ripple if you need
to. What's more promising for me is using Ripple to connect
alternative currencies like Bitcoin into a single global payment
network where holders of Bitcoin don't need to find someone who
accepts Bitcoin in order to spend it. Instead, they can pay anyone
who uses Ripple through someone else who uses Bitcoin.
In other words, Ripple is a clearing house for alternative currencies.
In Ripplepay, I've emphasized personal currencies, but really they
could be any kind of currencies.
> For anybody interested, there is no third party liability in Bitcoin because
> there is no central issuer; the rate of money creation is fixed, and the
> supply asymptotically approaches a cap. Also, transactions are verified by
> a p2p network.
>
Hmm... Bitcoin looks interesting, but I can't figure out what gives
the currency value. Simply making a coin hard to create doesn't imply
value to me.
Ryan
Daniel Lidstrom
Can Ripple aggregate and broadcast market data somehow?
The initial value thing (the regression theorem) with Bitcoin is definitely a problem. I've been thinking about possible ways to organize a decentralized reserve network to overcome this. It would need a suitable local trust metric in order to recommend a safe bitcoin price floor for everyone to agree upon, given their individual reserves and the properties of the network, as well as optimally safe individual reserve targets, given this price floor, so their risk is appropriately balanced. An agreeable spread would also have to be decided upon somehow. Indeed, I've also been thinking about whether Ripple could be used among reserve holders in order to balance this risk by moving reserves around between them. Since Bitcoin is basically an inflation-free currency, the reserves would have to keep up with demand in order to maintain a true peg, though. The longer and more transparently and consistently the peg can be maintained, the more confidence is created, and the more liquid the currency will become. But there's no loss to the reserve holders when demand eventually outstrips reserves. This would be a risk for reserve holders to take on, but with the potential for reward; although I don't really know what would happen if all the reserve holders started competing with each other... Based on your suggestion, I think this would all work very well (if it would work at all :P) within the Ripple network, provided we can get Ripple to take off.
Would be nice if Ripple became a common financial network to tie them all together.
I have to finish writing up my thesis first, but afterward I'm happy to help if the project is indeed starting back up. My background is mathematical physics, though, so I don't know how useful this will be. But I'm more than willing to learn any new stuff. Any thoughts on top priorities?
Dan
Ryan Fugger
> I have to finish writing up my thesis first, but afterward I'm happy to help
> if the project is indeed starting back up. My background is mathematical
> physics, though, so I don't know how useful this will be. But I'm more than
> willing to learn any new stuff. Any thoughts on top priorities?
>
Top priority is always to come up with a feasible plan to grow Ripple
to the point where it's useful as a payment system in its own right,
and not just an IOU tracker between friends.
rg
> Can Ripple aggregate and broadcast market data somehow?
>
> The initial value thing (the regression theorem) with Bitcoin is
> definitely a problem. I've been thinking about possible ways to
> organize a decentralized reserve network to overcome this. It would
> need a suitable local trust metric in order to recommend a safe
> bitcoin price floor for everyone to agree upon, given their individual
> reserves and the properties of the network, as well as optimally safe
> individual reserve targets, given this price floor, so their risk is
> appropriately balanced. An agreeable spread would also have to be
> decided upon somehow. Indeed, I've also been thinking about whether
> Ripple could be used among reserve holders in order to balance this
> risk by moving reserves around between them. Since Bitcoin is
> basically an inflation-free currency, the reserves would have to keep
> up with demand in order to maintain a true peg, though. The longer
> and more transparently and consistently the peg can be maintained, the
> more confidence is created, and the more liquid the currency will
> become. But there's no loss to the reserve holders when demand
> eventually outstrips reserves. This would be a risk for reserve
> holders to take on, but with the potential for reward; although I
> don't really know what would happen if all the reserve holders started
> competing with each other... Based on your suggestion, I think this
> would all work very well (if it would work at all :P) within the
> Ripple network, provided we can get Ripple to take off.
