mmeijeri
|
|
May 15, 2013, 07:26:31 PM Last edit: May 15, 2013, 08:44:35 PM by mmeijeri |
|
Thanks for all the explanation, I'm getting a much clearer picture now. Is there a special reason the consensus process starts with a threshold value of 50%? Near 0% it would be like set union, while near 100% it would like set intersection. It seems logical to start at a lower value to mop up some transactions, and gradually ramp up the required level to come closer to a consensus. But why 50%?
|
ROI is not a verb, the term you're looking for is 'to break even'.
|
|
|
JoelKatz
Legendary
Offline
Activity: 1596
Merit: 1012
Democracy is vulnerable to a 51% attack.
|
|
May 15, 2013, 08:39:51 PM |
|
Thanks for all the explanation, I'm getting a much clearer picture now. Is there a special reason the consensus process starts with a threshold value of 50%? Near 0% it would be like set union, while near 100% it would like set intersection. It seems logical to start at a lower value top mop up some transactions, and gradually ramp up the required level to come closer to a consensus. But why 50%? The level is gradually increased to speed convergence. 50% for the initial operation was chosen at first arbitrarily -- the choice easiest to justify and the one that seemed the most "fair". But then simulation showed that it avalanched the fastest. If you pick a number much higher than 50%, you don't have as much room to ratchet up, and that can cause pathological behavior. If you pick a number much lower than 50%, you can find a transaction in the consensus set that can be hard to acquire because you aren't connected to nodes that have it.
|
I am an employee of Ripple. Follow me on Twitter @JoelKatz 1Joe1Katzci1rFcsr9HH7SLuHVnDy2aihZ BM-NBM3FRExVJSJJamV9ccgyWvQfratUHgN
|
|
|
cdog
|
|
May 15, 2013, 08:49:01 PM |
|
Ill give OpenCoin this. These guys are really, really smart. One of the best hustles Ive ever seen actually...
|
|
|
|
🏰 TradeFortress 🏰 (OP)
Bitcoin Veteran
VIP
Legendary
Offline
Activity: 1316
Merit: 1043
👻
|
|
May 15, 2013, 10:26:29 PM |
|
Can you say more about that avalanche algorithm? Or is it just the process of switching to the majority (or plurality) ledger if there is one?
Yes, you switch to the majority valid ledger. And what does a node do if for whatever reason all of its UNL nodes suddenly endorse a ledger that's on a different branch?
The node basically uses the following algorithm: 1) It switches to the majority of the ledgers it has not rejected. But it does not endorse that ledger. 2) The node tries to determine if that ledger can be endorsed -- determining whether it's a valid following ledger or not. 3) If the node determines the ledger is invalid, it rejects that ledger. 4) While this is going on, the node keeps tabs on any other nodes that have changed their position. Over time, more and more nodes will veto any invalid ledgers in the mix and soon all honest nodes will be endorsing valid ledgers. The valid ledger with the most endorsements will eventually win. This doesn't take nearly as long to actually happen as it might sound like. The ledgers will tend to be very, very similar and only differences are analyzed. Perhaps you'd show the source code, but that prevents you from stealth changing network core rules like you already did.
|
|
|
|
ihsotas
Newbie
Offline
Activity: 28
Merit: 0
|
|
May 15, 2013, 10:30:55 PM |
|
Listening to David Schwartz ("JoelKatz") describe Ripple -- really good overview around some of the use cases, better metaphors than you'll find on the Ripple site. Starts around 12:30, Episode #007 http://letstalkbitcoin.com
|
|
|
|
JoelKatz
Legendary
Offline
Activity: 1596
Merit: 1012
Democracy is vulnerable to a 51% attack.
|
|
May 15, 2013, 11:11:53 PM |
|
Perhaps you'd show the source code, but that prevents you from stealth changing network core rules like you already did.
What are you referring to? We do change network core rules all the time, primarily to add features, but what do you think was a "stealth" change?
|
I am an employee of Ripple. Follow me on Twitter @JoelKatz 1Joe1Katzci1rFcsr9HH7SLuHVnDy2aihZ BM-NBM3FRExVJSJJamV9ccgyWvQfratUHgN
|
|
|
anti-scam
Sr. Member
Offline
Activity: 476
Merit: 251
COINECT
|
|
May 15, 2013, 11:15:35 PM |
|
I can't believe Google Ventures and IDG Capital Partners (a $2.5 billion fund) just invested in OpenCoin. Man, these guys are pulling the wool over everyone's eyes! </sarcasm> Nothing makes me trust something more than when Google gets involved. </sarcasm>
|
|
|
|
Coinseeker
|
|
May 15, 2013, 11:23:22 PM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
|
If your ignore button isn't glowing, you're doing it wrong.
