On Nov 29, 2012 12:38 PM, "Apostolis Xekoukoulotakis" <xeko...@gmail.com> wrote:
>
> People could verify the validity of a transaction with zero knowledge.
> http://en.wikipedia.org/wiki/Homomorphic_encryption
>
That would be nice, but these systems only support addition and multiplication of encrypted numbers. Verifying Ripple transactions requires many more operations than that. But if you can come up with a clever scheme, please let us know.
Ryan
>
> --
>
>
On Thu, Nov 29, 2012 at 11:14 AM, dwilliams <dudewi...@gmail.com> wrote:
> A few questions:
> * Is it a decentralized solution?
Yes.
> * Does the user have control over who is able to see their transaction data
> (I'm guessing no if it's a blockchain approach)?
Good question. No, all transaction data is public, which is probably
the main drawback to this approach. It is the tradeoff required to
have a system in which is extremely resilient to denial of service
attacks (since all machines process all transactions, attacking a few
of them accomplishes nothing). In contrast, my two-phase commit
design offers more privacy (although some credit information needs to
be exposed to achieve good routing performance), at the cost of being
more vulnerable to DoS (since only my server processes my
transactions).
The way to achieve privacy in blockchain Ripple is to act through a
proxy, which is called a nexus, and direct the nexus to make and
receive payments on your behalf. Unfortunately, while your funds are
at the nexus, they are not available for you to use to enable through
transactions at your personal node, so it's a bit of a tradeoff. But
in the general case, where you have some excess funds not required for
through transactions, it should be fairly straightforward to keep
funds at one or more nexuses for long enough to avoid having your
private transactions linked to you through the timing of transfers to
and from the nexus, especially if the nexus is busy.
It's not a perfect solution, and I still would like to develop my
two-phase commit design for applications requiring greater privacy. I
can even envision the two systems being linked up in various ways.
But in either system, there will be a need for continued development
of privacy schemes as more sophisticated attacks emerge, much as
occurs in Tor. And neither system will offer complete privacy, except
to those who forego their ability to act as an intermediary, and
choose to operate completely through proxies/nexuses.
There are also potential social benefits to openness, like the ability
to build a strong public reputation by one's behaviour in the network.
I'm quite interested to see how these develop in the new system.
> * Does the potential user have any inkling of how their transaction data,
> metadata, and out-of-band data can be used to make inferences about them
> that they'd rather not be public information? There has been at least one
> academic study of these dangers wrt Bitcoin, so if this Ripple/BTC hybrid
> doesn't take this into account it's irresponsible to say the least.
>
I don't think the new system will be claiming to offer anonymity with
respect to Ripple transactions. I can't speak for how well-informed
users will be, but it would certainly be risky to let users assume
their transactions will be anonymous when they are not.
> Also, consider the following:
> 1 public physical mailing address gets x unsolicited messages where x is
> bounded by nonzero cost of sending 1 message
> 2 email lowers cost of sending messages to nearly zero
> 3 x becomes practically unbounded
>
> Now, consider:
> 1 commercial banks extend credit where y is bounded by the nonzero cost of
> getting a loan
> 2 Ripple lowers the cost of extending credit to practically zero
> 3 y becomes practically unbounded
>
> x is obviously "spam". My point is that while you've defined the benefits
> that could stand in for "y" on the Ripple website, you haven't outlined the
> costs. What are they in your view? (Keep in mind they aren't the same
> costs as those of BTC which does not need to leverage human trust in order
> to facilitate transactions.)
>
Don't worry, the cost of offering credit to someone who is not worthy
of it is never zero. Actually, what I hope Ripple accomplishes is to
create a non-hierarchical financial network that does not require a
central authority who must bail out those in privileged positions, so
participants must bear responsibility for their own decisions. In
that respect, Ripple could in fact lead to a decrease in lending, were
it not for the fact that it will enable so many more relationships to
be mapped into the financial trust network than ever before.