>
"credit lines".
A credit line can be seen as the capacity to move value between two
nodes, without any commission or with very low cost or effort.
Ripple is agnostic about the origin of credit lines. It is not important
what kind of warranty, deposit, trust mechanism is behind this credit lines.
If Bitcoin is a way to move value between two nodes, it can be used on
ripple.
In my vision a Reserve is something that stays in a safe place and
usually do not moves. A reserve can improve accountability and help
create a Credit Line. That credit line can be used in Ripple net.
Maybe you can describe it the other way around and consider Ripple a way
to move "virtual reserves". But I do not think this makes things more clear.
I hope we can collaborate now or after you finish your studies.
Romualdo.
Daniel Lidstrom
I completely agree with you everything you've said here. I do, however, think that Ripple has some limitations that require something like Bitcoin to complement it (and vice-versa). This was just a proposal for a way to make Bitcoin more liquid using Ripple, and vice-versa if it would work.
It shouldn't be much longer before I've finished my studies, and I look forward to collaborating with you then.
Dan
If something like Bitcoin could become very liquid, then it would greatly complement Ripple by
--
Daniel Lidstrom
Mats
could you elaborate a bit on this? Would be most delightful to hear
your thoughts about it.
Mats
On Jun 21, 5:03 am, Daniel Lidstrom <lidstro...@gmail.com> wrote:
>I do, however,
> think that Ripple has some limitations that require something like Bitcoin
> to complement it (and vice-versa).
>
Daniel Lidstrom
It seems to me that nodes will have to be able to scale beyond the friend-to-friend model and make businesses out of extending credit in order to accommodate the less socially connected users, or those whose friends simply aren't able to extend to them the credit they require. And don't forget Shakespeare's, 'neither a borrower nor a lender be'; lending money to friends is not for everybody. These, I think, will be reflected by a high degree of 'preferential attachment' in the Ripple network, if it's successful. But nodes in Ripple are going to have a very hard time scaling up if they can't become independent of the traditional financial system. The Ripple network will always require external debt clearing mechanisms to function, and while Paypal and the traditional banking system will work fine for small and infrequent payments among friends, if any node tries to make a successful business out of extending credit, they will face high fees from them, invasions of privacy (KYC), frozen accounts (see Paypal's 'Acceptable Use Policy'), and legal bullying (see Liberty Dollar, and e-gold). A viable alternative financial system is a competitor to the traditional one, and will be treated as such. I see Bitcoin is a potential solution to these problems. It would also be nice to be able to eventually stop pricing things in inflationary fiat currencies, and to reduce the creepy amounts of power these give governments.
Conversely, Bitcoin needs a financial network to complement it; people need to be able to lend what they don't need, and borrow what they don't have. This is particularly important for Bitcoin market makers, since they need to be able to take on short positions in anticipation of times of falling demand, in order to smooth out the exchange rate and add liquidity to Bitcoin.
I guess, ideally, I'd like to see something like Ripple as being sort of a standardized banking protocol and a financial network that anybody is free to be a part of, and something like Bitcoin would be the cash that this financial network is built upon.
What are your thoughts on this?
Dan
Thomas Hartman
your customer) privacy invasion, or other onerous aspects of
civilization.
This dooms it to be a victim of its own success (and predictable
crackdown). After all, people that really want to get off the grid can
use well established hawala networks, or trade with trusted people in
untraceable commodities (gold, drugs, whatever). And people that care
remember what happened to egold.
The ideal thing would be a business model that is appealing to a
certain population, *including* the tax and KYC stuff. I admit I am
not sure what that would be.
But I think something that would increase transparancy and
accountability, rather than some kind of underground thing. These
things are just imho impossible to get bootstrapped given the current
situation, where after all you can do drug deals and paper cash is
"anonymous enough". Personally I like todd boyle's ideas about shared
transaction repositories. (Bitrotted but still available via wayback
if I recall correctly.)