|
|
|
ihsotas
Newbie
Offline
Activity: 28
Merit: 0
|
|
May 16, 2013, 01:38:27 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
The more I learn about Ripple, the more I think that it's too important to leave in the hands of a private company. You almost want to fork the code the moment it becomes available, and replace the transaction-throttling XRPs with some tiny amount of BTC (like 100 satoshis per txn), so it's truly an open financial network. I can see large corporations signing onto 'Bipple' knowing that it's a consortium model outside of the whims of OpenCoin.
|
|
|
|
SGExodus
|
|
May 16, 2013, 01:41:20 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
How would Ripple help? Opencoin can ignore a Seize Warrant? Or the Exchanges that are federated with Ripple can ignore a Seize Warrant?
|
|
|
|
oakpacific
|
|
May 16, 2013, 01:43:29 AM Last edit: May 16, 2013, 02:25:57 AM by oakpacific |
|
LOL, I love this: Note: There is currently no way to tell if you are in the network minority. This feature is coming. It's actually trivial -- if there are, say, 1,000 generally trusted validators and you see validation from fewer than 700 of them, you might be in the minority. This has been implemented in the server about a month ago, the wiki is just out of date. The results of the Ripple consensus process are never sent to clients directly as verified. The ledger must pass a "validation gate" before it's considered confirmed and this validation gate will never be passed if you're in the minority because you won't get back sufficient validations. In contrast, if you're in the minority, you won't know until manual intervention with Bitcoin. If the world splits (as it did recently) people who are on the minority blockchain have no way to know that they are transacting on a block chain the rest of the world does not think is the longest chain. At best, you could statistically infer you were in the minority eventually based on blocks being found too slowly. Bitcoin could adopt the method Ripple uses to solve this problem. Major mining pools could periodically broadcast a signed object stating the hash of the tip of the blockchain they are currently building on. You could then know if you were in the minority. What consensus is it you have been talking about anyway? Global consensus? If my transaction with someone only gets validated by about 20 nodes, how do the decision, say, propagates to the 1000 other validators globally? Especially if I try to send two transactions simultaneously to two different non-overlapping groups of validators? And the way you talked about Bitcoin is, to say the best, dubious, it's up to the miner to work after the block it received from another, or persist on his own chain, even if you somehow choose the wrong chain due to propagation time, it's unlikely you will make the same mistake when you receive a second, or a third block. Moreover, Ripple never have miners, and Bitcoin common users never have to worry, their transactions will eventually be included.
|
|
|
|
ihsotas
Newbie
Offline
Activity: 28
Merit: 0
|
|
May 16, 2013, 01:44:19 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
How would Ripple help? Opencoin can ignore a Seize Warrant? Or the Exchanges that are federated with Ripple can ignore a Seize Warrant? Presumably, the Ripple network would 'mark down' the Dwolla-MtGox connection in the graph and find a new route for BTC/USD exchange instantly, a la IP routing. Coinlab and Coinbase are both MSB-compliant, for example.
|
|
|
|
solex
Legendary
Offline
Activity: 1078
Merit: 1006
100 satoshis -> ISO code
|
|
May 16, 2013, 01:45:39 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
How would Ripple help? Opencoin can ignore a Seize Warrant? Or the Exchanges that are federated with Ripple can ignore a Seize Warrant? As Ripple clearly facilitates fiat transfers we can be sure that Opencoin have complied with all the MSB regulations. Absolutely sure. https://bitcointalk.org/index.php?topic=200443.0
|
|
|
|
ihsotas
Newbie
Offline
Activity: 28
Merit: 0
|
|
May 16, 2013, 01:58:43 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
How would Ripple help? Opencoin can ignore a Seize Warrant? Or the Exchanges that are federated with Ripple can ignore a Seize Warrant? As Ripple clearly facilitates fiat transfers we can be sure that Opencoin have complied with all the MSB regulations. Absolutely sure. https://bitcointalk.org/index.php?topic=200443.0Yeah, that's a key question to ask them this weekend at BTC2013 -- what licenses do they have, and which ones do they intend to get? Most startups in the financial space tend to assume that the regulations are for the other guys. But Chris Larsen has been around finance for a while, so I'm hopeful.
|
|
|
|
smoothie
Legendary
Offline
Activity: 2492
Merit: 1473
LEALANA Bitcoin Grim Reaper
|
|
May 16, 2013, 02:52:27 AM |
|
Anything centralized can be broken. Ripple can be hacked seized and shut down in an instant if the powers that be don't want it around.