Ryan
> --
>
>
>
--
> --
>
>
Sincerely yours,
Apostolis Xekoukoulotakis
Ryan
--
Will this allow:
* Chaumian cash transactions
* that thing that probably has a name, which is whatever
FellowTraveler was talking about when he described how to audit that
the books and verify all accounts have been settled without requiring
a transaction history in OT-- if you have no idea what I'm talking
about I'll go digging to find it
* microtransactions
If this can facilitate practical microtransactions and blind
signatures then you've found the holy grail of digital currency.
This is not really necessary: blind signatures can be substituted withOn 11/29/12, Ryan Fugger <rfu...@gmail.com> wrote:
> On Nov 29, 2012 12:38 PM, "Apostolis Xekoukoulotakis" <xeko...@gmail.com>
> wrote:
>>
>> People could verify the validity of a transaction with zero knowledge.
>> http://en.wikipedia.org/wiki/Homomorphic_encryption
>>
>
> That would be nice, but these systems only support addition and
> multiplication of encrypted numbers. Verifying Ripple transactions requires
> many more operations than that. But if you can come up with a clever
> scheme, please let us know.
"registries" for Ripple. Two phase Ripple can have a commit method
similar to the proof of work chain one [1].
The main advantage is the
faster latency of this new chain, but also including less data in the
new chain. There's still use cases where it's necessary for ownership
to be public, smart locks [2], for example. The other advantage is if
IOUs are public, you don't need intermediaries to be online when
creating a new Ripple transaction. In this sense, public Ripple is
more p2p than private Ripple.
Since the old proof of work chain 2PR commit was slow, you still
needed registries for POS payments, for example. Now maybe we can base
it all in this new faster chain.
I have an idea to make both public and private Ripple assets
compatible within the same atomic transaction. Of course, some review
is very welcomed. I want to turn Ripple transactions into merkle trees
where each actor can be required to sign different branches of the
trees and may not have access to all the leafs.
I'm not sure my explanation is clear. I'll start another thread.
Ryan
--
Ryan
--
- Walter
-
--
Ryan,
I have an idea that adds a decentralized option for this: http://en.wikipedia.org/wiki/Redistribution_of_income_and_wealth
consider it in the context of this: http://www.youtube.com/watch?v=QPKKQnijnsM *Wealth Inequality in America*
If you see that it would add value,
connect with me
//Johan
--
You received this message because you are subscribed to the Google Groups "Ripple Project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rippleusers...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
Moreover isn't this a wealth transfer from people with strong identities to Anonymous?
Oh. Damn that is going to get expensive. And it is going to make identity and definition of recipient important as you wouldn't necessarily want every *node* to be redistributed to but every human being no?
I'm happy to announce that there is finally a team seriously building
a distributed Ripple network at Ripple.com. The team is led by
founder Jed McCaleb, who also founded the MtGox Bitcoin exchange and
created eDonkey2000, and CEO Chris Larsen, founder of Prosper.com.
I've been talking to Jed, Chris and other members of their team over
the past few months, and while their plan is very ambitious, I believe
if anyone can develop the Ripple concept on a global scale, they can.
Their system is based on a Bitcoin-style blockchain, much as we have
discussed here over the last few years as an interesting possibility,
but with a novel miner-less consensus mechanism that allows
transactions to be confirmed nearly instantaneously.
After discussions with Jed's team, and some long-standing members of
the Ripple community, I've agreed that Jed's project should use the
name Ripple and be considered our primary implementation. It was hard
for me to let others step in to this role, but from the beginning I
always intended for someone else to implement the concept, and I'm
lucky to have finally found a group that is more than worthy of taking
the project to the next level.
Please check out http://ripple.com, sign up for the beta if you're
interested, and watch for the launch coming soon...
Ryan
the simplest solution I´ve come up with is that people receive relative how much they invest (sum of their transactions),
one person can receive from multiple different networks, but only relative his transaction activity with those networks,
--
You received this message because you are subscribed to a topic in the Google Groups "Ripple Project" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/rippleusers/IVin3Qwrp7k/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to rippleusers...@googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to rippleusers+unsubscribe@googlegroups.com.
--
You received this message because you are subscribed to a topic in the Google Groups "Ripple Project" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/rippleusers/IVin3Qwrp7k/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to rippleusers+unsubscribe@googlegroups.com.