--
Need somewhere to put your code? http://patch-tag.com
Want to build a webapp? http://happstack.com
Daniel Lidstrom
Oh, and the lesson I learned from e-gold is that digital cash has to be totally decentralized, like Bitcoin is ;) But yes, the bootstrapping problem is a doozy.
Thomas Hartman
somewhat , especially intractable in the current environment when you
are trying to bootstrap a community oriented open source deal that
doesn't tether transactions to anything of value.
Sure if you had a few million dollars you could do like egold, or set
up your own hawala network. But it doesn't look like that would happen
with ripple.
In terms of "striving to becoming a target" :) I'm actually more
interested in what happened with prosper.com than egold. prosper also
got shut down, and seemingly for reasons that I think could be an
issue with ripple if it ever got going. namely, lenders abusing
creditors crust, and creditors taking on more risk than they expected
given prosper's pitch and generally getting screwed.
I think this is a case where most people were idealistic but it still
didn't end well.
Thomas Hartman
I think even getting to the point where you can *be* a target is
somewhat , especially intractabl
should be
I think even getting to the point where you can *be* a target is
somewhat intractable, especially in the current environment...
On Mon, Jun 21, 2010 at 10:51 AM, Thomas Hartman
Daniel Lidstrom
Lending club seems to be working much better than Prosper. I wonder what they've done differently to avoid similar problems?
Also garbled: "lenders abusing creditors crust" :D
Thomas Hartman
> tether to something of value.
I pretty much agree
Chris Wagner
> My worry is that Ripple won't be able to scale much beyond occasional
> small purchases and micropayments for too many users unless nodes can
> safely tether to something of value. Haven't done any sort of
> detailed analysis, though - this is just my intuition. What are your
> thoughts on this?
Depends on what exactly you mean, but I think you can look to the
current credit-card regime for some inspiration in what is possible in
terms of credit networks. My intuition leads me to believe that what is
essentially a more distributed version of the current, credit-card-based
network (i.e., Ripple) would be more robust and more reliable. Seems
more a matter of gaining critical mass; small purchases and
micropayments could be the way to get there.
But if you're talking more in terms of the units of measure (gold versus
USD versus "bitcoins"), I'd probably be more in agreement. I think that
the popular fiat monies of the world will continue to deteriorate and,
eventually, each will die... but that doesn't necessarily mean you
couldn't have a stable fiat-style currency, if its "creation" were
somehow severed from the control of any one person or group of people.
(Is that how Bitcoin works? I haven't looked into it too much.)
Daniel Lidstrom
That's sort of how Bitcoin works. The total production rate is fixed and distributed according to CPU cycle contributions, but it halves every four years, so the supply asymptotically approaches a fixed quantity (twice the supply created in the first four years). So there is nobody setting 'monetary policy'. I agree that fiat currencies can be stable long-term, but they are irresistible to abuse, so they never are. Bitcoin also doesn't inherently come with any redemption promises, though, so in that sense it's similar to fiat currencies - except there isn't any fiat either. So bootstrapping it is a really difficult problem. But governments have managed to solve this with coercion, so I'm interested in whether or not it's possible through voluntarily means.
--
Ryan Fugger
relationships in a Ripple network -- it doesn't have to be just
friendly trust. Bank accounts would work on Ripple just as easily.
So there's no theoretical worry about being limited by one's friends'
credit any more than there is today. Just take your credit rating
somewhere and get a line of credit based on a full legal agreement,
just like today.
Ryan
Daniel Lidstrom
I think this could be avoided in Ripple, but at the cost of debt issuers having to set up KYC compliant accounts with everyone who would at any time hold their debt, and the trade of such debt instead being done indirectly, through the issuer. Again, please correct me if I'm wrong, cause I have no idea of what I'm talking about here.
Miles Thompson
Daniel Lidstrom
--