Since source is closed it isn't like someone can build on top of the ripple system....but why would you want to given things would likely still be centralized.
|
███████████████████████████████████████
,╓p@@███████@╗╖, ,p████████████████████N, d█████████████████████████b d██████████████████████████████æ ,████²█████████████████████████████, ,█████ ╙████████████████████╨ █████y ██████ `████████████████` ██████ ║██████ Ñ███████████` ███████ ███████ ╩██████Ñ ███████ ███████ ▐▄ ²██╩ a▌ ███████ ╢██████ ▐▓█▄ ▄█▓▌ ███████ ██████ ▐▓▓▓▓▌, ▄█▓▓▓▌ ██████─ ▐▓▓▓▓▓▓█,,▄▓▓▓▓▓▓▌ ▐▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▌ ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓─ ²▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╩ ▀▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▀ ²▀▀▓▓▓▓▓▓▓▓▓▓▓▓▀▀` ²²² ███████████████████████████████████████
| . ★☆ WWW.LEALANA.COM My PGP fingerprint is A764D833. History of Monero development Visualization ★☆ . LEALANA BITCOIN GRIM REAPER SILVER COINS. |
|
|
|
JoelKatz
Legendary
Offline
Activity: 1596
Merit: 1012
Democracy is vulnerable to a 51% attack.
|
|
May 16, 2013, 03:16:17 AM |
|
Anything centralized can be broken. Ripple can be hacked seized and shut down in an instant if the powers that be don't want it around.
Since source is closed it isn't like someone can build on top of the ripple system....but why would you want to given things would likely still be centralized.
We 100% agree. That's why we designed Ripple not to require any central authorities and why we're working to open source the server code.
|
I am an employee of Ripple. Follow me on Twitter @JoelKatz 1Joe1Katzci1rFcsr9HH7SLuHVnDy2aihZ BM-NBM3FRExVJSJJamV9ccgyWvQfratUHgN
|
|
|
cypherdoc
Legendary
Offline
Activity: 1764
Merit: 1002
|
|
May 16, 2013, 03:43:21 AM |
|
Recent events just highlight exactly why Ripple is necessary. Bitcoin needs Ripple in order to have a chance at legitimacy and mass adoption. It may already be too late but for now, I stay hopeful.
The more I learn about Ripple, the more I think that it's too important to leave in the hands of a private company. You almost want to fork the code the moment it becomes available, and replace the transaction-throttling XRPs with some tiny amount of BTC (like 100 satoshis per txn), so it's truly an open financial network. I can see large corporations signing onto 'Bipple' knowing that it's a consortium model outside of the whims of OpenCoin. wow. now we know why they're hesitating to release.
|
|
|
|
jbreher
Legendary
Offline
Activity: 3038
Merit: 1660
lose: unfind ... loose: untight
|
|
May 16, 2013, 06:22:27 AM |
|
Some people say 'the third time is the charm". Is there any chance my questions might be addressed upon third airing? Are my questions too basic? Too obscure? Too repetitive? I really have been looking, but I've not found what seem to me to be answers to the questions below:Neither shill nor premine screamer, I am one of the hopefully still rational trying to understand the implications behind Ripple. Allow me some continued discussion...Source code is not "open", you can do what you want in your closed code and nobody can really control you (except governments of course).
Not true. All ledgers are public, and signed by us. All differences between ledgers must be justified by a transaction signed by the account that issued it. All transactions are public and each transaction includes the precise "delta" is applied to the ledger. We can't just do whatever we want.
This is a point I have missed until now - "All ledgers are public and signed by us." While public ledgers seem on the surface a good thing [assuming that fine-grained pseudonymity is still available], I would like to focus upon the "signed by us" portion. Shall I assume that "us" is OpenCoin? Now and forever? Now until some future unspecified eventuality occurs? Is this not then a centralized point of attack?Also, because you are not Open Source, governments can shut you down at any given time they want.
The code would automatically become open source if that happens. Developers have that in their contract and any number of people have the source code and understand these terms. I've discussed the contingency plans for that case several times, I think at least once in this very thread.
Similar to misterbigg, this struck me as new information. What plans are in place for this to happen? While I do not doubt that an intent has been discussed and generally assented to, there are a litany of chances for failure from that position to actual concrete plans, with M of N safeguards, clearly defined trigger points, and failsafe and backup mechanisms. I'm sure we'll all recall that Pirate and GLBSE claimed to have such plans in place. A public disclosure of the specific plans would likely go a long way of easing at least this particular doubt in the minds of many. If you truly discussed this in this very thread, it must have been disclosed only very obliquely. Or perhaps I (and demonstrably at least one other) were sleeping during that portion of the thread.
...As I catch up with the thread's head, I ran across this... why you think there is a consensus mechanism that actually carries weight in a decision making process moreso than just being able to view balances and transactions on a gateways ledger (which i think is a long shot feature especially if we're talking about banks).
Each validator signs the consensus ledger each time a new one is created. So you have a set of cryptographic signatures for each ledger produced by a large number of independently-operated validators, none of which gets to choose the rules by which new ledgers are created from prior ledgers. Further, the ledgers contain hash chains which lead to prior ledgers and signed transactions that justify the changes between them. Also important -- the gateways don't get to choose the rules by which transactions are executed, nor can they make exceptions to them. Are the "us" in "All ledgers are public and signed by us" above actually the validators? Who are the validators? And what makes them able to judge what is and what is not a valid transaction? Can anyone be a validator? What is the criteria?
As far as the "rules by which new ledgers are created from prior ledgers", are these rules encoded in immutable protocol? Or are the subject to change over time? And if the latter, who can participate in these rule changes?
Last in the above, what happens if a gateway refuses to honor their previous commitment to carry out all transactions on a non-discretionary basis? Would this be synonymous with refusal to honor their IOUs? What if a gateway becomes insolvent?
|
Anyone with a campaign ad in their signature -- for an organization with which they are not otherwise affiliated -- is automatically deducted credibility points.
I've been convicted of heresy. Convicted by a mere known extortionist. Read my Trust for details.
|
|
|
🏰 TradeFortress 🏰 (OP)
Bitcoin Veteran
VIP
Legendary
Offline
Activity: 1316
Merit: 1043
👻
|
|
May 16, 2013, 06:24:35 AM |
|
FYI, the core network rules have already changed. Transactions from addresses with less than 300 XRP are now accepted as this limit is reduced.
This is a "hard fork" in Bitcoin terms.
|
|
|
|
JoelKatz
Legendary
Offline
Activity: 1596
Merit: 1012
Democracy is vulnerable to a 51% attack.
|
|
May 16, 2013, 06:27:40 AM |
|
Some people say 'the third time is the charm". Is there any chance my questions might be addressed upon third airing? Sorry, didn't see these. This is a point I have missed until now - "All ledgers are public and signed by us." While public ledgers seem on the surface a good thing [assuming that fine-grained pseudonymity is still available], I would like to focus upon the "signed by us" portion. Shall I assume that "us" is OpenCoin? Now and forever? Now until some future unspecified eventuality occurs? Is this not then a centralized point of attack? Every validator signs every ledger it wishes to validate, that's what validators do. Right now, all validators are run by either OpenCoin or those working closely with us. But that will change. Similar to misterbigg, this struck me as new information. What plans are in place for this to happen? While I do not doubt that an intent has been discussed and generally assented to, there are a litany of chances for failure from that position to actual concrete plans, with M of N safeguards, clearly defined trigger points, and failsafe and backup mechanisms. I'm sure we'll all recall that Pirate and GLBSE claimed to have such plans in place. A public disclosure of the specific plans would likely go a long way of easing at least this particular doubt in the minds of many. If you truly discussed this in this very thread, it must have been disclosed only very obliquely. Or perhaps I (and demonstrably at least one other) were sleeping during that portion of the thread.
We'll probably have to wait for the source code to open to make this argument academic. Are the "us" in "All ledgers are public and signed by us" above actually the validators? Who are the validators? And what makes them able to judge what is and what is not a valid transaction? Can anyone be a validator? What is the criteria? Anyone who wants to can be a validator once the source code is open. Validators participate in the consensus process (assuming people listen to them). They judge what is a valid transaction by applying the network rules, checking signatures, and so on. As far as the "rules by which new ledgers are created from prior ledgers", are these rules encoded in immutable protocol? Or are the subject to change over time? And if the latter, who can participate in these rule changes? They are subject to change by supermajority. Validators participate in the rule change process -- again, to the extent that others are willing to listen to them. Last in the above, what happens if a gateway refuses to honor their previous commitment to carry out all transactions on a non-discretionary basis? Would this be synonymous with refusal to honor their IOUs? What if a gateway becomes insolvent?[/color]
Gateways have no control over how Ripple transactions take place. They can, of course, refuse to honor their commitments outside the Ripple network. This will hurt those, and only those, who chose to trust that gateway. Gateways and validators have nothing to do with each other. As far as the Ripple network is concerned, gateways are just accounts that issue transactions that follow network rules just like everyone else.
|
I am an employee of Ripple. Follow me on Twitter @JoelKatz 1Joe1Katzci1rFcsr9HH7SLuHVnDy2aihZ BM-NBM3FRExVJSJJamV9ccgyWvQfratUHgN
|
|
|
|